字幕列表 影片播放
The Internet of Things is the giant network
of interconnected devices
from these devices are capable of making decisions
without any human intervention.
Edureka welcome you
to the full course session on the internet of things
and what follows is a series
of fundamental concepts on the internet of things
which will help you get started with IoT.
But before we begin let's look
at our agenda for today. so we can start out
with introducing you to the concept of iot.
We are going to discuss.
Why do you need I will take
what is iot of you benefits of iot for biofuel features
and finally the very
important five layer architecture of authority.
Then you're going to discuss Raspberry Pi in this section.
We are going to talk about why you need Raspberry Pi
what does Raspberry Pi of you Hardware specifications
and the installation of Rosman fire also,
This module.
We have a few demos including the Raspberry Pi
a few cents hats and Raspberry Pi camera module.
Next.
We are going to discuss the top seven projects
in iot in this section.
We're going to talk about the best projects
that we found in Rog next.
We will talk about iot devices
are going to talk about various devices
which came into Inception
after the concept of iot was introduced
then we want to talk about iot applications
in this Section we are going to discuss
practical applications of iot
in various domains
such as Healthcare security traffic governance.
So on and so forth and finally we are going to discuss
all things career in Ironton.
They're going to talk about the job market
the opportunity the salary in different geographies
and experience so on and so forth,
but that we come to the end
of our agenda also going to take up this time to subscribe to us
and don't forget to hit that Bell icon to double missile.
Update from The Edge Arabia YouTube channel.
So without Much Ado,
let's get started what is internet of things
now to help you understand what is internet of things?
Let's look at an example of a mobile phones Forest.
Okay.
Our mobile phones has GPS tracking.
It also has mobile gyroscope you have adaptive brightness,
which gets adjusted
based on the light falling on it you have voice recognition.
And you also have face detection
which identifies who is the user now again iPhone x
is coming with face detection.
So again, these are a lot of features
that are coming pre-built
on the mobile asset
and most of these have a common interaction between them
because let's say one application can use all
of these features.
I can also have these features
in itself being interacting with each other.
Let's say based on a GPS location.
My brightness could be adjusted or based on the direction.
My phone is being held the brightness in itself
can also The adjusted as well.
So there are a lot of features
but when they interact with each other these features
come together to bring in a better system
that anything
that they can provide individually
as such that's what internet
of things is it basically is a platform
where we can connect every day things
which are embedded with either Electronics software
or sensors to the internet
and this in turn enables us to collect as well as exchange
between these things.
Now when I say things it can be anything and Nothing,
let's say I have an internet platform where
and I can connect these things
if I take the example of my house,
I can connect my lock.
I can connect my AC.
I can connect my light
and all this can be managed on the same platform
since I have a platform
so I can also connect my car to this.
I can keep a track of my fuel meter.
I can keep a track of my speed limit.
I can also keep track of the location
of the car as well.
Now if there is a collective platform
where all of these are connected wouldn't it be great
because I would love
to Of the AC on and set a cool temperature
at my home by the time I reach back from office
if I have a platform
that knows my preference and that keeps track of where I am
and where I'm going to then it can also identify
that I'm going
from work back to Hope and my preference suggest
that it would be best
if there was an easy temperature of about
22 or 23 degree centigrade and this is something
that is definitely possible through internet of things.
Now, this is just one of the examples now,
let's say tomorrow you come back.
Home, it would be great.
If I wouldn't even need a key to unlock my door
my home system should be aware
that I have come home and should unlock all the doors
that are needed.
And now this can be done
if my mobile and my home devices are connected
onto the same platform based on the location of my mobile.
It can identify
that I am at my hope so it will automatically
unlock the door and let me come in as well.
These are some of the real world implications
of Internet of Things.
These are something that are already happening
but going forward
what Need to understand is
that when I have a specific component with me,
which can do a lot individually wouldn't it be great
if I can collaborate this complaint with my system
of different components and build up at a system.
This is what internet of things is helping us to okay,
you provide a platform to which all these things are connected
through the internet.
So internet becomes the medium through which you're connecting
all these components were things to a platform.
Moving forward let's try to understand.
Why do we need internet
of things now to help you understand why let's look
at an example.
This is a patient at home.
Okay, he's on constant life-support wearing.
His status is being checked
to a health monitoring system present on the cloud.
Let's say at a point there is certain issue
with respect to his health.
Let's say there's some irregularity with his heartbeat
or his blood pressure is low.
There's some fluid being developed or so far now what?
Offenses since the system on the cloud is connected
to a hospital as well.
This information would get passed on
to the hospital as well.
They're in what would happen is
that they would get the complete details
with respect to the patient
and the important information with respect
to the current situation of the patient as well.
We made it aware with respect to what issue exactly is
the patient facing as well as enable
them to dispatch an ambulance
immediately to bring the patient back to the hospital as well.
Now meanwhile,
once the patient has been picked up
and brought back to the hospital there could be prescriptions.
There could be medicine.
They could also be an operating theater made ready
in case of an emergency situation as well.
There'd be doctors on standby
who have the complete history of the patient
who have the complete details
of the present condition of the patient as well.
So this in turn brings in a lot of transparency
and reduces a lot of effort
and time involved with respect to this same thing.
Let's take in today's scenario.
There has to be someone wandering this patient's health.
If there's a fluctuation they need to call the hospital they
need to call request for an ambulance.
And meanwhile, once the ambulance is here.
They take the patient and their back into the hospital
then again there needs to be a lot of checkups
that need to be done
because the doctors are not fully aware again.
There is a lot of tests that need to be run
and this in turn leads a lot of delay as well
in certain emergency cases
as such if a system can do this then this is exactly
where our future lies in and
what we have to definitely move forward to to now Internet
of Things basically is expanding the interdependence
of humans to interact contribute
and collaborate with things around us.
Now.
What do I mean by interdependence of human see
how we depend on each other like
how I'm explaining this concept to do
if you're not clear you would reach back to me.
You would request me for help.
Okay, this becomes an interdependence between us
when I'm giving you knowledge
when you're not here you're coming back to me
and you're helping similarly tomorrow
if there's something that you can do for me,
I would reach out to you.
We all are interdependent on each other for something
or the other
if we can expand
this interdependence to interact collaborate
and contribute with respect to the different things around us.
Then we would be building a proper Internet
of Things environment.
This would be a much more safer secure effortless
and time-saving environment into existence moving forward.
Let's Talk about the various benefits of Internet of things.
Now.
The first thing
that would be as a benefit
of having an Internet of Things platform would
that you could efficiently utilize the resources
that are available
if I have a smart system which can interact
with everything if it has enough computational power
if it has enough understanding
of how things work between each other.
I'm quite sure the usage
of the resources available will be more efficient as well.
This resource could be in terms of monetary.
It could be in terms of natural resource it.
Also be an input taken
up by the thing as an input and so far.
Okay.
So all this can be more efficient
if I have a platform
which is more smarter and interconnected as well apart
from this it minimizes the human effort involved
if my system is smart enough to interact it to filter smart
enough to do things
that I do need to get involved
with then my interaction is always going to be minimum.
This is same with everyone and that's one of the major reasons
why Internet of Things has become popular today.
The and the concept
of smart home is always growing as well in the same perspective
because if the system in itself is able to do most of my work
at hope then I don't need to put in much effort.
I can relax at home without having to worry about anything.
Okay, the next benefit would be it saves time.
If it rains my human effort definitely it is going
to save my time.
Okay apart from that
if the resources are utilized more efficiently then again,
it is going to save a lot more time as well.
All-in-all any benefit of Internet of things
will in turn help you save a lot of time as well.
Okay.
So time is one of the major factor
that can be saved on an Internet of Things platform.
Now if I have an artificial intelligence platform
through which all of this is managed
and maintained then the personalization
and the human touch also comes into the picture.
Now today most of us
have had a level of interaction with an artificial intelligence
or a virtual intelligence as well.
This could be a personal assistant
like see Hurry,
or it could be an assistance application
like Google assist now
if I have system
where all these components
and things are interconnected
then in turn all the security present
on each of these things is going to get multiplied
and it's going to build a much more secure system apart
from that the level of security
that we would be integrating
to the platform in itself is going to be quite huge.
So the overall security booth respect to everything
is going to increase multiple times as well.
Now let's look at some of the major features
of Internet of things.
Now any technology
that is available today has not reached
its hundred percent capabilities and it always has
that Gap to grow Internet of Things is one
of the major Technologies in the world today
that can help any other technology reach its true
and complete potential as well.
Now there are mainly three aspects to Internet of Things.
Things as to how it works first is the connect aspect here.
Basically, what you need to work on is you need to ensure
that there is a connectivity between all the things
around you all the necessary things to The Internet
of Things platform.
Okay, then comes analyze now, I have my things around me.
They each are going to generate some amount
of data now this data needs to be collected
and it needs to be analyzed to build
a business intelligence solution
if I have a good insight,
The data that is gathered from all of this then
definitely I can call my system as a smart system.
Finally what happens is in order to improvise
and improve your system.
You need to integrate it
with various models to improve the users experience as well.
Let's say there's a personalization module
that is there or let's say there's going to be an aspect
when I can directly connect to my providers.
Now.
This provider could be Amazon.
It could be Flipkart.
It could also be my retail store that is next to me.
Let's say I'm out of milk
to retail store will get a notification
and he would send milk right away
and the same thing can be done.
Let's say I have my coffee machine requires special beans
and this be isn't is almost complete then
what my system could do is it could go online
to Amazon and purchase these beans for me as well.
So there are a lot of things that I can do
when I integrate this with respect to various models
and improve the overall experience to a single user
or a group of users on the world itself such.
Okay.
Now, let's talk about each one of One
by one let's start with connect.
Now.
The first stage of connect is device virtualization
because what you need to First do is
that you need to standardize the integration of the device
to the Enterprise platform,
which is present on the flower.
Okay.
Now, it could be present on a cloud it could be present
on a so, but again,
it's all going to be connected through the internet.
So what I need to do is that I need to ensure
that certain level of standard is present on the device
so that it can go on
and connect to my Internet of Things platform.
Now to help you understand this better,
there could be a standard power plug and there could be
a power block which has an inbuilt Wi-Fi support
so that it can connect to my lap.
No to build a smart home system.
I need the second power point
because there only I would have the access to control it
over the internet
and my system could integrate with respect to the same.
Okay.
Now if I take the first standard PowerPoint then
what would happen is
that I need to manually switch it on and switch it
on how in the second case I could send a signal to it
and In don't we'll switch it on and switch it off.
So there's supposed to be a level of standardization
through which I can integrate all of these devices
to my platform.
Next comes high-speed messaging.
So now what I have done is
that I have connected all these devices to my platform
but these devices in turn generate a lot of data
and this data is
what is going to help us understand better
on how we can improve the overall system
and help and provide the user with better experience.
So for that we need to To have high-speed messaging.
Okay.
This basically means
that there needs to be a reliable secure
and a bidirectional communication Channel
between the devices
and the platform not the purpose of it being bidirectional is
because you need to control each one of them as well.
Let's say I want to switch on the AC then the signal
would be going on from the cloud platform
to the device.
So this is how it works on.
Okay.
So every communication
needs to be reliable it needs to be secure
and it needs to be bi-directional as well.
Moving on to the third point
of connect you need to have endpoint management.
If I don't have an endpoint manager,
I have established a way
through which all my devices can connect to my platform.
I also ensured
that the data is going to be sent
from the device to the cloud
and the cloud can send back to the device as well
through a secure Channel,
but if I don't actually identify
from which device which data is coming
and how this data has to be processed.
Then it becomes a failure of the system.
This is where Management comes
into picture endpoint management basically helps you
in managing the devices endpoint identity the metadata
and the overall life cycle involved with respect
to these things are such.
Okay, so to put it quite simply it basically helps you identify
from which device which data is coming
and what needs to be done with this data as well.
Now coming on to the next feature is analyzed
and the first thing that you need to do for analysis,
is this reprocessing now
if the data coming from the device Not on a real-time basis,
then my system is often Williams.
There's no use
if I tell my system to switch on the AC at my home.
And by the time I reach there
if the AC is not even turned on then it's
a failure on my system.
Okay, so real-time analysis
of the incoming and outgoing data and must be done.
I have with respect to different aggregations
it filtering correlations processing and so far.
Okay now apart from this what you need to do is
that this is raw data
that is being streamed from all the things you need.
Need to identify
which is contextually important information
which is going to be taken forward.
So once I have the irrelevant information,
then I can even generate composite streams of information
which can be taken ahead for future analysis
and understanding as well.
This is what your data enrichment process does.
Then you have events to now in events
to basically any information
that you want can be queried
and visualized from the vast amount of data
which is present on my cloud.
Okay.
Now this in turn can also help me get
a better insight and Analysis
if I have all the enriched data present on my cloud platform.
I have a tool
which helps me identify what is needed help me analyze.
This data helps me visualize it then definitely it becomes
more useful as well.
And when I have data coming
from different things as such this in turn can also lead
to being a collection of big data now
when I talk about Big Data,
it's not just a few GB of data
as it's going to be terabytes of data because
because the data generated from the things around us is
that was and if you doing it over a period of time then
definitely it is going to grow into a big data domain as well.
Now coming to the third feature
of Internet of Things comes Enterprise connectivity.
So this is what I was basically telling you about.
Let's say I have a requirement
from my retailer or even an Enterprise organization
which is present.
It could be Amazon.
It could be flipcard anything as as any Enterprise organization
which provides me a service.
If I can connect to them through this platform,
then definitely My overall process also
becomes easier as well.
Let's say this is service provider.
Okay, let's say there is a leakage in my Plumbing
or let's say there's some issue with respect to my electricity.
Then it can contact to the corresponding service provider.
It can send them a detail
and corresponding they would be dispatched.
This would in turn reduce my effort
of having to check the problem having to call someone wait
for them to come back all
that gets reduced to minimum required effort as well.
Well, now how does this communication happen?
So for that we have rest API.
Okay.
Once I've integrated my rest API with respect
to the cloud application
and my internet of things then communication between
the Enterprise communication between the platform
and the communication between the things around us
can be made more efficient
and can be more easy as
per the third aspect is command and control
if I don't have command
and control on my platform then it's awful noise.
Yes, I built a very great.
Table something that's quite extraordinary.
But if you cannot command it
if I cannot control it as per my requirement,
then the system in itself is not useful
if I cannot tell my door in a smart home to unlock
when I want it then it becomes failure on my part.
If I cannot control the AC on my horse motto then again,
it's a failure.
So always the major aspect
when you integrate with respect to these things.
The major thing is
that you need to ensure there's a huge control on system
and you're able to Um
and it as per your requirement now this command could either be
through a voice based recognition.
It could also be a message
that you can send through your mobile application
and so forth as well
when we come down to the iot ecosystem.
There's no single consensus or again.
There's no singles architectural design
that's out there which is agreed universally
because each company each organization each user
for that matter has different requirements
and We look down to it.
We can break it down to a simple three level architecture
where and we have a perception layer
where centers actually gather the information
from the environment around it.
Okay.
Once this is done I'm going to use I'm going
to pass this information to the network layer
the network layer in itself takes up the responsibility
of transferring this data
from the sensors to the next layer,
which is the application layer now here the main objective
of the responsibility is the application
in itself delivers this information to the end user
or the When platform for that matter this architecture
can also be expanded to a five layer architecture.
Now when I talk about a five layer architecture,
it's quite similar here.
It's a the difference mostly comes around with respect
to the transport layer the processing layer
and the business layer mostly here
when we had three layers doing the earlier task.
We've just broken this down
so that we have an easier operation
or a smoother system for that matter.
Now again, the perception layer remains same
when it gathers the information from the sensors,
but the transport layer actually Transports the data
between the sensor to the processing center.
Now, this could be through a wireless system.
It could be through Bluetooth.
It could be two RFID 3G NFC or any medium
that I choose to once.
The information has been transmitted the processing layer
comes into picture
which actually stores the relevant information analyzes
this and again processes it as per the users requirement.
Now again, this could employ various databases
Cloud Computing Services as
well as big data processing modules
to store this information
as as well as process it for that matter.
Once this is done.
I give the information to the application layer,
which is actually responsible
for delivering Superior Services to the end user for that matter
on top of all of this
stands your business lamp now any device for that matter
when it is working on a large-scale environment
a business layer is usually used here.
Now, let's say I'm working in an organization
where we using multiple pumps
for different views features in different locations
for that matter a business layer here actually monitors
the complete functioning of these parts.
You can also have these in various cars as well.
So what do I would note here is
if a car is going to break down then I also get
an awareness with respect to that individual car
and it also helps me enable
or help me reach out to the closest customer care center
so that it can assist the user coming down
to how I can process it again.
This can be divided into two segments.
I have my cloud computing based processing.
We're in here.
It's quite simple once I have the information.
I pass it on to the cloud platform
which then in turn also.
Processes it and also has various applications to deal
with this process information for that matter.
Now again, this is something that I can do on a system
which does not require any immediate action
and requires a large amount of processing for that segment.
But let's say I am in a system
where I need immediate response in those cases,
I can go with my
for computing now again for computing is something
that represents a layered approach wherein we actually
insert monitoring pre processing
and storage with the security layer.
Between the physical and the transport lab.
Let me just go back a few slides here to help
you understand this.
Now.
If you actually look my
for composedly comes between these two layers
when I add four new layers for that matter.
Now again, this is used in order to make a system quite smarter
or effective with respect to it now between my physical layer
and my transport lab,
I have a monitoring layer.
I have a pre-processing layer a storage layer and
a security line now to help you understand this.
Let's take in real-world example out them.
Let's say I have a complete traffic system.
Which is built on my internet of things.
Now, let's say at one point.
There is an ambulance
that has come to a first signal I detect this and what I do is
that I allow the ambulance to move from this traffic signal
by giving it a green line,
but what you need to understand is
if I'm using a cloud
computing this message has to be passed on to the cloud.
This information has to be then process
and then correspondingly a map has all stood to be created
at the same time
when I use a fog approach what happens is
that all the pre-processing
and the storage happens on the Gateway level itself.
The information from the sensor goes all the way
to the Gateway there
and it actually processes this stores the relevant information
and sends this back
to the discourage ponding sensors as well.
So let's say if there's an immediate track
that I can create to the closest hospital.
I would highlight all the traffic signals to be green
so that the ambulance can move smoothly as well now security
here is very essential
because if I implement this tomorrow anyone can
actually try to manipulate this for their requirements as well.
Imagine there is a high High Speed police chase happening
and the culprit uses this
in order to move fast from the traffic signal.
So this is something that is really Ascension.
Although there is a pro to this.
There's also a con and this is
how we try to overcome this issue now talking
about each one of these new layers
when I come to the monitoring layer.
What it actually does is it monitors the power consumption
it monitors each
of these resources as well as their response
and the services
that are running on these resources.
Now this in turn helps me monitor
or gives me a complete idea.
Of which other services are
which are the sensors which are working?
Where are the challenges?
What is the power consumption and how it works with respect
to that same now
once I have information from these sensors
what you need to understand is
that usually you work
with thousands of sensors in a real-world environment.
Now, I need to understand which are the necessary information.
So I'm going to do a level
of filtering I'm going to do a level of processing
and then I'm going to apply a level
of analytics to understand
what is needed
and what is not from these information as well now.
Now the temporary storage area is something that I use
in case I want to store any relevant information.
Let's say I'm creating a route today for an ambulance as well.
And this is going to be stored in my temporary storage area.
But this also needs to be used in future scenarios.
So once I am done with this usage,
I can also push it onto the next transport layer
which can send it to any other storage system
that has part of my environment.
Now as I said security plays a very important role,
although my for computing is something
that makes my system faster.
It should not be easily.
Inaudible now in a cloud-based system.
I have the assurance
that it's not really easy to break down the security.
It's quite hard where there are various layers of security
which are part of the system.
But when I have a fork system,
it's essential that this factor or the security
which deals with the encryption
which deals with the privacy of the information.
The Integrity of this information is maintained.
Now, there's also a very interesting variation of this
which is called an edge Computing system wherein
rather than doing all these operations are
After I have gathered this on the Gateway,
I can do it on the individual notes
or individual sensors as well with respect to it.
So where I have edges these becomes point
for me to perform operations on the data
that is being collected.
So that's the slide variation
of our for computing system is such next let's talk
about the various taxonomy associated
with internet of the X.
Now.
These are the key Concepts or these are the key layers
which are present with respect to most architectures.
That's out there.
As I said,
this is a general guide ER e Person
on each system that's out.
There requires its own level of customization requires
its own level of approach to solve that problem.
But these always remain the fundamental layers
which are included
in all the architectures out there first.
We have the perception layer,
which is usually the layer
where we gather the information from the various sensors
that's out there or we use the various sensors
which are required as part of our system.
Then we have the processing layer
where in we perform filtration.
We summarize the data we again do a level
of analytics on This data
before we decide to send this relevant data to the system
that's above this.
Now.
Then I have a communication line
of communications layer is very simple
as in here will Define the protocols
and standards as well as the medium
through which the information has to be passed from my sensors
to my main system
as well middleware is something that's quite essential here.
What it does is
that it creates an abstraction as well
as it makes my system work much more smoother.
Now what you need to understand is
that there are various components in what hear me?
That really helps
me integrate the information coming out from each
of these sensors or each of these individual systems.
Once it's present
that I can pass it on to my application layer where
and I have various applications
which help to improve the overall experience
of the user as well as
provide much more accuracy and efficiency to the information
that's present now coming down.
Let's talk about each layer one by one now
before I talk about the perception layer
what you need to understand is
that one of the most important aspects of Annette
of things is context awareness.
That is what you need to understand with respect
to the change of environment is very important.
And this is extremely impossible without the usage of sensors
as such now sensors
in themselves are very small in size.
They again cost you very little
and at the same time they consume very little power again.
There are various constraints
with respect to the factors as the battery capacity
and the ease of deployment as well,
but let's not go into them as such.
Now when I talk about sensors again,
we have various types
of sensors as such one of the easiest example of sensors
that can be seen on a daily basis is the sensors
which are part of your mobile.
You have a location sensor you have movement sensor camera
in itself is actually
another sensor your microphone your light sensor.
These are all various important aspects of your mobile
that we use on a daily basis
apart from this neural sensors medical sensors,
like the fitness bands that we use Healthcare bands,
which are used for heart patients.
In environmental sensors,
which check the temperatures around the environment make
you aware of the changes chemical or biosensors
which are very useful on a daily basis
as 12 again infrared sensors are something
that's quite common as well.
When we talk about RFID.
This is something that's really important or not.
This is something
that really gathers a lot of attention with respect
to it RFID stands
for radio frequency identification now
unlike a traditional barcode.
It does not actually require a line of sight of communication
between the tags.
And the reader and can identify itself
from a distance without even a human intervention
or a human operation
for that matter RFID is are technically
of two types you have active and passive active tags actually
have some amount of power source associated with it
and passive sources do not have anything related to it.
And when we talk about the RFID Technologies as well,
there's near and far
and near RFID reader uses a coil through which we
actually pass AC current and Create a magnetic field.
Now when we generate a magnetic field anything
that comes in its vicinity it registers with respect to it.
Now when I talk about afar RFID it basically is
a dipole antenna in a reader.
Now this again propagates an electromagnetic waves and tax
themselves also have a dipole antenna now again,
these are something that's used in various applications.
That's out there.
Now one key factor,
which is associated
with the perception layer is an actuator now
when I talk about an actual It actually is a device
which can affect a change in the environment
by converting any sort of energy into another now.
This could be a motor which is generating electricity.
This could be a windmill
which is converting the wind outside to you
to of electric form as well in these are just some
of the examples
that's out there
and actuators themselves play a very essential role
in the perception lab now,
the next layer that we have is the pre-processing layer.
But before I really talk about the pre-processing layer,
let's actually try to
And the limitations of trying it to process everything
that we have on the cloud system,
which is part of our ecosystem.
Now when I talk about this one of the key
or one of the biggest challenges for me is mobility.
Let's say my sensors are
on devices which are in constant motion.
Then it becomes a really high challenge
for me to pass
this information continuously to my cloud environment again,
this could be true the challenge of Transport layer.
This could also be due to the challenge
of power consumption associated with it
when my Smart device
or when my sensors for that matter are
in constant motion run constant Mobility.
Then it cannot completely pass all the relevant information
on to the cloud.
Now.
This intern actually causes a challenge
for me to have some latency.
This could also lead to early 10 C with respect
to real time processing of the information
that it gets as well.
Now if I'm working on a critical system,
then real-time information is something that I will highly
depend on and that becomes a challenge as well now.
Now if I really want to scale up
if I want to use a lot of devices then
my cloud computing system also needs to scale
but there's always a chance to increase the latency
because I'm working with multiple sensors
or multiple devices.
That's out there.
Imagine today.
I have a system with just includes thousands answers,
but let's say in a smart home system There is close
to about 10 to 20 thousand sensors associated with it.
So this is just one small system.
Imagine if I'm trying to build a smart city in that case.
Is going to be hundreds of thousands millions of sensors
that's out there
if my cloud computing system cannot process this
on a real-time.
Then there's going to be a high challenge
with respect to that sick.
And this is exactly
where the usage of smart gateways comes into a picture.
This layer actually helps me process my data on real type.
It also helps me filter the data on based on the priority
or the requirement
and creates a local copy of whatever is needed
or whatever needs to be taken forward.
Word now when I come to the pre-processing layer
or when I come to the features
of for computing for this matter,
there's a very low latency
because the information does not have to go
to the cloud system
wherein the processing needs to be taken up.
It always is done on the Gateway level itself.
This information really is faster as we've seen
in the previous example about the ambulance now,
I can also use distributed nodes
where in the information does not have to be
other processing has to be
on one single node for that matter
when I'm using Distributed notes then I can also distribute
the effort or the work
that is needed as well
when I'm on a mobile environment as well.
These Smart Systems can communicate with
the gateways present in its closest proximity.
It does not have to connect to just one single Gateway.
If I'm setting up gateways across multiple points
in my city,
then it makes it quite faster and more efficient
and this in turn can also lead to a real-time response
from the Gateway for that matter.
Once I have a real-time response,
then it is big.
Taking my system faster and as
in the previous example rather than just clearing one signal.
I can clear an entire path for the ambulance as such now
once I have relevant information those
which are necessary or those which are really something
that I need to ponder
or I need to analyze can be sent to the cloud system as
and when as it's needed as such so this is something
that really makes
my pre-processing layer important and efficient
and effective coming to the next layer,
which is the most important layer.
It's out that which is my communication lab now
as your Internet of Things environment actually grows.
This is a compromise
or this is a combination of various heterogeneous devices
which are connected to the internet.
What you need to understand is
that these devices in then self need
to pass these informations and some of the challenges
that the communication layer should actually address is
with respect to let's say first start off with addressing
and identifying of each
of these informations wherein I know
which Each device is sending me the information.
What is that device
if I want to communicate back to that device as well.
How do I do that again?
When it comes down to the communication in itself?
This should also not cost me a huge amount of consumption
with respect to power because
if I save up a lot
of energy with respect to how the information is gathered,
but I waste a lot of energy in transmitting this information
then I don't have a smart system again information itself,
which comes should use various route.
Protocols which actually require very low memory
and should be very efficient for that matter.
If the information itself needs to be bounced
around different layers going from different segments
of your Echo System,
then it needs to use very less memory
and this itself should be very fast as well as seamless
for that matter.
Now when I come down to each
of these components for that segment,
let's talk about NFC or the near field communication.
Now NFC is actually
a very short-range Wireless communication.
Technology through which usually mobiles interact with each other
or a distance of a few centimeters for that matter.
Now all the type of data can actually be transmitted
between two NFC enabled devices
in seconds by bringing them close to each other.
Now this in turn is actually based on the RFID concept
and it uses a variation
of the magnetic field to communicate data
between two NFC enabled devices.
If again, we go down
into slide specifications and FC usually works The frequency band
of Thirteen point five six megahertz.
But again, this is very similar to high frequency RF.
Id I'm not going to bore you more with respect
to the technical details will talk about the next segment,
which is your RFID and wsn integration for smart objects.
Now again, many a times
what you need to understand is the data
from one single sensor is
actually not useful for monitoring large areas
and complex activities now here
what you're going to use is
that you're going to use various sensor nodes to it.
Interact with each other in this also has to happen wirelessly.
Now the disadvantage
of a non IP technology such as RFID NFC or Bluetooth is
that it's range is very small,
so they cannot be used
in many applications wherein large area needs to be monitored
through many sensor nodes deployed in various locations
for that matter a wsn
or a wireless sensor Network consists of 10.
Mm sensor nodes connected using a wireless technology.
They collect the data about the environment
and communicated to the Gateway to Ice
and relay the information
to the cloud infrastructure over the internet as such now
when I come down
to the iot network protocol for that matter,
usually what you need to understand is
that the IP for protocols themselves can only
be used for communication of close to 20,000 device.
Now again, the internet protocols used by
these devices is something
that's quite a challenge as well because when you look at it,
the predicted amount of devices
that's going to be available by 2020 in The Internet
of Things domain is Close to 40 billion imagine
40 billion devices communicating with each other.
Now if I don't have a smart system or
if I don't have an effective low power system,
then I cannot communicate
or I cannot gather the information from this now
usually are low-power IPv6 is used for these Network
on these communication
which helps you in passing of information from these sensors
onto your processing
or onto your Cloud infrastructure now again,
when I come down to the low energy technology my main,
With respect to communication is always to ensure
that low energy technology is most probably used
in this segment.
We have your Bluetooth low energy,
which is usually referred to as a ble
and this was actually
developed by the Bluetooth special interest group.
Now what you need to understand is
that it actually has a shorter range for communication
and consumes lower energy
as compared to its Computing protocol.
Now the BL a protocol stack is actually quite similar
to the stack used in classic Bluetooth technology,
however it As two parts,
it has a controller and it has a host as well.
Now the physical
and the link layer are implemented in the controller
and the controller is typically an SOC
or a system on chip with a radio for communication.
Now the functionality
of the upper layers again are included in the host and BLS
actually not compatible with classic Bluetooth.
Now, the next is a low-power.
Y5.
Now.
Again, the Wi-Fi Alliance has recently developed a Wi-Fi Halo,
which is based on your IEEE 802.11.
It had standards this in turn consumes,
very less power than compared to your standard.
Wi-Fi devices also has a longer range.
Now this exactly is why it is most suitable for The Internet
of Things applications for that segment.
Now any device
that supports Wi-Fi also supports IP connectivity,
which is very important for an iot application
for that matter now.
The last is ZB ZB
is also based on the IEEE 802.15.4 Station protocol
and is used mostly in personal area networks or pan.
Now again, the range
for zigbee devices
to communicate is very small usually between 10 200 meters
and the details of the network and the application layers
are also specified by zigbee standards
as such unlike the ble the network layer which is part
of the zigbee provides for multi hopping routing the
when I come down to more details about this agree Network.
I have three tabs
which is an ffd full functional device and RF.
Reduce function device and one zigbee coordinator as well.
Now with this.
I just hope you have a simple understanding
of how communication is essential for The Internet
of Things architecture
and the various ways
that you can implement the communication
between the devices as well.
Now the next concept or the next layer with respect
to the iot architecture is the middleware segment.
Now when I come down to the Middle where one
of the key challenges and one of the key issues
that comes into the picture is the interoperability.
Operate ability as well as the program abstraction imagine.
I have 40,000 devices communicating with each other
40,000 devices may not all use the same programming language
or may not pass the information in the same way as well.
I need to build or I need to have something that ensures
that these devices communicate with each other
and there is an abstraction maintained
between the information pass from these as well.
Now if I have multiple devices also
what I need to ensure is
that these devices Independently discoverable
and I can manage each of them today.
I need to be aware.
If one single sensor also breaks down
because the information coming from the sensor is
extremely important as well
when it comes down to scalability.
It is extremely useful
because when I need to grow my ecosystem this middleware
really comes into picture
if I can replace an existing middleware
with something I can help me scale up,
then I don't have to completely V8 my entire ecosystem as well
with respect to it usually When I use
a highly capable middleware,
then it also lets me perform big data analytics
and Implement security and privacy as well
and this in turn usually helps me communicate
with my cloud computing
and also context detection the again
when you come down to the middleware segment,
you need to understand
with respect to the various specifications
of the application.
Which kind of database is it oriented.
What is the semantics it's based on what kind
of events can process
and what kind of service can it process or White ass
but these are some aspects
that you need to keep in mind
while you're selecting your middleware for
your architecture.
Now the last layer is your application lab Now
application is something that really is
what your end user gets or is
what usually maintains
or helps process your information to the best
that's out there.
Now this in turn can be used in different domains.
It can help you achieve different things.
It can help you have a smarter lifestyle it can help
you have a smarter environment your entire home system
can be Managed
with respect to an application your car management
can be done using application.
You can build an entire social life
and entertainment system based on a smart application now,
although the end user usually only looks
at the application layer.
This actually is the front face of your entire architecture.
Now with this.
I hope you at least have gotten an idea
of what the entire skeleton
of your iot architecture comprises of now
what we have discussed here is just the skeleton.
It's always up to the user to add.
Muscles to it and complete it with a skin as
per your requirement Raspberry Pi today has become so common
that you can find it across most households as well today.
It has become one of the most cheapest
and common Computing device
that can be found almost everywhere,
but let's actually go back to understand the ideology
which bought Raspberry Pi into development itself.
Now Raspberry Pi basically
was bought by the Raspberry Pi Foundation to Introduced
or to bring in the information technology back to the schools
where in students can learn
how to program from scratch the growth
of technology today has grown to such a level
that everyone today
has ease of access to do anything on a computer
but back in the early
and late 90s and 2000 for using a computer you needed to know
how to program and
how to work around with respect to it.
So this in turn have to build
a very strong foundation for programming knowledge
And with the growth of UI everything has become so easy
that today you don't need to learn programming to do much.
But at the same time this has made it harder
for people to identify
and understand good programmers from those who are not.
So in order to build a generation
which starts with a very strong programming foundation
and fundamentals.
The main ideology was to introduce
or take back it to the basics
and make it accessible across every school as well.
Again, as I said,
it was introduced by the Raspberry Pi foundation
in 2012 as watching.
The moving forward what exactly is Raspberry Pi now.
This is one of the questions most of you would have a cross
your mind as well because you might have heard
about Raspberry Pi body might still not have a Clarity.
So let me help you clarify that right away.
Now as per the definition Raspberry Pi basically
is a series of very small single board computers,
which actually have additional features as
Bluetooth Wi-Fi USB capabilities General input Output ports
and so forth now it
basically is a very small low-cost credit card computer
which has actually can be plugged
into any monitor as well as you can include a keyboard
and mouse and it increases the opportunity
for people to explore learn and understand
how to program as well.
Now the latest version
of Raspberry Pi is Raspberry Pi 3,
which was released in February 2016 as well.
Now Raspberry Pi basically is a combination
of raspberry operating And Pike
which basically stands for Python programming language.
But before we move forward,
let me just show you a very simple video
which Raspberry Pi has put out
so that you can understand
the ideology of Raspberry Pi Foundation as well.
This is a Raspberry Pi it's a credit card sized computer
that costs around 25 pounds
designed to teach young people's program
and is capable of doing all kinds
of wonderful things back in the 80s kids
had to learn how to code computer too.
You use them and as a result these kids grew up
with an inbuilt understanding of how computers work.
Now, we need more programmers
than ever before so to deal with this problem.
Some clever people came up
with the Raspberry Pi to reignite the spark.
It runs minutes a free operating system
from an SD card just like the one in your digital camera
and it's powered by USB phone charger.
You just plug in a mouse and the keyboard.
To a TV or Monitor
and you're ready to go in schools.
Not only is best berry pie
a great way to learn programming skills as part of ICT.
There are also dozens
of cross-curricular applications next science and music
and all over the world.
People are experimenting
with Raspberry Pi's and attending raspberry jam events
where people of all ages are learning
what can be done with a Raspberry Pi
since the first Raspberry Pi was ship.
We've seen examples of people using the pipe in a variety
of amazing interesting projects
taking advantage of its size portability cost programmability
and connect ability.
And whether you want to learn to make games build robot
or even teach a bath parachute
with Raspberry Pi the sky's the limit.
So I hope you guys had a great learning experience
with respect to that.
That was basically the ideology of the Raspberry Pi foundation
on introducing Raspberry Pi into the market
as well moving on some of the capabilities
that Raspberry Pi enables you to do is
that it helps you to browse the internet as well as
what complete HD videos on the same device as well.
So all you need to have is a HD supported display
as such now apart
from that even basic operations like making spreadsheets.
Creating words presentation all these can be done
on Raspberry Pi and you have a huge set of games
that are available
which can be played on Raspberry Pi making
it quite interesting
and easy as well for people to enjoy the component as well.
Then you have various add-on capabilities,
like infrared cameras and security system,
which can be built keeping Raspberry Pi
as the core Hardware as well.
Then you can also use many music machines
as well as detection of weather stations,
which we actually had done in our previous session.
On as well.
Now.
These are just some of the top capabilities
that I have picked up from the list Raspberry Pi
today has become one of the biggest component
that has enabled users to achieve
and create a lot as well today.
The capability of Raspberry Pi is restricted just
to your imagination whatever you can imagine can be done using
Raspberry Pi given the amount of effort put into that.
Mmm now moving on let's look The Raspberry Pi Hardware
as well before that.
Let me show you
how I would be using a Raspberry Pi as part
of this tutorial session.
So let me just give you a simple overview
of the component as such now what you basically see is
that the Raspberry Pi Hardware.
So let me just switch
over the camera and let me show you the Raspberry Pi now here
what you're seeing is the Raspberry Pi 3 Hardware as such
and this is your processor and system on chip,
which is an adrenal a few P3 processor.
Now two important ports here is the CSI.
Which is the camera serial interface
where you can connect camera directly to the Raspberry Pi
and the DSi Port which is the display serial input Port.
Okay.
Now, let me just flip this over slightly.
And if you see this is a 2.5 M. M--
micro USB connector.
So your standard mobile charger can be used to charge
the Raspberry Pi
and this is a standard HDMI port for display.
So your standard TVs and monitors
which have an HDMI support can be used to become
a video interface for that.
Then you have an audio jack,
which is used to connect your audio input now,
let me just flip the slightly over and let me zoom out a bit.
Now if you see we have an Ethernet connector
following which there are four USB slots as well.
So when you see here,
there's a 40 pin General input output pin present here.
So this is something
that's quite interesting
and configurable from your Raspberry Pi a software
where you can configure for what each purpose pins can be used
which will be silly little head.
Now the final thing that I want to show you is behind here
when you flip backside,
so there is a micro USB slot present here.
This is mainly to insert your memory card.
And when I say memory card,
I would generally recommend at least a 16 GB memory card.
This is mainly
because the Operating system and itself is about 4 GB
and if you use an AGP,
then what happens is there is very Less storage space
for your operating system to work on as well.
So 16 GB memory card would be really helpful.
And one thing
that you want to see is this is an Evo class,
which is something similar to a class 10 memory card.
So if you get a memory card,
make sure it's at least a Class A Door class 10 memory card
so that you can have high disk read
and write processing now coming back.
Now bow your head let's just skim through the various changes
that has happened over the hardware
of Raspberry Pi now.
We are not going to devote too much time into this
because this is a completely in detail session
how I would be stopping at these slides for 10 seconds.
So if you wish to know more you can pause the video and get
a complete look as well.
talking about the processor the first generation
of Raspberry Pi initially
came with a broad Camby CM 2 8 3 5 SOC
and it basically
was similar to the first generation smartphone chips
and the architecture
that was used also was an armv6 architecture now
the over the years Raspberry Pi 3 has grown a lot
and today it has a capability of 1.2 gigahertz
which basically is because of the a arm cortex a53
64-bit processor it is considered to be 10 times faster
in comparison to Raspberry Pi
wow now I'm talking about the change
in raspberry pi over the years
as you can see here the various models have been listed
and the architecture change
has also been mentioned here as well so in case
if you are interested in Rome or you can pause the video
and take a look at this now
when you come to the memory the first model of Raspberry Pi
came with a 256 MB RAM
and which basically were shared by the GPU
but today with Raspberry Pi Version 2 and 3
you have four times
that you have about 1GB of ram
which again is shared by the GPU as well.
Now the default split was
at hundred and ninety-two MB RAM for the CPU basically,
which was more than enough to play a full HD
that is 1080p video or perform simple 3D operations,
but again, not too complicated operations as well.
Now moving forward
when you look at the networking capabilities in terms
of networking capabilities,
the Model A of Raspberry Pi did not have any such features,
but from model B of version 1 itself,
you had an ethernet port
which was introduced here and from version 3.
You also have Wi-Fi
and Bluetooth capabilities introduced here as well.
Now talking about the peripherals in Raspberry Pi
there are basically four USB ports.
Now, this has been introduced in model B version 1 plus onwards
and today you have a lot more
that you can do with respect to that even audio input ports
and audio output ports have not Over the years
but the on-board storage
has changed a lot with respect director.
Now again,
when you look at the video capabilities video controllers,
basically here you can watch complete HD videos
but Raspberry Pi does not have a hatch 256 decoding Harbor
but the CPU in itself is way more capable
and helps you in decoding the h26 included videos
through software operations now
in terms of GPU the GPU
in Raspberry Pi 3 runs at a higher clock frequency,
which is about 300
or 400 in comparison to the previous versions
which is at about 250 six megahertz.
Now in terms of video input you have a 15 put CSI connector
that has always been present with the Raspberry Pi
and we output has changed over the years.
You have a HDMI port you have a DSi port
and now you also have a 3.5 mmm trrs Jack as well.
Now in terms of the connector capabilities.
There are 17 pins
of the gpio port switch can be configured as
per your requirements as well.
No, moving forward.
These are the various details with respect
to the general input output pin.
This data has been taken from Wikipedia.
So in case if you're looking for more information,
you can definitely check out wiki space
or the official Raspberry Pi documentation as
well now coming down.
Let's begin the Raspberry Pi installation process here.
But first,
let me help you understand the different operating systems
that are available at present.
Okay Raspberry Pi in itself supports.
Has multiple operating systems as such but we will mainly
be working on raspbian.
Okay.
So these are some of the most popular operating systems
supported on Raspberry Pi
so you have risc OS you have FreeBSD operating system.
You have net BSD operating system Plan 9 is again
from Bell Labs as well as Windows has its own Windows
10 of iot version.
Okay.
So for this session,
we're not going to go into the Windows 10 will stick
to the core Raspberry Pi operating system
now for downloading the operating system.
You can go on to the original site.
That is Raspberry Pi Dot o-- r--
g--
And there in the downloads tab,
you can download the operating system.
Let me just show it to you.
Okay.
So this is the Raspberry Pi home page and here if you go on
to the download section,
you can download the operating system.
Now if you are starting off with Raspberry Pi an Internet
of Things trust me.
This is the best place that you can find
a lot of informations with respect to it.
Okay, most people actually
published their details with respect to the projects
that they're working many popular projects
are also available.
You can find a lot of help as well as good information here.
Okay.
So this is definitely the place to go to if you go on
to the download section here, okay this
I'll show you 2 options you have nodes or you have raspbian.
I would recommend that you go with notes
because this is a complete package with respect
to the different operating systems as such so here again,
you have noobs and Loops like so my recommendation would be
you download notes
and make sure you using
a 32 GB memory card with this would be really helpful
because it gives a lot of memory for your operating system.
So in case if you're going
for a 16gb or below I would recommend you go
with the nope slide,
but make sure you keep the bare minimum of 16 GB in case
if you working with Raspberry Pi but do not mistake me.
When I say use a 16 GB memory card.
You can install it on an 8GB as well.
But again, the operating system has less memory to work.
So at least go
for 16 or higher 32 is what I would recommend.
So once you download this ZIP file,
or you can even download it by a torrent.
Okay, but I would recommend download it as a zip file.
Let me show you the file.
So this is the noobs file.
Let me extract this.
Okay, it may take some time.
It's about a 1.5 GB file.
Okay, the lights were seen as slightly lower in comparison,
but this is something that I would recommend in case
if you're working with Raspberry Pi.
Okay, so I have my noobs folder here.
So what you need to do is
that you need to copy everything inside the nude photos
and then paste it inside the SD card.
Okay, make sure you are not directly copy
pasting the folder that you have extract.
So what happens is if you do that,
it does not actually recognize the operating system as part
of the Raspberry Pi so copy everything
that you've extracted and then directly paste it
inside the SD card.
Now again, this is why I recommend
that you use a high speed memory card
because in case
if you are not using it then
what happens is But it takes a lot
of time with respect to fetching the data.
Okay.
Now as I had mentioned
if you're going
for a memory card for your Raspberry Pi
you can use something like a 16gb at least
and make sure it has a high transfer speed.
So this in turn will give you a lot of memory area to work
around with as well.
As I would recommend you take something
like a class 10 memory card
that is available in the market today.
So these have a very good read and write speed ratio.
So definitely this is something that you should consider
while you're getting a memory card for your operating system.
Okay now Once you're done copying with this,
let's go on to install this memory card
onto a Raspberry Pi and let's begin the installation process.
Okay, so now we've successfully copied all
that is needed for our operating system.
So safely remove this make sure you rejecting it safely
because sometimes what happens is
if you do not see
if rejected the files get corrupted
before you load it into your Raspberry Pi this
is certain issue with respect to the memory cards.
So just eject this.
Okay.
So this is my Raspberry Pi and if you actually flip it over.
This is the slot where you have to insert the memory card.
Now, we've copied the noobs folder
into this memory card.
Okay, let me just insert this into this and let's begin
with respect to the installation process.
Let's begin installing this memory card
onto my Raspberry Pi and start with the setup.
So, let me just turn the on my Raspberry Pi.
Okay, so this is a screen.
I'm getting by default.
Now what happens is the installation steps initializes.
So the Raspberry Pi is getting loaded.
Now there are two operating systems
that are present here by default and we want raspbian.
So let me just select raspbian and click
on install option present here.
If you see here it basically is telling me
that it requires 4.5 GPA.
So this is why I
was recommending you to get at least 16 GB
or more because
if I have an 8 GB memory card
if 4.5 GB goes off with the operating system itself,
then there's not much memory left
for me to work around with now.
Let me just click on install.
Just before I do that in case
if you are using a different language,
or if you want to use a different keyboard as well,
you make sure you're specifying that okay.
So there are written language is present here.
So choose the corresponding language as
per your requirement.
Once you selected that description install
and this is just basically going to ask you
whether you want to format your data
and install the operating system on top of this.
So click on yes,
and the installation process has begun now it may take some time.
So I request you to be a bit patient
because this is an Operating system installation process,
so please be a bit patient and hope you have fun mean wire.
So now as you can see the operating system installation
has completed successfully once you hit on okay,
it's going to ask me
whether I want to reboot or it automatically reboot as well.
Okay.
So yeah, it's automatically rebooting at this point.
Okay.
Now as you can see the Raspberry Pi operating system
has successfully been installed.
Okay.
So this is what your Raspberry Pi operating system
would be looking like okay.
So this is raspbian as such so there are
different operating systems that are available today
for you to install that we have discussed already now
that we've installed raspbian on our system.
Let's look at some of the accessories
which can be used with Raspberry Pi now one
of the most popular
and interesting accessory to Raspberry Pi is the sense hat.
Okay, so that basically is a hardware attached on top.
There are various hearts
that are available but censored is the most popular
because you have various onboard sensors present here.
You have a temperature humidity
and pressure sensor present here as well as a gyroscope
and a joystick also present here apart from this.
The reason that it is quite popular is
because it has an 8 cross 8 LED Matrix display
that is present to him.
So let me just switch over
and show you the sense hat component as well.
And okay.
So here you can see the sensor know how it is
slightly different from the the other accessories is
because it's got an 8 cross Aid LED Matrix present here,
which can be used for displaying various things
and we will also be seeing how you can display this
as well now coming on to the other components here.
Now if you see here,
it's already got multiple sensors here.
It's got an accelerometer
and the gyroscope meter present here.
It's also got we just this part the axe low
and gyrometer present here.
So you can see it's got labeling as well.
Then you have a humidity pressure
and temperature sensor on board with respect to this
and it's quite interesting and useful as well.
Now one of the easiest things with respect to this is
that it can directly be placed on top
of the Raspberry Pi's gpio pins
and thereby making it quite easy to assemble it as well.
So it directly sits on top of the gpio pins as well.
So this is something
which makes the entire Raspberry Pi ecosystem quite easy
and effective one other interesting key point or factor
that makes the sensor quite interesting is
that it's gotten on board joystick as well.
So in case if you are a gamer and you want to try out
the games on Raspberry Pi you can use this choice to catch.
As such now another additional component that most people use
with Raspberry Pi is the camera now in 2013 itself one year
after Raspberry Pi 1 was launched.
It had introduced the Raspberry Pi camera
along with the firmware update as well.
Now the r a spy cam is basically an 8 megapixel camera
and can directly be connected
to the Cs airport present on the Raspberry Pi itself.
Now, it's also very interesting
and can record up to thousand ATP videos as well.
So moving forward.
Let's also look at the infrared camera.
So this is a very interesting thing.
So this was something
that helps you capture video in infrared mode
and is called pine oil.
Okay.
So again, these components can be bought online
and is one of the most easily
by now the final accessori announce the most interesting
and essential accessory to the Raspberry Pi is
the Gert board.
Basically it is used for educational purposes
and it helps you expand the operations done
using the general input output pins here,
so, Using this you can connect to various LEDs
which is Analog Devices sensors and much more.
This also helps you connect
to Audrey know which is also another DYI Hardware device
which has gained a lot of popularity but in comparison
to Raspberry Pi order,
no is still growing Stitch moving on.
Let's begin with our sensor tutorial
and what we'll be doing is
we'll be exploring five different demos to understand
how the sense had component can be used effectively.
Now the first thing that we'll be doing
is we'll Implementing a very simple code
and we'll be trying to display a certain letter
on top of the sensor.
Okay, so I'm going to display p&d but the value
of that color is going to be generated by random.
Okay, so let me just run you through the code
that's present here.
So the first line is from censored import sense.
So what it basically is doing is
that it's helping you import sense
at and establish a connection to the censored component.
We're going to use time for mainly sleeve function.
Oh and random for generating a random integer.
Now.
What I'm doing is
that I'm initializing the censored which is
sense equal to sense
at basically initializes it then what I'm doing is
that I'm generating a random integer
between the range of 0 to 255 now for those of you
who would have guessed why I'm doing this.
Definitely this is
because the RGB value lies in this range,
so it's somewhere between 0 to 255 then
what I'm using is that I'm using a censored function
which is show later
and then I'm specifying which It means to show
and I'm going to specify the color combination.
So this is the RGB values present here.
Okay, so it's going to show the letter e and it's going
to create a random input with respect to that same.
Okay, then it's going to sleep
for one second again generate a new random integer
and then it is going to show D.
Okay.
So again, it will sleep and then it's going to clear.
So let me just show you this to you practically.
So now what I'm going to do is
that I'm going to connect to my Raspberry Pi.
Using VNC Network.
So it's virtual Network,
which I'm connecting it to because it's not quite easy
for me to record the video
on the Raspberry Pi so I'm going to use VNC server for that
and I'll just give you a simple idea of how you
can connect to VNC server to your Raspberry Pi as well.
Now one essential thing
that if you are connecting to the VNC server,
then you both need to be on the same network.
Okay.
So what I've done is I have used a hotspot
connected both my Raspberry Pi and my laptop to the Network
so in case you are wondering
how to connect to your Raspberry Pi y remote server,
this is how you do it.
Just go to your settings go to preferences.
You have Raspberry Pi configuration here.
Now in your interfaces make sure that VNC is enable.
Once you've done this then VNC server gets enabled
and onto your system just download VNC server
and VNC viewer and you can connect
your Raspberry Pi quite easily.
Okay.
So let me just show you the code
which we're going to be executing now,
it's the same that we have seen earlier.
So, let me just make a small modification to this.
So this is awkward again
as we have seen we are first initializing the censored
with sense equal to censored
then we're generating a random integer
and then we showing a letter
e on that so let me just change this.
Let me just set it to 0 and let me just copy this once more
and I'll also show a third combination here as well.
So I'm just generating a random value between 0 to 255
and I'm displaying based on that color.
Letter e d and f okay,
so let me just save this once so let me show you the output
that is going to be displayed on the sense
that as well.
So so let me just execute the program.
Okay.
Now let me just switch over to the censored
so that you can see the output.
It's a ready green d and a blue f as well.
So this is what So based
on the inputs that I have given it has displayed
a certain letter on the censored LED display as well.
So let's go back to our presentation
and look at the next program that is involved here.
So the next program basically is going to display an image.
Now what it does here is
that it basically is going to display an image
that Gives you a complete idea
of the various color combinations possible
on the sense at as well.
So what I have basically done is
that I have identified
the RGB colors red orange yellow green blue indigo violet
and E stands for empty,
which is null.
Okay and based on this.
I've created a random Matrix here,
but each of these values defines for a pixel of the sensor now
as I have mentioned to you earlier,
this is an Eight cross it led Matrix.
So each of these values
is for one of the individual pixels themself
and with respect to how they start the numbering starts
from left top position.
So accordingly it goes in a row by row, man.
Okay, so let me just show you the output for this as well.
So this is our program and let me execute this
and show it to you.
And now let me just switch over to the Raspberry Pi.
To here you can see the different color combinations
that are present on the Raspberry Pi sensor.
The first row was left completely empty.
The next row has to our values present here.
So this is how it has been defined.
If you notice the code here,
there are two are following by three empty spaces.
And before that.
Also, if you can consider three empty spaces
that is one rule.
So every specific value is pertaining
to one specific row as such now coming back to our presentation.
And so the third one is quite interesting.
What we're basically going to do here is
that we're going to rotate
a letter based on the different orientation of the screen.
So let me show you
how this is done rather than just tell it to you.
So what we basically going to do here is
that we're going to rotate a letter
J on different angles as well.
So we've defined the different angle 0
90 180 270 and these are the various angles
which is going to rotate for.
Okay.
I'm going to run a for Loop
where it rotates the same on different.
Angles as well.
So let me just show you that so just let me add
since dot here here.
Now.
The reason I'm doing that is mainly to ensure
that the sense that is cleared after the rotation process.
So let's execute this.
Let me switch over to the Raspberry Pi
and if you see here A Jay has been displayed
and it's rotating.
Okay, it has stopped.
So let me just rerun this once again.
So what is happening here is
that I'm rotating the letter J by 90 degrees
and it's completing to complete circles on the Raspberry Pi.
Okay, so it's a very simple thing but it's important
that you understand how this is done
because what we're basically doing is
that we are setting the rotation using the sensor.
Okay, and then we're sleeping
at point zero five milliseconds as well.
Again, you can increase the delay here.
You can change the letter here.
You can even change the angle to understand
how it works as well.
So this is something that you should definitely explore
about and get a better understanding with coming down
the next and most basic thing
that you can do.
Using sensor is that you
can measure the temperature pressure and humidity.
So we've already tried to do this
in our previous session as well
where we've measured the temperature the pressure
and the humidity now here I'm just rounding it off
and based on a specific value.
I'm going to ensure
whether the display is going to be red
or green as well.
So let me come back to my senses at let me just open the code
and we will modify this
on the go to get a better understanding.
Let me explain.
In the code once again,
so first I'm establishing a connection to the censored
then I'm running an infinite Loop
where I'm getting
the temperature the pressure the humidity.
I'm rounding it off and if the temperature is about 36,
so this is just a value
that I have set then the background color
should be green
and the color of the text should be complete wide.
Okay.
So this is a simple modification I have made else
in normal situation
if it is less than 36 degrees then what it should be.
Is that it should have a background color of blue
and the text color should be yellow.
So this is basically to help me identify the different ranges.
Then I am creating a message variable
which is storing the following format
where temperature is equal
to temperature value pressure is equal to pressure value
and humidity is equal to the humidity values.
Finally.
I'm going to show this message on my sense
at LED display screen.
So let's just see how this works and execute the program.
So let me just execute the program and let
me also switch over to the Raspberry Pi.
So here let's begin with the execution of the code.
So now you can see by default.
It is having a blue background
and the text is actually in yellow.
Okay, so if you would have noticed the temperature is
about 34 degrees Centigrade.
So to this letter
or Scroll once more let's just validate
the value of temperature.
You can also print this into the screen.
You can also send it via an API and much more.
So if you can see,
it's thirty four point two degree Centigrade now
to this what I'm going to do basically is
that I'm just going to introduce a hot water source.
Okay, so I'm just trying to raise the temperature here.
So let me just shift a little bit
and please keep a note on the temperature value.
So the temperature has started Rising.
So it is taking some time.
I think the water has cooled down a bit.
All right, so it's started now the color background color
has changed from blue to green
and the text color has changed from yellow to white as one now.
These can be used in critical measure environments as well.
So let's say you are in a situation or an environment
where you need to have a constant Monitor
and constantly be aware
with respect to the environment change in example,
like space or
any other environment that you're working with.
Definitely.
These are certain things that can really be important
and helpful as well.
So Coming back.
Lets us.
Look at the last program of the sensors tutorial,
which is basically detecting the motion of the sense
that now what I'm basically going to do here is
that I am going to rotate my science at and I'm going
to identify the value of pitch roll and yaw on basically,
these are the orientation with respect to the x
axis y axis and Z axis
and this is just from the orientation of the sense
that as well.
Now this is mainly done using the gyroscope
but to get a better understanding and more precise.
Value will also try
to include and incorporate the accelerometer as well.
So let me come here.
Let me just show you the program.
So this is the program as well.
So let me just stop the execution here.
So as I had mentioned
this is going to run infinite number of times
because the while statement here is indefinite.
So let me just stop this and let me just run
the orientation program
and Let's just see the default values
of pitch yarn and roll first.
So now by default
if you see pictures about zero role is about
96 and Ian is about 0 itself.
Now, let me just switch over to the sense
at and what I'm going to do parallely is
that let me try radiating the value of the sensor as well
or let me try rotating the sensor to give you
a better understanding of how it works simultaneously.
You can also see the change in Pitch Ian and road.
So if you see Here I'm slightly lifting it
and the value of which has risen from zero
to six role also has changed Ian also has changed.
I'm going to try tilting it same pitch roll and yaw
on also has changed.
So I'm just filling it
with respect to various accesses and you can see the change here.
You notice the changes I definitely hope you do.
Okay.
So now let us actually stop the execution of this program
because this is also an infinite Loop
and to same orientation program.
Let us actually try to incorporate
the accelerometer and understand
how more precise we
can make now instead of just taking the pitch roll and yaw
on I'm here going to directly take the value
of acceleration here.
We have taken the value of orientation and here
we're going to take the Of Aksum duration,
okay how basically
the first is with respect to the gravitational force
that acts on this.
So now let me just show you the code here.
Now again, what you need to understand the difference
between gyro measure and Axle row measure is just
that its senses both the static as
well as Dynamic change with respect to that
but your gyroscope many measures the rotation
of a specific device as such.
Okay, so let me just execute this program as well.
so let me just first begin by clearing the screen
and then we'll execute the code once again,
so let's just execute the acceleration code to get
a better understanding and simultaneously
what I'll do is
that I'll just show you how the change with respect
to the axis is or how you
if you change the Raspberry Pi
how the value will also change so by default
if you can see the value of x is 0
the value of y is also 0 and Z is 1 so now
what To do is
that let me Begin by slightly rotating the Raspberry Pi now
if I change the Raspberry Pi
into a vertical position like this.
Okay, so it's a perfectly vertical position.
But if you notice the value of x it is become minus 1,
okay, why has become 0 and Z has become 0 same time.
If I rotate it in such a manner.
I'm just keeping it on different axis the value of x
becomes plus 1 instead of minus 1 same with respect to Y
if I change it in such a horizontal way it
is minus 1 and if I change it
in such a manner then it becomes plus one.
Okay.
If you flip it completely over then the value
of Z becomes minus one as well.
So if this is the vertical position then the value
of Z remains one,
I hope this is clear for you guys.
So again, these are all different parameters today
what I have done as part of this is
that I've just help you understand
how you can explore the various parameters
and various components
that are associated
to the Raspberry Pi censored as well.
Now what how you use it
and what you use it for is completely left
you these are just some of the basic things
that you can do using Raspberry Pi
the limitation of Raspberry Pi stops
where your imagination stops.
Okay, now that we've installed raspbian on Raspberry Pi it's
time we begin with the demo for this session.
Now, what we're trying to do is
that we're trying to build a weather detection system
to get the real time weather analysis
when we talked with real term analysis.
It basically is the pressure temperature and humidity.
And what we want to do is
that we want to get a live mobile notification for this.
So for that will be first using censored board
which will help us measure the various metrics
like temperature humidity and pressure.
Iran to a Raspberry Pi then we'll use Raspberry Pi
to transfer this data and display it on the sense
that as well.
So we'll just be displaying the temperature as
of now on the set.
And finally we'll be using insta push API
and Python program through which we can send this notification
across the Internet to our mobile application.
So let's begin this demo part first.
Let me show you what's and satis
and how you can assemble sense at on Raspberry Pi.
Okay.
So this is my Raspberry Pi unit
and this is That's what I was referring to.
This is sunset boat.
Okay.
Now if you see this is a simple additional board
that you can add on top of your Raspberry Pi.
Okay.
Now it's quite simple,
but how it's different is
that in comparison to your Raspberry Pi
there is a neat cross 8 digital LED display present here.
So you can use this
for various different purposes as per your requirement.
Okay.
Now, it's simply sits on top of the Raspberry Pi.
Okay, just make sure the alignment is correct.
And once you've done with that just push it on top.
And the pins would merge accordingly.
Once you do that you can just place the four screws
across the different ends and tighten it up.
And once you're done with this,
you can integrate and use the censored
for various different purposes.
We will use sensor to measure the temperature the pressure as
well as the humidity across in this room
that is present.
Okay, so this has different sensors present on it
and we'll be using them for today's demo.
Okay.
So let me just put this up and once it's all set up.
Let's go back
to our demo and let's begin with understanding the programs
before we begin the execution.
Okay.
So moving on now
that you've understood about Raspberry Pi's
and Saturn V.
We also install raspbian on our Bye,
let's move forward
and start with respect to today's demo part.
So again going back just to remind you.
We trying to create a better control system
wherein we try to measure the temperature the pressure
and the humidity around the sensor as such.
Okay.
So the first thing we're going to do is I let me just connect
to my Raspberry Pi.
Okay.
So this is my Raspberry Pi interface.
Now.
Let me just show you the program as well.
So I've divided the actual program into three parts
so that it's easier
for you to understand
how this program works as such now,
the first program
that we are going to execute is the sensing environment.
So this is a simple program
that's going to sense the environment variables.
That is the temperature pressure and humidity.
And then it's going to correspondingly display
these values on the censored LED display panel.
Now, let me explain this program to you one by one now firstly we
need to import the He boosts the values on the censored.
So for that we're going to use this line from sense
underscore hat import sense at so this will give me
the corresponding values
from the sensor to my Raspberry Pi interface.
Then I'm also importing type.
Now.
The main objective
of time is to help me keep a log of all the values
that are measuring.
Okay, so that will be seeing little ahead.
After that.
I'm going to assign the censored values
to a variable known as sense.
Okay, so through this variable.
I'm going to correspondingly relate to all the values coming.
From the sense at okay moving forward we
have written a while loop and this is an infinite
while it's going to continuously
measure the temperature pressure and humidity
and is going to correspondingly be displayed on the sense
that as well.
Okay.
So let me help you understand
how the temperature humidity and pressure are measured.
Okay.
So what you need to do to measure the temperature
is used the function sense dot get underscore temperature.
So this function
that is get underscore temperature will give me
the temperature value
measured by the sensor in degree centigrade.
It and hear what I'm basically doing is
that I'm just converting it into fahrenheit.
Okay, so the value
in degree Centigrade into 1 Point 8 plus 32 is just
a Fahrenheit representation of the current temperature.
I'm using a round function to round
of the corresponding value.
I don't need any decimal point values here.
I want a whole number.
So what I'm doing is
that I'm using a round function and once I have the temperature
then I'm going to be storing it inside a temp variable.
Okay.
Similarly.
We are going to use sense not get humidity
and sense dot get pressure to get both.
The humidity as well as pressure
and store it in humidity and pressure variables are such.
So basically I have all my three variables
that are needed now.
What I'll do is I'll create a string called message.
Okay into which I'm going to store the temperature humidity
and pressure with a specific notation.
So T is going to be equal
to the temperature H is going to be equal to the humidity
and P is going to be equal to the pressure.
Now comes the part
where have to display this value onto my sensor display screen.
Okay, so here I'm going to use the function.
Sense dot show underscore message.
So this is basically a call through which
I can send any message to my sensor display screen.
Okay.
Now here I am just passing the message with respect
to the temperature humidity
and pressure variable
and then I'm also defining a scroll speed.
Now.
This is something that you can play around with
if you want to have a fastest-growing you
can modify it.
I have chosen a 0.08 scroll speed then the other two factors
are the font color and the background color.
Okay.
Another text color I have set as 200 240.
Tea and 200 now if you can guess these are
basically the RGB values for the corresponding text color.
So this is something that I have said as
per my requirement you can play around as well
as background color is something
that I have set to 0 0 and 0.
Okay.
So you have complete freedom
to modify this as per your interest.
I would definitely recommend that you try playing around
with this to get a better feel of how it works.
Okay.
Now what I'm doing is
that I'm putting my system to sleep for four seconds.
This is basically to introduce
a delay between the measurement carried out.
On the sensor now,
if I don't do this
since it's a continuously running infinite loop
it's going to keep measuring at one after the other.
So I want a simple delay between the measurement.
So I'm putting the system to sleep for four seconds.
After that.
What I'm doing is that I'm opening a file
called weather dot txt now here.
I am going to create the log into this file.
I'll take the present time and then pass the message as well.
So these both with respect to the present time as
well as the message are going to be stored in form of a law.
Okay, so in case in future
if I want Want to refer to the measurement,
this is something that will be really helpful.
Finally.
I'm also going to print message on my console.
Now.
This is something just for our understanding as well.
We'll be seeing it on different places will see it
on the console will see it
on the sensor as well as we will be seeing it
inside the log file as well.
Finally.
I'm closing the log
and then I'm putting my system to sleep for 5 seconds.
So these sleep delays are just to ensure
that the consistent values are being measured
in regular interval
and there is no hindrance with respect to Okay.
I hope you guys are clear with this
because I'm going to go on to execute this program.
So let me open my console.
Okay here let me write the command for executing
a Python program and it's very simple.
It's just python followed by the filename.
Okay.
Now in our case it's sensing environment.
Okay, so let me execute this.
And now if you see the values are being computed
and on the sense
that the values are being shown
as well so you can see here the change with respect to that.
Okay, firstly it showed me all the temperature humidity
and pressure and now it's already come
on the console as well.
Similarly the going to have a simple delay and then you
can Already it showing me the next values on my sensor
and it's going to be stored in a log file.
And then it's going to come onto my console.
So very interesting.
Right?
So definitely you can play around with respect to the font.
You can play around with respect to the background color as well
as if you are interested,
you can try bringing different temperature items
near the sensor to see the variation with respect
to that as well.
Let's say if you have a glass of hot water
and you bring it near the sensor
the temperature value will correspondingly change as well.
Okay.
So this is something that you can really Experiment on
and have a good understanding of how things work.
So let me just close
this program and let's see the log file.
Okay.
So this is my weather dot txt file.
So I have taken different temperature measure variables.
It's been log from a long time
that we've been trying this out on.
So definitely there are a lot of values here.
Okay.
So let me just show you the latest value.
So this is a range of values
that I had just tried out some time back.
So all the log all the temperature
and pressure log are being stored here continuously.
Okay when you feel comfortable.
This is the latest log
that we have just created with respect to the three values.
This is the current status as such.
Okay.
So again, this is something that you can experiment on.
I am just told it because of my future reference as well.
Okay.
So I hope you guys have got a simple understanding of
how you can connect
to your sense at measure the temperature pressure
and humidity and also display it
on the console as well as on your sex act.
Now coming on
to the second part second part is the slightly Advanced version
of the same program here.
We're trying to do something a little more apart from just
seeing it on the screen.
Okay.
Now I want to send an email with respect to all the logs
that I have or let us say at a regular interval.
I want to have an email
which notifies me with respect to the present temperature
and the current pressure and humidity.
Now, this is something that you can definitely use
while you're working on a social experiment
as such usually in the Laboratories.
It's important that you always keep Check
on the temperature pressure and humidity.
So this is something that you can try out.
So let me show you the code.
Alright, so here I am just
including something new as well apart from our previous code.
Okay.
Now if you see our previous code is still present here.
I'm taking the temperature.
I'm taking the pressure humidity storing into a message
and then I'm also printing that message
but apart from that I'm using a few libraries
which are very essential for mailing me this detail.
Okay.
So for that,
I'm going to use first SMTP Library.
Okay, because your I'm wailing Protocols are
SMTP based protocols.
So these are something that I will be using
in case I need to send a mail with respect to all the details
as such so these are standard protocols
through which I be communicated.
Okay.
Now coming down with respect
to what Mi Ami is a Miami basically is
an extension or multi-purpose internet mail extension.
Okay through which I can send emails
which support text characters
non text characters audio video images and so forth.
So sending a mail.
Is an essential important Library as such?
Okay moving on forward.
Let me explain you
the code now here we are repeating the same thing.
We are also importing
the time function and then we are establishing a communication
between the sense at and the program.
Okay.
So here first you need to specify the from address
that is from Coos mail ID you're sending this.
So I have created to test Melodies one is
a durocher test 11 at gmail.com.
So from this address the mail is going to be sent
and the address which is going to be receiving.
This is a deal-breaker.
A Raspberry Pi.
Okay.
So from this email-id I'm going to send a log
with respect to the data
that is measured from I censored Raspberry Pi.
Okay, then I'm going to use
my Mi Ami multi-part function from here.
I'm going to distribute my mail into multiple parts.
So this is what your Mi
Ami multi-part function will help you to that.
So in my message part the from is going to be the from address
that is from whom it is going the two is going
to be the to address and the subject.
I am defining it as temperature Raspberry Pi so temperature
Then I'm measuring all the temperature as
well as the pressure
and humidity and storing it inside a message variable.
So here what I'm doing is
that I'm basically going to attach the message
in a plane format to my message variable,
which is going to be the details
of the temperature and pressure.
Okay.
Now, let me come down
and help you understand this part of the code.
Now.
I'm going to establish a connection to my server.
So I'm going to use your server variable
where and I'm establishing an SMTP protocol do Google.
Okay.
That is your or smtp.gmail.com sensor.
This is a standard mail protocol SMTP for Gmail.
It is smtp.gmail.com other meals you can just replace
the corresponding me and then you have the port number.
So again with respect to SMTP.
This is a standard Port
that you would be using for communication.
That is 25.
Okay, then I'm going to start my server.
So that is server dot start TLS.
Okay, once I start this
then I'm going to specify the login credentials.
This is basically for authentication Factor.
So server DOT log
in you have to provide the Username and password.
So my from address is my username and my editor
a kpi is the password for this email-id.
Okay once have specified this then what I'm doing is
that I'm converting the message as a string.
Okay, sometimes what happens is
while you are trying to send this.
It may not get sent
as such because it has values also associated
with it some directly converting it completely to a string.
Okay, and then storing it inside a variable known as text.
Then I'm calling the function sin.
Email through which I'm specifying from which address
is going to which address it should go and the text
which is part of it.
Finally.
I'm quitting the connection between the server.
I hope you've understood till here with respect
to the program now,
it's time we execute this program.
Okay, so let me call python email dot p y
okay.
So now what it has done now, it's connecting to my server.
That's the delay here.
Okay, its measured the temperature its measured
the humidity as well as measured the pressure
and then it's going to send the mail.
So let me show you this meal ideas were so
as you can see here.
I've already received a mail
from anyway car test lab it's called temp rasp
that we had set
and it's given me the corresponding value
of temperature humidity and pressure.
Uh sure so this is what I was referring for
so you can do this across to let me show you my sent box as well.
Okay.
Now if you see a new Sandbox,
I can actually see
that this male has been said so there are two ways
for me to verify this as well.
So this is something that is really helpful
for me now coming back
to my Raspberry Pi let us now see the final program
in which we want to incorporate this
and do a little more.
So let me explain you the code little more better here.
There are few things do here.
So first you have pikul
and Jason being imported now pikul, basically,
Is going to help you establish a connection between your API.
So we're going to use an API
through which I'm going to pass my data.
So for that I'm including bikers here apart
from that you have string IO,
which is a standard string input output management package
and you have your our PI dot gpio.
So that is Raspberry Pi's General input output.
I am just referring to it as gpio.
Okay, so moving forward whenever you see gpio it
is the general input output from the Raspberry Pi
and then you have the similar function you have sense.
Go hat from censored and time function
and I'm establishing a connection here
as we have done earlier between my sensor
and my application
and I am clearing the screen on sensor.
Now.
What happens is sometimes
if you've executed a program before
and you've stopped the execution of the program then
what happens is the display
will still have the remaining data present on it.
It stays stuck on the sensor.
So before we move on to with respect to our program,
we will clear the data out.
Okay, so that's what sense dot cleared us now here.
Two things that are present here hot and cold.
Okay now why they're being used I'll come back
to that a little ahead.
Okay apart from that.
There's also a new variable known as push message.
So we'll be talking about this also
when we come down later
and you'll understand what the push message is used for.
Now.
Let's try to understand the code
which will help me display numbers on my sensor.
Now, you might be wondering
if I have already displayed number and data on my sensor.
Why do I need a separate part of the score
because here we're trying to do something different.
And from what we have already done,
so here firstly what we're doing is that we're setting an offset
from the left and the top.
Let me go back to my presentation here.
Okay.
Now here this is
the standard program that we are talking about.
But let's see.
This is what we need.
Okay.
So this is what we're going to explode ahead.
Now.
This is my actual sense at okay.
This is a neat cross 8 LED display function.
Okay.
So here what I'm going to do that I'm going to eliminate
the First Column
and the first two rows from the top.
Okay, so this We will give me
a 6 cross 7 Matrix area remaining with me.
Okay, this is
where I'm going to do all my operations from now onwards.
So this is something interesting and I'll help you understand why
I am taking this region.
Now before we move on let's take a simple look with respect
to the core present here.
Okay.
Now before I do that,
let me try to help you understand why I have taken
a 6 cross 7 Matrix area.
Now.
This was the remaining area.
I have divided this remaining area into two parts.
Okay.
Okay, I'm just going to give a gap of one column
between these two
and the left hand side is going to represent the temperature
or the value in tens place
and the right hand side segment is going to give me the value
in once place.
So this basically is a 3 cross 5 Matrix where
and I have 15 positions to play around with boom.
But this number again,
this is 15 areas that I can play around
with in a 3 cross Phi Matrix.
Now, let's go back to our code now.
Let me help you understand how it works now.
I have basically 10 numbers with me.
Okay.
So each one is going to have a specific way of representing
in a 3 cross five Matrix.
So that's what I have written here
in this number Matrix as such.
Okay.
So each of the line is a representation
of a corresponding number in a 3 cross 5 Matrix
and this one and zero basically is a configuration
for the LED to be either turned on or turned off.
So normally when we write 0 the first line
will always be full.
The first three numbers are one.
Okay, then you have 1 0 1.
Okay again 1 0
1 1 0 1 and my last three numbers will be again one so as
and when you represent zero in a normal form,
this is how you do that
so it may not be so clear to you right now.
Okay.
So what I would recommend is
before you try to understand
how each of the numbers are displayed set an offset
on your sensor there try playing around
with respect to different values to understand
how you can represent each
of the numbers once Made two or three mistakes,
then you'll get a clear-cut understanding of how each
of the numbers can be displayed.
Now.
This is for a 3 cross Phi Matrix tomorrow.
You want to try it across the across it you can do
that as well.
So definitely make sure to try
out make mistakes only then will you learn?
Okay.
So each of the line is a representation for a digit
in a 3 cross Phi Matrix.
Now, let's try to understand
how we can display a single digit for that.
I'm going to define a function called show digit, okay.
Okay.
So this is the function it takes a value as an input.
Okay.
It takes a value for x
d OK XD is the x-coordinate for where the digit
has to be present yd is the again the y-coordinate as
to where the digit has to be represented
and the last is a simple RGB value present here.
Now first thing what I'm going to understand is
that I'm going to compute the offset position.
Okay.
Why do I need to compute an offset?
Because each of the numbers are present
in different rows of the call.
So I Need to understand which is the present number
and which value do I need so for that
what I'll do is the incoming value into 15 will always put me
at the start of that room.
Let's say the value is 3, okay.
So 3 into 15 tells me
that I need to move 45 positions.
So as I said each row has 15 so 15.
Okay, then I have 30 in the next row.
Okay, and once I'm done with the 45th transfer then I'm here.
This is the standard representation for three.
3 so I have this complete row at my disposal presently.
I am pointing to the start of three value.
Okay.
So this is something important to remember.
So this is why we are Computing an offset
by multiplying the value in 250.
So I have the complete access to this row.
Okay now we'll try to execute a
for Loop through which I'll place these values
into the different parts of the LED display.
Okay.
Now see I have three cross 15 area.
So I need to identify in which area should
which Look up that is which led should be turned on.
I have three rows and 5 columns.
So I need to identify this.
Let's go back.
Now.
The first thing
that I'm going to do is that I'm going to compute x
t OK XD basically is going to be the value of p modulo 3.
Okay.
This basically is going to give me value of 0 1 or 2.
Okay.
So if you go back to our presentation,
it's either going to start from this 0th position
or it is going to be
in the first position or in the second position.
Okay, similarly in the left hand side.
Also, it could be this 0th position first position
or second position.
So basically we are dividing this that's
what you need to understand.
Okay coming back.
Let me just remove this.
Okay.
Similarly.
I'm going to compute my y
of T by have to basically is to help me understand
which value should I take from this complete row,
so this will basically start from zero and go all the way
till 14 as such.
Okay, so it will basically represent which column value
should I take So I hope you understand
from this which column value should I keep taking
for my next value.
Once I have this then
what I'll do is I am going to call
the sense dot set pixel Rumba how we had called sense -
roh message so there
and we had sent a complete message.
Now, what we're doing is that we just sent to
each of the pixels individually as such.
Okay, so we're going to use the x coordinate.
Okay, which is going to be my x t plus XD.
Okay.
Remember this?
Okay.
We'll Define X.
Still it'll add but just keep this in mind XT
plus XD is going to give me the x coordinate y t
+ yd is going to give me the y coordinate
and then basically my RGB number comes so here
I have just done a multiplication of P into R.
So whichever I want I'll just convert it accordingly.
Okay.
Now comes the choice of
when you have to display two digit number.
So instead of going back to the same function.
F defined a new function,
which is called show number
now show number is The original function
that will be caught
because the temperature
measured is most often a two digit number.
So this is where the actual code was going to start from.
Okay.
So number takes four values.
It takes a variable called
while which is the present temperature.
I have my RGB value as well as I'm going to begin the function
here firstly I'll begin
by taking the absolute value of the given input.
Okay, when I say absolute value,
I am basically trying to identify the number present
in the tens place.
He's okay.
So this is what absolute value
will be whatever the digit is present in the tens place.
I'm going to store it in my abs underscore value
so that I'm just providing it here
to the tens place then let me also identify
what is in units place and this is simple mathematical operation
where I divide the number by 10,
I'm going to get a remainder
and that's what I want.
So I'm going to use a modular douche.
So my absolute value divided
by 10 is going to give me the unit's place as well.
Now, what I'll do is I'll run.
If condition so here what I'm going to do
if the value is greater
than 9 then I'm going to call the show digit function
which I had defined earlier
and I'm going to pass the tents value.
Okay, the value present in the tens place.
So and then I'm going to also pass the XD n yd.
In this case x d + yd is going to be the offset left
and offset top why let me help you understand here.
My offset top is going to be with respect to this position.
It's going to start from here.
Okay, so it will point to the third row.
Okay my offset.
Left will point to Second column here.
So I know that this is where my tens place has to come.
So that's why I'm passing it here.
Okay.
I also am going to pass the standard RGB value
that I have got input here then for a second time.
I'm going to call the show digit function where and I
will pass the unit's place,
but with an offset of additional four spaces.
Okay.
So this coming back to our segmentation.
This is going to be four positions from our left.
So offset left is going to give me the first place but offset,
Plus four is going to be the start place
where I have defined for my units place
when I say from top it's worth going to be the same
because I'm not separating with respect to different levels.
So I'm just going to change with respect to my offset left.
Okay.
So that's the only difference present here now coming down.
We are Computing the basic temperature the humidity
and pressure and then storing it
into a message string now comes the place
where I need to do my actual implementation now
if you remember our problem statement,
We have to send the pressure temperature
and humidity as a notification on our mobile device.
Okay, so for that I'm going to use in stop push
which is an application through which I can send
notifications across my mobile.
So my python application
that is the final application will communicate to install push
and that will in turn communicate
to my mobile application
that on which I have configured it.
Okay.
So the notification will be received
on my mobile application now coming down.
Me remind you again why we are using this
in a real world scenario.
This is something that is being used by NASA.
Okay, they're they're using it to identify
if the exterior temperature is falling between a specific range
or even let's say
if there's a change in the pressure inside the cabin
or if the humidity is too high,
so all this can be measured
and you can get notified with respect to any change
that you want.
Okay now coming back to our program.
This is why I have defined a range of cold and hot
If it falls below 37 degrees,
I want a notification if it goes beyond 40,
I want a notification.
So this is my range by value.
It should either be only 38 or 39.
Okay.
So this is why I am using cold and hot variables now.
Let me help you understand
how the in stock push application is working for that.
Let me open my browser and show you how you
can configure in stock push for your Raspberry Pi application.
Okay.
So this is my insta push homepage.
Okay now first.
Seeing what you need to do is
that you need to begin by creating a user ID credential
here and it is very important.
When you logging into your application,
you use the same credentials.
So make sure you do not lose this
because it's very important only
if you use the same credential will you get
this notification so remember
that so you have an option to sign up for free as well?
Okay.
All you need to do is provide your username provide
your email address and then specify a password now,
let me just log in.
Okay, so I'm just logged in.
Okay now here you can see there's already a timeline
because I have already configured this application.
Let me help you understand
how you can do this now in my application.
I already have a temp notify application.
Okay, which will be using as part of our understanding.
Okay, firstly let me click on ADD application now
once I do this I need to specify a name.
So let me just call it sample.
Once I click on Sample
once I press add application sample gets created
now it is asking me to add event here.
So click on add event.
So let me call this temp rasp.
Okay, and then tracker would be message.
Make sure you create a tab here after you specify whatever it is
that you specify and here push message will be message.
So make sure it's present inside curly braces.
What are we are specifying here.
Once this is done just click
on add event and this event is going to get added.
Okay, so the event has been added now it's time.
I helped you understand why I had specified message
there so rumor let me go back.
To my code.
Remember we had defined something known as push message.
So this is why I have defined message here.
So this is
what is going to be sent between my final program application
and my insta push application.
So this is why we had defined this variable earlier as well.
Okay now coming down with respect to this.
Let me go back to my insta push and let me help you understand.
What all do you need before you move ahead?
Okay, just go to the basic info
Tab and here there are two things that you need.
Okay first.
First you need your application ID
and then you need the application secret.
So this is very similar to how you have a user ID and password.
So this is what you're going to use in case you're working
with an application.
Okay.
So make sure you note this application ID
and the corresponding application password.
Now if you come back to my program you can already
see I have mentioned the application ID as
well as the application secret
and here the push event is to my temp notify.
So this was the previous event that was already set.
So I will be reviewing.
Bring to that as well.
Now, what you need to do is
that you need to use Curl to post whatever data
that you have to the insta push API.
So for that you're going to use biker dot curl function.
Okay, and this in turn will help you communicate
between your application and insta Porsche.
Okay.
So all the communication is going to be
through a variable called C. Now C dot set up.
Okay here what I am going to specify is
that I'm going to specify the URL first, okay.
The URL is going to be
HTTP colon slash slash a pi dot insta pushed IM /
version 1 / post.
Okay.
There's also another alternative URL which is
HTTP colon slash slash a pi dot insta pushed IM /
post itself now in case
the other is not working you can use this or vice versa.
Okay, moving forward.
I need to set up the customer headers basically
for my authentication and content type specification.
Okay.
So for that you're going to use the command C dot set of forward
by C dot HTTP header.
Then I'm going to specify x -
install push - a pi D plus my app ID,
which I have specified earlier.
So this will help me identify to which up.
This has to be pushed.
Okay.
The next is going to be x - insta push - app secret.
So basically the app secret your your password in this case.
Okay, then comes the content type.
That is what What type of content I go to pass
so this is going to be application dot Json.
So I'm basically going to pass a Json file to my API.
Okay.
So I hope you guys are clear till here.
Okay.
Now what I'm going to do is
that I'm going to define the function
through which the message is going to be passed.
Okay.
So here what I'm going to do first is
that I'm going to use a dictionary structure
for passing the data as Json in my post function.
So Jason -
field is going to create a dictionary
and in through that I'm going to send this data.
Okay.
Now what I have to do is
that I have to specify the field values.
So first comes the event, okay,
which is basically what event I need to relate
to then comes my tracker.
I'm not specified any tracker as such then comes my message
where and I'm going to pass the push message push message
basically is the temperature the pressure
and the humidity as such now
if you want you can print this
but I am not going to print it here.
Finally.
I'm going to use Json dot dump to make a Json file
and going to assign it to post Fields variable.
Okay.
So post field is the variable
which I am going to pass to my APA.
Okay next to send your Json file
with post use the following command C dot set-top.
Okay inside which you're going to specify CDOT post field
and you're going to pass this post field,
which is the Json file
as such then if you wish to capture the response,
then you can capture it inside a buffer.
So for that I am also going to use a buffer to write operation
and this can Done with the following command C dot
set up inside which I'm going to write C dot right function
and buffer, right?
That is the right back has to be into my buffer.
Okay.
Now if you want to be updated
whether the post has been sent then you
can use the following command
that is C dot set top followed by C of C were Burrows.
So this will make sure
that I am getting posted on my terminal with respect
to the communication finally comes our operation
of checking the temperature.
This is very similar to what we have.
I've done earlier
where we are running an infinite Loop.
I'm measuring my temperature in degree Centigrade.
I'm measuring my humidity measuring my pressure
storing this value inside a message variable
and then putting it to sleep simultaneously.
I am also writing into a log file
and this is going to be stored with respect
to the present time.
Okay, so till here I hope you guys are clear
now comes the part where I have identified my temperature.
I'm just making sure it's an integer
and then I'm going to call my show number function.
But if we had defined a function
based on which it is going to be displayed
on my Raspberry Pi so this is a call to that function.
Okay, once I have done with this,
I'm going to store it into temp 1 then I will make sure
that I am writing it into my file and then
comes the interesting part
if my present temperature is either greater than or equal
to the hot message
or if it is lesser than or equal to a cold temperature
that we have defined rumor the two variables.
We had defined earlier.
This is where they are being used.
Used.
Okay, the push message will start with either
if it's hot or if it's code
following that would be the present values.
Okay, the pressure the temperature and the humidity.
Okay, then what I'm doing is
that I'm going to pass this push message to my P function
that I have just defined.
So basically it will first convert the push message
into a Json data
and then it will pass it through the communication.
So this is what is happening here.
Okay.
Same thing will happen in case it is the temperature Falls
below the cold rain.
Guess what?
Okay, then I'm going to create a Json file
and then call CDOT perform.
So this basically will initiate the curl operation
to start finally to capture the response from the server.
You can actually use the buffer rage.
Okay.
So with this we have come to a conclusion.
Let me just show you the final part of this.
Okay, so make sure
that you're closing the communication through curl.
So you see door closed.
Okay.
This will ensure
that the communication done through curl is closed as
well as you ensure
that the General purpose input output is cleaned up
so that new values are being taken.
So now let's execute the final program.
That's python final program dot P by okay.
Now as you see on the Raspberry Pi it showing me
that it's presently 35 degrees Centigrade, okay.
Okay, I've started getting notifications on my phone.
Let me just share this phone screen with you.
Okay, and you can see there's a variation with respect
to the temperature shown on the Raspberry Pi as well.
Okay, let me show you first my phone screen.
Now.
If you see here
it's actually showing multiple notifications being generated.
Okay, it's showing me that it's cold presently.
And here's the complete list of the temperature notifications.
So I'm getting a constant notification with respect
to the change in temperature.
So you can see it's continuously being Generated here as well.
So this is something that is really helpful
when you're working around with respect
to a real-time scenario.
So like I said NASA extensively uses this
for their own purposes
of measuring the temperature pressure both
in their scientific research area as well as the shuttles
and space stations that the user.
Okay.
So with this we come to a conclusion
of our demo session Raspberry Pi has become the heart.
Of The Internet of Things domain
and camera is one of the most interesting modules
that you can add onto this interesting Hardware as well.
Now, let's begin by exploring the pie camera for that matter.
Now the pie camera was one
of the best additions to the Raspberry Pi component
because this module has helped the users to use the perspective
of camera into one of the smallest
and the most
effective combination of the internet of things.
Now the main objective
of the by camera module is to help you capture images.
It's capture videos time-lapses even slow motions as well.
Now apart from this the pie camera also helps
you capture videos in full HD at 30 frames HD at 60 frames
and even Vijay videos at 90 frames as well.
Now there are two versions of my camera that are out there.
Now the latest version being the camera module version 2,
which was introduced in 2060 now,
although there's not much of a difference
between both these cameras.
There has been a slight increase
in the resolution wherein you had five megapixels
for the first module you have a
It's not and even the sensor on board has been swapped out
where in there was omnivision for the first version now,
it has been replaced by a Sony sensor as well.
Now talking about the by camera.
You need to be quite careful
while you're working out with the by camera.
So let me help you understand
how you can set up by camera with your Raspberry Pi as well.
Now, there are five different stages on setting up by camera
will be looking at each one of them one by one first.
Let me help you understand how to connect your pie camera
to the Raspberry Pi now this Is very important
because we've used three camera modules
and we managed to burn out two of them.
So please be careful
because this is a very delicate ribbon
and this is also a very delicate module
as such so let me help you understand
how you can attach the CSI cable to the Raspberry Pi
and then I'll also help you understand.
Once you've done this
how to enable the camera as part of your Raspberry Pi as well.
So this is my Raspberry Pi camera,
and this is the version 2 of the camera now
if you notice this is the camera that you can see
and this is something that's really difficult.
It gets a piece be careful.
Now one thing
that I really want you to notice here is the connector here.
So this is called The Sunny Side Up and this is called the silver
and now you need to be really careful
when you're plugging this so,
let me just bring in my Raspberry Pi here as well.
So here we have our Raspberry Pi 3,
and I'll just show you how to connect this.
There are mainly two precautions that you really need to consider
while you're working with the Raspberry Pi camera.
One thing is just please ensure
that the Raspberry Pi is Switched Off or disconnected
before you connect the Raspberry Pi.
Camera as well.
The this actually leads
to a high chance of blowing up your camera.
We've already successfully managed to do that once so,
please ensure firstly
that the Raspberry Pi is Switched Off and also
that the connection is correct.
If you notice the silver side present here,
this should be facing the HDMI port
that is present here.
Now in order to connect this let me just slightly push it off
and this slot needs to be slightly raised up.
You insert it like this.
Okay.
Please be very careful.
Once you've successfully inserted it then just push
down the hinge as well.
This will lock it.
So make sure the silver side is towards the HDMI
and the Sunnyside is to your 3.5.
Mm Jack side as well.
Now.
It's recommended that you get a case wherein you can also
support the camera
because it's slightly challenging because there are
rubber in itself is delicate.
So when you get a case,
which has the camera support, it will be really useful
to you as well.
So it is a very delicate items.
So please be careful while you're working.
Thing with the Raspberry Pi camera module moving on.
Let's look at how to go into the interface and enable
the Raspberry Pi camera module for your Raspberry Pi as well.
So now that we have successfully connected
to our Raspberry Pi let me show you
how to connect or enable the Raspberry Pi camera
on your Raspberry Pi as well.
Now, what I have done is
that I have established a remote connection
between my Raspberry Pi
and my system using VNC viewer as well.
So this makes it quite easy for me to show you
what I'm doing on my Raspberry Pi.
So once you have successfully connected and turn down
your Raspberry Pi you can go into the menu
and then go to the preference section here going.
To the Raspberry Pi configuration.
And what you need to do is go into the interface section here.
Once you're here make sure
that you have enabled the camera option as well by default.
This would be disabled.
Now.
I have any able most of these for different purposes
so you can just enable the camera option as well.
Once you're done with this then we'll go on to the next stage,
which is understanding
how you can capture images from the Raspberry Pi camera
and how to capture video from Raspberry Pi camera
as well now rasp eyes still is actually a shell command.
Which lets you Is the camera module
and capture an image now,
the command for capturing an image is rasp.
I still -
oh and name of the image that you want to store it as now -
oh is basically to open the camera
and capture the image and at the same time
if you want to rotate the image vertically or horizontally,
you just need to pass the parameter - VF or -
Hatcher now both of this will either flip it vertically
or flip it horizontally and when you pass both of them,
it will completely flip it around as
well now apart from this
if you wish to capture a video
from the Raspberry Pi you can use the command
r a spy with doing this will help you
again capture the video from the camera module.
Now, this is something that gets stored by default as
a h.264 video as well.
If you wish to convert this you can use the following command,
which is sudo apt-get install -
why g-pack now?
This video will actually help you install MP4 box
onto your machine
and help you convert your h.264 videos
into MP4 format as well.
Now once you have it in MP4 format then Quite
easy to play this video and check it out as well.
Now again,
if you want to run it for a fixed duration,
then what you can do is that you can also pass
the duration as part of the parameters
while you're specifying a spy with as well.
Now one thing that you need to keep in mind
while you're doing this is
that the timeframe
that you're passing here is in milliseconds.
So please be careful while you're doing this now,
although I am specifying 10,000.
It's only for 10 seconds
that the video is going to be captured as well.
Now, let me help you understand how you can do all this.
By using a python function rather than a shell command
as well for that.
What you need to do is
that you need to install a package called by camera.
Once you've done this then you can easily access
or you can easily perform all the operations as part
of your Python program using the pike Ram module as well.
Now, what you need to do is
that first you need to update all the packages that you want.
Then you need to install
the python by camera package that's present.
Now if you have two versions of python on your machine
that I would recommend that you run both these commands
that is python -
by camera and by In 3 -
bye camera this in turn will ensure
that you have the 2.7
and version as well as the three-point X version
of Pi camera present on your machine as well.
You can also install Pi camera using pip install as well.
So the command is pseudo pip install Pi camera as such moving
on let's talk about how you can actually create a program
or how you can use a Python program
to record the videos for the same 10 seconds.
Now earlier.
What I had done is
that I had just mentioned that run for 10 seconds
using the Shell command r a spy with but Technically speaking.
This does not really account for all the challenges
that may come across
while you're actually capturing a camera now that is
where you use a Python program to help you ensure
that even if an exception
is come across then you can handle it
and due to that will be actually using the
with statement here firstly
what I would like to do.
I would like to import the time package of now
importing the time package also helps me access
to the pi camera module
and the by camera function will in itself connect
to the pie camera module and the data
that comes With them not when I'm using
with function here this
in turn will actually help me in handling all the exceptions
and encapsulating the proper tasks
and cleaning up the tasks as well.
So basically becomes a best practice
while you're using the by camera
or while writing any codes for that matter.
So most of us programmers would like to consider
all the scenarios and take care of it.
And in this case python has given as the with statement
to it take care of it as well.
Now if I use the command camera dot start with you,
what is going to happen is
that it's going to give me a Review of what my camera is C
and I call a sleeve for 10 seconds
this in turn will ensure that for the next 10 seconds.
The camera preview is on once that's done.
I'm going to stop the preview.
So what I have here is a preview
of what my camera is seeing for 10 seconds as well.
But this is not where it restricts.
There are various other commands that my pie camera brings in
as well talking about these commands
when I use the command camera dot capture.
It captures an image and Camera dot start recording
and stop recording will in turn help me.
Did a video as well if I go back to my previous program
here here I can also write camera to capture
in order to capture the image between the 10 seconds
or what I can also do is
that I can also start recording at the same time as well.
So I started recording Let it wait for 10 seconds,
and then I stop the recording as well.
So these are different commands
that you usually use together apart from this
if you want to capture a stream,
then you can use camera
to capture my streams followed by what type of stream
that you want to store it as well now
if you wish To stream the video
that is coming from your Raspberry Pi then you can use
the command camera dot captures followed by my stream
and the type of stream
that you wish to store.
Now again, this also
can be recorded wearing you use camera dot start record.
It captures the stream and you specify the quantization
that is needed to this as well.
Again, if you wish to play with some of these properties
of the camera as well,
you can play around with the brightness.
You can change the resolution
and you can even add a weight with respect to it.
Similarly as you have
a timer function in your standard cameras the same.
Be used here as well.
So let's say I want to start the recording after 5 seconds.
I can do that as well.
Now with this,
let's come to the demo of this session.
Now, this is a slightly interesting demo.
I'm quite sure most of you are looking forward
to this as well.
Then we'll try to help you understand
how you can actually set up a security camera
at your home as well.
Now talking about the overall system here now here,
what I would like to do is
that I would like to capture the image of anyone
who comes to my door.
Once that's done I would check
if this person is a recognized person or is it someone
that already has access to the house?
If not, I can always prompt my user
and And give access to it.
If not, I can deny access to it.
Now to those of you actually see in our iot projects video.
You would have noticed that in Mark Zuckerberg section.
He had actually allowed his parents.
Now, I do not want to manually have to do this
when I just automatically allow people to do this as well.
That's a smart system or that's an enhanced system
that we really would love to achieve as well.
So we would not be completely able to achieve it
but we will take our tiny steps towards it
and we can all enhance it as per our requirements as well.
So firstly what We're going to do is
that we're going to detect the face.
We're going to gather this data.
We're going to compare it
and then we're going to start recognizing data as well.
Now again, this is a learning process
that takes place.
This is something that involves a lot
of understanding and this slightly gets interesting
for most of you as well.
Now when we talk about face detection now,
the first stage
of face detection is Verena program actually decide
whether an image has a face or it does not and this stage
in itself is called a classifier as well.
In order to achieve this opencv actually provides us
with to pre trained
and ready to use face detection classifiers,
which are part of it.
We have the half classifier
and we have the LBP classifier as well.
Now one thing that you need to notice or one thing
that you need to keep in mind
while you're working with this is
that the classifiers themselves process the image
while they are in grayscale as well because it gives
the machine quite Clarity on
whether to classify it or not as such now
when I talk about each of the classifiers
whether I talk about the heart classifier
or the LBP classifier,
Each has its pros and cons.
So let's actually try to understand this.
Now the concept of phase detection is not something
that's quite alien to most of you and I'm quite sure
you've all seen it be it in our Facebook applications
Snapchat.
Some of us have also seen it on the ATM machines as well.
We're in Phase detection is incorporated.
And if you wish the most simplest methodology or
if you want to see the most common example
that's out there for face detection
or face recognition.
You can see it in most
of the phones today as well again here.
What happens is that The two algorithms
that are most widely used are the higher classifier
and the LBP classifier as well.
So again, both of these classify
are mostly processing the images in grayscale
because we don't actually need
the color related information to decide
if there is a face or not now again,
there are going to be two classifications present here
the computer program which actually decide
whether there is a positive aspect
where in the phase is present
or the negative aspect where the face is not present or so.
So this is how you decide or this is
what you Call a classifier for that matter now in opencv
as we have seen.
There are two different classifiers,
which is the higher classifier and the LBP classifier as well.
Now again, when you talk about the higher class
where it is actually a machine learning approach
that we take a
and this was created by Paul y Le and Michael Jones
wherein they are trained many positive images
and negative images without faces as well.
Now this intern actually started helping them understand
that extracting feature takes quite some time
and if you consider both What segments
when you consider higher
and LBP had actually takes up slightly more time.
But at the same time
it gives you a slightly more precise solution as well.
Now what it actually does is it actually considers
the adjacent pixels to the region
where the face is protected
and then it actually sums up the intensity
of these pixels in each of these regions as well.
Once it's done.
It actually calculates the difference between these sums
and then it uses them to differentiate
between the section
and the subsection of the Ages as well now
when it comes to the LBP section of this here,
it creates a 3 cross 3 window at the time of image as well.
Then what happens is
that it starts comparing from the Centre pixel
with all the other pixels
which are greater or equal to the value
which is set as 1 or 0 as well then again,
it goes on to read the binary value
in a counterclockwise order and converts it to decimal
which are Center values of the pixel as well.
Now what happens is
that this process is completely repeated one after the other
and you An easier or a smarter solution to this as well.
Now again,
as we have discussed both of them have an advantage
and a disadvantage
as we are talking about the half classifier there.
It has a very high accuracy when it comes to detection
of face and there's quite minimalistic false positive.
But the challenge is
that the computational time that it takes is quite slow.
And again, the training time
that it takes is quite long as well and even in dark segments,
it's not quite accurate as well because it fires High Clarity
with respect to the images or the scanning area
that it takes in the picture.
But at the same time
when you look at LBP classifier its computation
and training time is actually quite fast and shorter
when you compare to the hard classified as
well now it is quite robust
when to any elevation changes
that is when there's a change in lighting.
It's still comfortable and capable of handling this
but at the same time it's not quite accurate
or it's not on the same level at the same time
has a high chance of false positive fatigue as well now.
Move on forehead.
Let me just show you about these packages slightly better.
Now.
If you come into your opencv package now,
I would recommend
that you download the opencv library from GitHub itself.
And if you come inside the segment
and you go inside data
what you can see here are the different hard classifiers
and LBP Castle path.
Now, if you go inside the hair Cascade section here,
you can find different XML files present here,
which will help you understand
or help you recognize different aspects of a body
it can start from I you can start go to I classes
And then sue forward as but now what will mainly be using is
that we'll be using the frontal face segment here.
Okay.
And again, these are XML files.
So even if you open them and if you see them you
would not perfectly get a clear-cut idea
but this is something
that we have to use as part of our program as well.
Meanwhile while this is opening.
Let me go back.
Let me show you the lpp segment as well.
Okay.
Now what you see here are different values
that are associated
in these are different values to help you identify
or help you understand
how the face recognition should take place as such.
Okay.
No again.
It's a very huge document.
So what they're trying to do is
that it again going to try to restructure it again.
I think you've got a simple idea at this point.
So let me just close this and let me go back now
if you check the LBP
Cascade segment as well here you have different segments.
So again, what we'll be using is
that we'll be using the frontal face segment as such.
So what I've done is
that I've just created a separate folder
where I've kept my program
and these two opencv files for different algorithms
as well now coming back
we've done with our detecting the face.
Let's talk about Using our faces
as well now till here what I have done is
that I have detected my face,
but I need to start recognizing this face as well.
So face recognition
mainly can be classified into three different steps first.
I've actually dealing
with the data Gathering stage now in the stage,
what you're mostly going to do is
that you're going to gather any relevant information
with respect to that face.
So usually this is done when you have multiple photos
of a person and then you start identifying
or you start training your machine on this face
and then it starts recognizing the Face every time
that it comes up with an image of the same as well.
Now there are three different functions
that opencv provides us
to do this and let me give you a brief idea on each of them
so that you know,
which to use well now first comes the Eugene faces now
Eugene freezes actually takes all the images
at once looks for all
the important and useful components in these phases
which are known as principal features
and then passes it on now one key factor
that makes or one key challenge with respect to the Eugene.
Phase function is
that it requires a good amount of Illumination.
But at the same time
when you have your fish your face here individually images
or individually features are extracted rather than
extracting one by one
and this in turn actually
does not have a challenge with the illumination segment
as such third is your lbph,
which is your low binary pattern histogram
and this is very similar to your LBP as but but
what happens here is
that you create a binary Matrix and convert into decimal factor
and then you actually Not a histogram with respect to it.
So what happens here is actually a comparison
with respect to the histograms that are present.
So each phase will have a correspondingly
different histogram
and I just compare these histograms to in order
to identify which phase it is belonging to now.
Let me come back to my place
where you buy and before I go forward
as we said this segment actually deals with two parts
first is actually capturing the detection
and the second is with the recognition part.
So let me first show you how you can detect it now.
We've already discussed above.
About detection or how to capture any intrusion detection
in our Raspberry Pi 3 tutorial
where we actually capture someone opening the door
and coming in as well.
Once I've captured this image.
Then I need to start recognizing now first.
Let me show you this program here.
If you want more details
on how it works out then you can definitely check
out our Raspberry Pi 3 tutorial as well.
Now if I use the following code, let me just expand this a bit
so that I think you may not be visible to you.
We just increase the font size.
Yeah, I know.
I think you can all see my code as well.
Now again, we're using different packages here
you using user Library CV to the time and numpy as well.
Now, I'm first going
to set the first frame then I'm going to decide
whether there's any detection as well.
Okay.
Now for this
as we have mentioned in our Raspberry Pi 3 tutorial,
you can use the IP camera application
which is an Android application
and you get an IP address as well or a URL
which have specified here.
Now, let me just run this and show you
what I'm seeing at this moment.
Now what it's going to do also is
that it's going to capture the last image
that the camera sees as well.
So any change with respect to it?
It's going to overwrite
until I quit or I exit from the program
as such so let me just run this.
I just zoom in a bit.
So let me go back
to my desktop and I'll run the camera application.
Now what it's doing is
that it's connecting to my phone camera
and it's going to create or show you what I'm seeing or what.
My phone is seeing at the moment.
So what you see in is my Raspberry Pi set up here
as you can see,
it's set up on a curve because it's slightly tends
to get heated up,
but you can see it's a separate device as well.
So this is my first frame and what I'm going to do is
that I'm going to put my hand in front of it and I'm going
to make this as my last frame as well while this remains.
I'm going to press Q to quit from this.
And it also has created
an image called test dot jpg here as well.
Let me just open this and
if you see it's my hand that has been captured.
So what happens here is
that it captures the last in frame as well.
So if you want to get a better understanding
of which image it captures will rerun this but before that,
let me just remove this file.
Okay.
So let me just send it to the waste bin.
And then what we'll do is we just re execute this program.
I'll start by keeping my hand in the front of the camera itself.
So this is my first frame, okay.
And any change to this let me I'll just move
it off completely.
So this is going to be my last frame
and I press Q to quit
from this and let's just recheck the image as well.
So you can see it has captured the last frame
as such with respect to this.
So it's a very interesting thing for capturing this
and now let's actually come now into our program
which talks about face recognition and
face detection as well.
So this is the program
that will help me detect the face train my application
and also understand which face belongs to whom as well.
So I have imported my opencv to because that is
where my face detection and face recognition.
Algorithms are present.
I'm going to use the OS module here as
well because I need access
to the drives our the path
where these images are going to be stored as well
and numpy is one of the very Key Programs of is one
of the very key modules,
which is required for any face detection recognition as well.
Because the information
that you're going to be using would be stored as a numpy array
for that matter.
Now again,
what I'm going to do is that I'm going to actually
identify the face of three people.
Okay.
I'm going to identify whether the image
that I'm passing is of Ronaldo is of messy or is off Robin.
Okay.
Our if it does not fall into either of these three,
it's going to give it as null as well.
Now our first program or the first function
that we're going to be working is with respect
to detecting the fixed.
Now, as I said,
all the images are going to be converted into grayscale.
So that's what I'm going to be using.
C V2 dot CVT color to grayscale conversion.
Once this is done.
I'm going to create a face Cascade for that matter.
Now C V2 dot Cascade classifier actually specifies which kind
of classifier to be used.
Now for this program.
We're going to be using the LBP classifier.
Okay.
And as you see it's for the frontal face as well.
We've seen there are different aspects but we're going
to be using LBP frontal face classifier as such now
this phase is going to store the list of all the faces
that I'm going to be detecting or I'm going to be identifying.
Okay.
So for this what I'm going to do is is
that I'm going to specify face underscore casket
and then I'm going to ask it to detect
in multi scale parameter.
So I'm going to say it has to identify in grayscale.
The scaling factor
of the image is set as 1 Point 2 and the minimum number
of pixels that it should consider is 5 okay.
Now if no faces are detected from the images
that I pass then it has to pass the parameter as none.
But if it is detected then I want all the four coordinates
with respect to the X and y-axis,
which I'm going to be storing in XY W and Okay,
X Y width and height with respect to smarter
or I'm going to specify which part of the image
does the face actually contain.
So I'm going to specify with respect to the four coordinates
that I'm going to be passing.
Once this is done.
Then I'm going to define the next function
which is to prepare my data now
as we've seen first comes detection then comes preparing
when it comes to it.
I need to specify the path
from where it has to pick up the images for that matter.
So let me come back.
Let me show you where the images are.
Isn't now when you see inside training data folder.
Okay, I have actually three folders present here,
which is my S 1 S 2 and S 3 if I go inside as one
you can see there is a Ronaldo's photos.
If I go back into S 2 you can see
that is Messi's photos.
And if you see inside S3,
this Robbins photos present here now.
Let me come back to the core now here.
What I'm doing is that I'm passing the path then
I'm calling the phase function as well as the labels.
Now again, the phase is going to hold the sub phases
of all the subjects and labels actually.
Help me identify which it belongs to as well.
Now what it's going to do is
that it's going to run a for Loop wherein it goes through.
Each of the images is present in the path.
Now again,
what we have decided or what we have started
with the naming convention is with s 1 2 3.
So what it's going to do is that it's going to check
if the folder name starts with s
because usually there are system photos separately.
So it is going to check if the folder name starts with s
if it does not then it's going to go into the next Loop.
But if it does it will check that folder as well.
Then what is going to do is
that it's going to create the label for it
or it's gonna identify which folder I'm presented.
If you actually go back inside back to my training data,
you can see I have specified 1 2 3 after this so it's just
to help me identify which label it belongs to as well.
You can use any different naming conventions
that you want as such then what I'm going to do is
that I'm going to specify where the folder
or where the images is present,
which is actually one level inside this
so I'm going to go into the next directory.
And then what I'm going to do is
that I'm going to run a for Loop for all the images.
Are present in this
if an image or a file for that matter starts with DOT
because system files usually start with DOT.
So those are going to be ignored
but others are going to be recognized using my opencv
where and I'm going to read the image first.
I'm going to show you a message that says training on the data.
I'm going to resize the image to a size
that is comfortable for me,
which is 400 pixel to find a picture.
Okay, and then I'm going to wait four hundred milliseconds
for that matter.
Now what you can see here is
that I'm calling to detect phase function
which is present earlier.
Yeah, so this again is a call back to the previous function
where it goes up and it actually calls the previous function
that we have created to detect the face as well.
Once that is done.
It's also going to revert
with me the position as well as the face
or the ID which it belongs to as well.
Now if a phase is present then I'm going to append it
to my face list
and I'm also going to add a label to this face as well.
Once this is done.
I'm going to destroy these windows
because again opencv creates a lot of windows,
so I'm going to ensure
that it's destroyed by doing it twice as well.
This is done.
I'm going to return the This
which I have and the labels correspondingly as well.
Now what you need to understand is
that here the faces here are of different players as well.
So each of the phases present
in each of these images is going to be segregated.
It's going to store this into a location
and it's going to start raining on them as well.
Then what I'm going to do is
that I'm going to prepare the training data.
Now this again is something
that is becoming a recursive function for that matter.
It's going to be calling itself again
where it goes back it goes to one level up
and it's going to keep calling itself
until I have trained on all the folders
which are present
and all the images present inside that folder as well.
So once this is done
what happens is that it prints me
that data has been prepared.
And then it also tells me the total number of faces
and the total number of labels also present here
as such finally what I'm going to do is
that I'm going to create our LVH be faced a cognizer.
So for this I'm going to call the lbs be faced
a cognizer function.
So if you remember we had discussed
about three different phase secondary functions,
we're going to be using the LBH be faced
a cognizer function figure.
The function has been defined
then I'm going to pass these faces
so that my machine can learn
and train itself to identify these phases in future.
Now, what I'm going to do is
that I'm going to just create a rectangle across their faces.
So that becomes a differentiating factor
for them or helps them separate where the faces with respect
where and once this is defined or
once I have created a rectangle across their face again,
these XY heads
that are the different coordinates of the face,
and I'm just going to draw a rectangle across them.
And again, I'm going to specify which color
so this is your RGB value.
So it's going to be a green color box
that's going to be defined and I'm also going
to print the name of the person present here.
So what it's going to do is
that it's going to identify which person it belongs
to and it's going to print it in hen seaplane.
Okay.
And what is going to do is
that it's going to be printed again in green color as well.
Now what I comes or now the most important function comes
which is prediction.
Now in production,
what I'm going to do is that I'm going to pass
an image the image copy is going to be created.
As we do not want to manipulate or work on a single copy
of an image because in case of any challenge or any issues,
we do not want to overwrite
or lose the image for that matter.
Once this is done then what I'm going to do is
that I'm going to call the detect phase function
which in turn will help me identify who the person is then
what is going to do is
that it's going to predict the image using our face a cognizer
and then it's also going to print the confidence
or how sure it is
that this face matches.
Now what I want you to understand here is
that if my confidence is low 30 that is
if I'm not sure
that this person's image is correct
or the phrase is detected then what happens is
if it is greater than 30 is going to draw a rectangle
and then it's going to write who the person belongs to
and it's going to return this image for me
on the screen as well.
Now why have two images
that I'm going to be passing here?
If I just go
onto my desktop images show you first is abc dot jpg,
and the second is wrong dot one dot jpeg as well.
So let me just show you the images.
So you can see here.
I have drawn dot one dot jpeg,
which is an image of Ronaldo.
And I'm also going to pass a b c dot jpeg
which is an image of messy.
Okay.
So these are the two images I want to identify
and I will be passing this to my program.
Let me just run this program
so that you get a full fledged understanding of
what we have written here.
And you understand how this works out as well.
Okay.
So once I've passed this image,
then I'm going to call the predict function
so that it predicts whose face
that this belongs to Two or who's this face is as well.
Once that's done.
I'm going to destroy this now.
I hope you've got a simple understanding of this.
If you are interested in the code,
please give us your mail ID.
We will be happy to share the code with you as well.
Now.
Let me just execute this
so that you get a clearer understanding of this.
Now, let me just zoom in so that you know,
so you can see it
has started picking up the faces for recognition.
So you can see it starting to prepare the data.
It's checking out all the images.
It's reading the faces is trying to detect the fuses for itself.
Now you can see
total faces are 22 wherein it has identified 22 images
and it also has started to predict it
and it's given me the confidence of the face detection
that it has taken place as well.
Now the prediction has completed for Ronaldo.
It has successfully predicted the face
and it also has put in the text and if you see for Messi also,
it has done the same but the text is very small here
because the image has been resized for that matter.
Okay, so it has successfully predicted this it has detected.
It also has recognized it as well now.
Has also told me who this face belongs
to as such for that matter.
So this is
how our face recognition system works completed.
Now, this is just one
of a very my new door one of the basics examples of this
and if you want to create a complete full fledged system,
let me give you an idea on that same as well.
Now if you wish to have a very great system out of something
that's really effective out there something
that's not too hard to build then this is something
that you should really consider place your camera,
which is a wireless camera on top of your door.
Okay, use the Raspberry Pi to detect the motion.
What will happen here?
Is that every time there is a detection Raspberry Pi
would detect this happens
and then it would pass on this information
onto the cloud the reason for doing this is
that it becomes quite fast
and it's way more effective
in that matter as well here the phase
would be detected recognition would happen all the information
regarding these phases would be present on the cloud itself.
And once it has identified
that this is a valid user then correspondingly.
It will allow the door to be opened as well.
But what you need to understand here is At this
in turn can also be configured
for you to have an application
or a notification sent on to your mobile application
from the cloud as well.
Because most of the mobile application Epi is
for pushing information is very easily accessible
and configurable on the cloud
as such the Raspberry Pi has become the most popular platform
for exploring the no House of Internet of Things.
It is redefining the nuances for a new era
of minimal operating software.
And if it's about operating systems,
you can be sure to find
Microsoft Windows just around the corner.
So we'll be talking about the Windows 10 today,
but not the one on your desktops or laptops.
In fact, not even the one on your smartphones
this Windows operating software runs devices even
smaller than any of those now,
if you've been tinkering with the Raspberry Pi you
must try installing the Windows 10 iot core on it
and know the operating systems
that come as a part of the Windows 10 iot windows
that is fairly popular
amongst operating systems is now available for Internet
of Things is actually a revamp
of the old operating system Windows embedded
but with iot taking the World by storm the Windows 10 iot is now
it's official operating software for all
everything's and systems.
Now the windows
for Internet of Things comes in three packages the Windows
10 iot Enterprise comes for the embedded systems
that draw more power and require more resources
like the bulkier systems in high grade
or industrial grade Machinery
the Windows iot mobile Enterprise comes
for more minimal systems like the ones in our televisions
or handled it.
Vices and the Windows
10 iot core comes for the smallest embedded systems
in iot like even minut combinations of microcontrollers
and microprocessors.
So you can scale up
or scale down these options according to iot devices
and the requirement now
the Windows 10 iot code is the smallest
of the three operating systems launched by windows
for The Internet of Things it in fact
is the smallest operating system
that Windows has ever had and you will be happy to know
that this one comes for free now this windows
for small devices is so minimal.
It's interface is
unlike any of the other Windows operating systems
and it is almost like that of a bear Windows kernel
but it is just about enough to power the things in iot run all
of its applications
and ensure the seamless flow of data to and from the cloud.
It also supports many other words used in iot development,
but we be focusing on its installation
on the Raspberry Pi today.
Now, it's minimal interface is for better convenience as
whenever an application gets deployed to run
on the Windows iot core the windows interface will disappear
and all you'll be left with is the application.
Running on it.
Now this is because the Windows 10 iot has been designed to run
both headed and headless
which essentially means with and without a display.
So your device
becoming the application will reduce confusion
while running headless.
This also means
that we can have only one active application running at a time
while there could be other background applications
running all that said I think it's finally time.
We went over the installation of the Windows 10 iot core
on the Raspberry Pi
so I have a Raspberry Pi 3 here with me already
and a class 10 SD card of 16GB
upon which Be flashing our operating system.
And please make sure
that you SD card is at least class 10 and has a capacity
of at least 8 GB
without these you might face compatibility issues
with the windows iot core operating system.
Now the Samsung Evo
or the SanDisk Ultra would be your best choices.
So the first thing to do is to format the SD card.
I recommend using the SD card format or software
that is available for free online.
You could also use the default format on your system,
but I'd avoid using it as it does not optimize the partition.
Required by the operating system and the storage on SD card
and it compromises on the entire capacity
that is available.
So go on with the SD card for matter,
so it is really easy to get we start by opening a browser
and typing SD card for matter into it and hit enter
so you can go to the first link.
And scroll down once the website opens
so you can see SD memory card format a download
for Windows and Mac
since we need the Windows operating system to flash
the windows iot core or to the Raspberry Pi will be going
for the windows version of it.
So click on it and you download should be initiated.
So once you download is done find the setup file
and install SD card format Iran to system.
I've already done that so I will straight away go
and open the app.
And here it is.
So you'll have to know
which Drive is your SD card mounted on for that.
We can go to my PC or this PC.
So in my case, it is the F drive.
So I'll make sure
that the card is wanting to the F drive
and I'll click on format.
It will give you an alert saying
that it is all data on this card hit.
Yes.
There we go.
Now we are ready for the windows iot core download.
So we go back to the browser and we type windows iot.
So the first website is the official website
for Windows 10 iot core.
Once the web page loads go to the download section
and you can see there's a download link saying
download the Windows 10 iot core dashboard.
So this is what will be downloading.
So we'll be installing the operating system
through this dashboard itself.
Let's go to the download folder find the setup file
double-click on it.
And click install.
So it's a 58 Point 1 MB file.
So it is getting downloaded on the Windows
10 iot core dashboard from where we will be installing
and flashing it onto our SD card.
So once the setup is done,
the windows iot dashboard should pop up automatically
if it is not you can search it on your system as iot dashboard
and open it.
So now with the dashboard in our system,
we can start installing the windows iot core
onto our SD card.
We can do that by setting up a new device.
Let's click on setup new device.
So once you do that, you can see a few columns
that you will have to fill so the first is device type.
So I'm using a Raspberry Pi 3.
So I'll choose Raspberry Pi 3.
What?
Was Bill do I want well,
I want the Windows 10 iot core.
So I'll click on that which Drive is my SD card
on its on the F drive.
That's correct.
And you can rename your device from here.
Let's say I'm gonna name it men
win RPI and you'll have to set a password for it.
So I'm going to go with our PI 3 win
10 have to repeat the same.
Our PI 3 2010.
So now that the passwords are in place accept
the license agreement and click on download and installed.
So I'll give you a pop-up about erasing the SD card
click on continue.
So you downloading has now started so you might get
another pop-up which is the iot utilities trying
to install on your system,
press yes, then you should get a command line window
where the OS will be flashed upon your SD card.
So with that done,
we finally have the operating system on our SD card.
There will be another prompt asking you to format disk.
Please hit cancel.
This is very important to not hit for my disc it on cancel.
That's it.
It says you guys d card is ready.
So go back to your mypcc.
Make sure you eject your SD card.
Once it is safe to remove take it out and put it back
in your Raspberry Pi.
So the next thing to do is to connect all your peripherals
to your Raspberry Pi
and power it on now most operating systems
will require you to configure some basic settings.
When you put them up for the first time
since the Windows 10 iot core is the smallest version of Windows.
It almost has negligible settings to be configured.
It will only ask you to set up your default language and
if you want to connect to any Wi-Fi network,
and that is it so now upon starting the pie.
You should see the windows logo.
You can see the welcome screen for the Windows iot 10 code now
since it has been designed for minimal and low-cost
devices the configuration
upon the first boot has also been kept minimal
so that does it for finally booted into the settings
of the windows iot core.
So let us set the default languages English and hit next
after which it will ask you to connect to a Wi-Fi network.
Let us say we will skip this step and You
can also do it later from the home screen
or the desktop.
So that's it.
You finally successfully booted the windows iot core
and it is ready for use.
So here we have Katana you can record your voice
for better speech recognition.
I'll skip this for now and let's say got it.
So this is your desktop
for the Windows 10 iot core you can see there's device info
which is mainly a home screen.
It tells you the IP address and everything.
There's a command line window.
There's also a browser and there are tutorials
for you to make the most of the windows iot code.
This is settings button.
There's a power button as well other reason
behind your desktop being so minimal is
that the windows iot has been designed
to run both headed and headless applications
so to develop applications,
You can do it on your laptop and then push it remotely it
onto your Pi upon which the pie will become their application.
You can use the visual studio for it.
So now you can go back to your iot dashboard.
Now the iot dashboard also offers some functionalities
for you to remotely control your pie
or your windows iot core.
So you can go to the my devices on your dashboard and you
can see there's a Raspberry Pi 3 with the name mean
when RPI the name that we gave it in the start now.
From here, you can open it
in device portal to our extended features.
You can launch a command line window.
You can launch iot Remote client.
You can shut it down.
You can restart it plus you can add another pie to it.
You can also connect it to the Jour Cloud
for intelligent analysis and advanced cloud computing.
And also there are three packages which indirectly
be deployed to your Raspberry Pi
so you basically won't need Visual Studio.
You will not need much of coding you can directly
download these packages and run them to die.
The power of Windows iot core.
So that's it.
You can see the settings offers
some basic preferences this network in Wi-Fi.
This blue to this.
Katana is a command line window.
There's an internet browser
and there are other tutorials to make the most
of Windows 10 iot core.
So that will be all I hope you followed
how to install the Windows 10 iot code now looking
at the rise in the Popularity of Internet of Things.
I've come up with this tutorial to make sure you people pick
or purchase the right development boards
for your iot projects.
Now the adreno
and the Raspberry Pi are you go to votes for all things iot,
but how are they any different
and which one should you be using?
Well, don't worry by the end of this session.
You could be educating your friends on the same.
But before we begin though,
let me quickly tell you what exactly I'll be covering
in today's session.
So for those of you
that are still confused on what these devices are.
I'll briefly talk about these integrated circuit board.
Odes then we compare their hardware
and software specifications understand their modes
of operation check out there available boards
and finally conclude with the purpose
that each of these boats have been designed
for so without putting it off any further.
Let's get started.
The computers were a breakthrough in the world
of machines making the operation of any device
or machine smarter than ever
and the motherboard in a computer is
what drove these operations
either performing logical instructions
that have been fed to them
or calculating outcomes based on the information.
They receive these motherboards LED all parts
of the computer communicate and act by utilizing the CPU
and the memory in fact any peripheral connecting
to the computer also needs to connect
to the motherboards programmable input/output.
Actors for their operation
now these motherboards are integrated circuits
and with the world trying to minimize the size of devices
and make them more portable.
We now have a variety of similarly capable circuit boards
available to help us build smart equipment of our choice
and according to our needs.
Only smartest development boards today zero down
to the Raspberry Pi
and the Arduino both these integrated
circuit boards might have a few things in common,
but differ majorly in the mode of operation
and capabilities and
if you've been experimenting with the embedded systems,
or if the onset
of iot has got you playing with these integrated circuit boards,
then you must understand the key differences
between them and know the one
that will be more suitable for your iot projects.
So let me Begin by telling you
that the Arduino is based on a microcontroller
and is mostly designed.
To control the electrical components connected
to the circuit board in any system.
Whereas the Raspberry Pi is based on a microprocessor
that computes data and produces valuable outputs and
if required it controls the electrical components
and it's system based on the outcome of its computation.
So basically the Raspberry Pi is a single-board computer
with the Arduino is just a single-board microcontroller.
So let us start with the comparison
by learning more about the hardware
and software specifications
of the latest models of both these devices.
Now the Arduino boards
have a very simple hardware and software structure the Arduino,
you know runs
on an 8-bit AVR atmega328 microcontroller on a chip,
which has a ram of two kilobytes upon an eeprom flash memory
of 32 kilobytes
and clocks are processing speed of 16 megahertz on the contrary
the Raspberry Pi boards have a complex.
Lecture of hardware
and software the Raspberry Pi 3 B
plus comes with a powerful broadcom BCM to a37 b0
microprocessor on a chip as its Central Processing Unit,
which runs on a 64-bit quad-core processor called
the a arm cortex a53
featuring a static Ram of 1GB running directly
upon the SD card storage at a clock speed
of one point four gigahertz.
Now this microchip also acts as the graphics processing unit
with the broadcom video code for integrated
graphics card building.
It for low-power multimedia processing now such
a multi utility microchip is being called a Soc
or a system-on-a-chip these days
and besides the processor and the graphics card.
They also house other microcontrollers to control
the electrical components.
So there's plenty of power packed into it.
Even for many add-ons or USB connections.
Now most of the other Hardware,
uh, pretty much the same for both these devices
like timers voltage Regulators serial,
peripheral interface bus has into integrated circuit serial,
you are and I/O pins, okay.
Connectors but again,
the I/O connectors in Arduino are more capable
than the I/O pins in the Raspberry Pi
which must depend on transistors to drive external hardware.
I'll be Raspberry Pi is still better logical processing
which means it can compute even complex mathematical operations
enabling it to process
and support audios videos images and other gu eyes.
So it's more like a computer in contrast to the Arduino
which can only control path The device and would need
additional circuit software
and communication buses to run audio video
or any visual graphic at all.
Plus the pie also offers HDMI support.
So all that you need is just a screen
or any display device removing the need for connecting it
to a computer all the time like the Arduino now this is
because the Arduino comes
without any operating software the only real software it has is
for compiling code to Machine level
and allowing serial connection for data transfer.
This is also why it requires less power
and could even get faster
as it only needs to compile code and use
the chip directly to control any component.
Whereas the Raspberry Pi
must enable all required layers of software
and utilize different programs
to make any task happen then again,
it is also why the Pike in multitask doing a variety
of tasks at a time
while the Arduino can perform any single task multiple times.
So the Arduino offers an ID with the compiler
and is the only one that this device supports
which means only the codes written here will work
and get compiled on to its physical board
this He uses basic C
and C++ paradigms and breaks down all its functions
into accessible packages making it quite easy to learn
and operate however with the Raspberry Pi
which supports multiple programming languages
and software environments.
You need to be proficient
at coding to get the most out of it.
Also the pie now comes with Bluetooth
and Wi-Fi support for remote connections.
In fact, its latest model features faster Ethernet
or dual band of 2.4 gigahertz and 5 gigahertz wireless,
LAN and an embedded web server
while the we know features Bluetooth on just one
of its boards all the Arduino BT
and will require additional circuit
for communication over wireless networks or ethernet.
Now these additional circuits for the Arduino
or called Shields
and are attached to it for extended functionalities
similarly in the case
of the Raspberry Pi these extended circuits
are called hats both are plenty
of available housing diverse functionalities
sensors Motors actuators
and software on them
for interacting with the physical world
and driving external hardware
according to It's host devices capabilities.
Okay.
Now I can see somebody has a question.
So Debbie wants to know
what is the most popular programming language
that is used for coding with the Raspberry Pi.
So since the Raspberry Pi is
essentially a computer you can actually run
code on it in almost all programming languages.
But the most popular one in my opinion would be python
as even the pies official operating system comes
with the default python ideally,
which is the IDE for python.
So I hope that clears you doubt if there is Any more questions,
please keep it coming.
Okay.
Meanwhile, I mean move on.
So let me now tell you
how you can identify the purpose behind using either an Arduino
or Raspberry Pi.
So the Arduino being more of a microcontroller motherboard
is best suited for repetitive tasks like reading
and Reporting temperature opening and closing
of doors controlling lights
or building motion detecting alarms.
It also works great
for creating quick and lightweight prototypes,
but the Raspberry Pi is more like a general purpose computer
and is good at complex.
Relations or performing multiple tasks or processing
a large amount of data.
So if you need a full-fledged computer
on a very small scale the pie is what you should be looking at.
So let us now move on to the different boards
that both have had over time
and look at some of its prominent ones.
So the Arduino has been around for quite a while now
and has had multiple boards over the years
with different functionalities.
Although the Arduino, you know,
the Arduino Mega they Do you know lily pad
and the Arduino Bluetooth have been the more popular ones?
Similarly?
The Raspberry Pi
despite being relatively new is on its third generation
with each of its models better than the previous ones
the Raspberry Pi 3,
B+ is its latest and most powerful development board
yet in the session.
We'll be talking about the top seven projects
which are outstanding themselves in the Internet
of Things domain now,
let's talk about Each of these projects.
Let's start with number 7, which is a biometric system.
Now a biometric system is something
that we always encounter on our daily basis as
well because we always either use a fingerprint sensor
or we use an iris scanning.
So it depends from organization to organization.
But how does this actually work the system there?
Let's take in case
of this example has a fingerprint scanner.
Now the first time that I am presenting my fingerprint,
it scans the fingerprint
and considers this as part of an enrollment.
Us from this fingerprint template what it does is
that it extract certain key features,
which makes it different from others and stores it
into a database from then forward every time
that I place my finger on top of this fingerprint scanner.
It creates a template
and Compares this with all the templates
that's present in the database
if it matches then correspondingly,
let's it gives me an attendance or lets me access at or
if it does not then it raises an alert
with respect to that same now.
This is just a foundation now
as I said this biometric system Fingerprint,
it can be is scanning
or it can be a combination
of both as well as voice recognition system is one
of the key products in the biometric domain next.
We have Smart irrigation system.
Now I am someone
who personally likes gardening a lot.
Now, this is something
that would really make my life easier
because usually what I do is
that on Sundays I spent
about two hours just watering and looking
after my plants through a smart irrigation system.
What it does is
that it checks the moisture present in the environment or in
the water lanes that I have created now to help.
You understand how it works.
Usually there are two main Internet of Things devices
that are used here,
which is the Arduino board and the Raspberry Pi
your Raspberry Pi becomes the main processing unit.
And I place an Arduino board for each of my water channels.
These addressing modes themselves are connected
to multiple sensors
which are part of this water channel.
So what these sensors do is
that we check the moisture present in these lanes
are such so let's say
a specifically in
does not meet the minimum required moisture then
what it would do is
that it will Send a signal to the Raspberry Pi again.
All these devices
are connected on the same wireless router Network
and the Raspberry Pi would identify the lack
of moisture and pass the signal to the relay.
The relay in turn would initiate the water pump
and water would be pumped now in order to ensure
that water is not wasted we would create gate controls
and only the gate
where the moisture is less.
Would the gate be open once my sensor detects
that the moisture level has gone
beyond my required level it would
Against another signal
to the Raspberry Pi asking it to stop the pump as well.
So this in turn helps you to save a lot of water
and also makes your life quite easier as well.
So after this your only task
in your garden would be is either setting up new plans
or creating new water channels as such the next project
in our list is a security camera and door unlock system.
Now, this is something that's quite interesting
and I have personally tried this out
and it's really something
that you should try out as well here.
What happens is that you place a camera
on top 40 Which in turn clicks the photo of a person
who comes into the frame now,
this photo is again sent to an analytical system
which in turn Compares this with all the photos
that it possesses in order to identify
whether to let the user open the door or not now
and evolution to this is
that if it does not find the photo of that person it
can notify the concerned person
that so-and-so person is trying to access this dough.
Would you like to authorize this person
and address information to the database or would you?
I'd like to deny the access to this person as well.
Usually this is used in areas
where you have high sensitive information stored
in order to maintain a strict control to the access
to this information as well another usage
of the security camera
and door unlock system can be even at our homes
when we can identify
who's come at our home when you're not there
and either decide to give them access to our homes or not.
Now the next thing is something
that we all really decide to have
which is a smart home system a smart home system
can be something that really makes.
Life quite easy starting from energy management
where the light control system is the AC the appliances
that we use the thermostat all this is managed
in short trying to cut down the power consumption
that's taking place
by door management system is also part of this.
My security system is also part
of this my water management system again becomes a part
of this as well.
Again, these are key things
that really stands out in the smart home system.
But again, what I would personally recommend is
that a Smart Homes limitations is where
our imagination stops anything
that you wish to automate or make wish to make
your life easier can be part of a smart phone system as well.
Now a smart home usually is going to be a base
for our next project
which is a smart City a smart city is an evolution
of a smart home here.
It's not just the sensors of a single home
that is connected here.
It's correlation or a network
or a connection between various organizations
various domains as well as various segments
of that city as a whole as such Judo the life
of every single dependent person
in that City becomes easier as a whole as well
and in turn will really help develop that City
to Greater extent such now the key factor here
for a smart city is a government support as well.
And if governments are really willing to take this step
then I hope we would see a smart City completely built
on internet of things maybe the next 5 to 10 years as well.
Now the next project is something
that really stands out on a personal level this this
is Zelda Ocarina controlled home automation system.
This is personally something
that I feel is the closest to a smart home system
where most of the elements
of the home is completely controlled by an Ocarina
what Alan pan has done here is
that he's created a node-based recognition system,
which completely automates his hope but rather
than telling more about this.
Let's just look at a quick video
which will give you a glimpse into how he has done this
and what has he done.
7:27 p.m.
Next on our list at number one.
If you've not already guessed
is Jarvis Jarvis is the artificial intelligence system
that Facebook's creator Mark second Berg has built
for his home automated system.
I'm quite sure you've already seen the video
of Mars are kannberg interacting
with Jarvis which has Morgan Freeman's voice.
And if you've not,
here's a quick glimpse into the same.
Good morning, Jarvis.
Good morning Mark.
It's Saturday.
So you only have five meetings room temperature is set
to a cool 68 degrees earlier this year.
I started building a simple AI to help run our home.
I talked to Jarvis using this app.
I built it uses artificial intelligence to understand me
and figure out what to do Max woke up a few minutes ago.
I'm entertaining her.
All right.
Let's go check on her.
Good morning.
Let's practice our man.
Zhao Xiang Hao Jarvis your Mandarin is so soothing
she she Jarvis also helps me get ready in the morning
fresh shirt Byron home.
Hell yeah,
Jarvis knows when to make me breakfast.
Your toast is ready.
All right, it's time for my call with shrap.
Can you get him on the video conference line
setting up the BC room.
Now remember to check on the AI guidance system for tequila.
One of the best things about Jarvis is it could recognize
people at the gate let them in automatically
and then just tell me about it
Mark your parents are coming in.
Thanks Geoffrey.
It's Jarvis and Jarvis can play all of our favorite music.
Hey play some good.
Go back songs.
I'm sorry Mark.
I'm afraid I can't do that.
There are no good Nickelback song Good
that was actually a test.
Okay.
How about just play some songs
that are whole family likes what?
I'm a dad.
Now.
This is what I listen to.
No, what you need to understand here is
that internet of things is not something
that's just dependent on a sensor or a few sensors,
which is connected to a Raspberry Pi on adreno put
when I look at the entire architecture of iot.
This is a complete ecosystem
where in my sensors gather information,
which is again stored on a platform
and then processed on this now
if there have been any issues or failures
that have been reported by these sensors,
then I need to create actionable items.
In order to ensure
that this is never really repeated again,
but the information wants to spin process
the process data is then passed into machine learning
and artificial intelligence in order to understand analyze
and identify various patterns
that's out there and help pass this information
on back to the sensor this in turn helps.
You have a better experience
and also improve the system as a whole a system
where I come home every day
at 8:30 start off my Casey wait 15 minutes
for it to cool down is a tedious process.
But it's a today my Internet of Things platform identifies
this pattern already switches on the AC at 8:15.
Then I just need to come home and I can relax right away.
This is just one of the key examples
of one of the key ideas
that's out there today
and the limitation here again is just your imagination.
It is so good to see
that so many of you are keeping up with the trends.
Of technology today.
So let us dig deeper into these devices
that Mark the beginning of every iot ecosystem
and look at the best ones in use today in different domains.
But before we start don't let me quickly run you
along the outlines of today's session
so that you have a clear idea on the topics
that I'd be covering.
So first I'll be telling you how you can identify any object
to be an iot device.
Then I'll tell you what they can do for you and cite some
of the major use cases on it and then I'll tell you
how iot devices are shipping the entire world
and code some leading examples
and then I'll tell you how secure is.
The device in iot and finally,
I'll be telling you all the important things
that you need to know to build an iot device.
So without putting it off any further,
let us begin our session today.
Now these devices
that play the role of things in iot are
what make the smartest systems possible today be
just a smart Appliance or an entire Smart City.
It could be complete Computing devices designed
keeping the portability in mind
or tiny electronic components operating
on minimal device software for data transfer
and Activity without any Computing abilities at all
and there are also objects
that are not made up of any electronic Hardware at all
and can still become the things in iot by having
these smart things attached to them in summer the other way.
So what I'm basically saying is
that even we humans can become a thing
in the internet of things.
But again,
that is a contextual top and you'll understand why
as we move on with the session today.
So for now we can just say
that any device however puny
or powerless with the ability to receive
or send data allowing it to communicate
over networks can be called a thing or an iot.
Ice also, it should be serving a purpose like either collecting
some information or delivering some outcome.
Now that is the least any iot enabled device
or thing in iot will require today,
even though the bigger devices have diverse features
and can be more powerful.
It is a miniaturizing of computer hardware
that has truly given the Internet of Things its Edge
in the market today.
It is made it possible
for these iot devices to even function and microscopic levels
in the remotest corners of our world.
And as also Borden focus
on more Target oriented features specific to their goals,
Plus with devices running on incredibly low power levels
and operating with such minimal resources.
The things in iot come with the promise
of implementing Solutions
in the most cost effective way possible without compromising
on the system's accuracy.
This is because at the smallest levels these things are embedded
with just enough technology
for them to communicate over the internet
and other networks to get the data analyzed stored
and processed over that Network
or over the cloud rather than physically housing.
All those Technologies on themselves.
Also such low power levels are perfect for the lead.
Our chips sensors and other low energy components
that is devices used for the operation.
But like I said,
there are substantial equipped to connect wirelessly for
exchanging data and Performing actions based on them.
Also such low energy modes of connectivity make it easy
for these things to operate just about anywhere
while letting us control and monitor them remotely
for almost any part of the world.
So it is as if the British
technology Pioneer Kevin Ashton who coined the phrase internet
of things back in the 1999 somehow knew of these potentials
that I would have brings out.
Devices today so let us move on now.
Let us see what things can do for you.
So one of the most popular implementation of these things
and their applications have been in the sector
of home automation today,
the use of Internet
of Things in this industry has got the world closer
to achieving a dream home powered by the smartest systems
using these iot devices
and these systems have been built such
that all its devices
can intercommunicate to allow owners a customized access.
To all aspects of your home
like your lights your logs or the inside environment
or security cameras
or even your total energy consumption with the brand
specific IOD platforms and clouds behind them.
These iot devices are now capable of exhibiting
ambient intelligence
and also stay connected to the internet for driving
real-time operations.
So let us get into some examples
of iot devices used in most Smart Homes today,
and we must begin with the tunable smart Lighting systems
like Philips Hue smart home lighting
or the tp-link.
Color Smart Wi-Fi
that can be easily controlled
using your smart phones with the option
of creating custom lighting automations
from anywhere in the world by just being connected
to the internet with smart lights
like these you could turn them on or off anytime
and tune the color
and the brightness of Lights in your home's remotely
no matter wherever you are moving on.
Did you know that locks
and keys are getting old fashion nowadays.
It's not a joke.
Smart locks are a part of almost every smart home today
and the August smartlock third generation is one
of the newest and the phones on the market today.
What makes it Stand Out Among its peers is
that it comes with bluetooth support
so that you won't need to worry,
even if your internet is down somehow with a device like this
that attaches to your existing that bolts you will always know
if your door is completely closed and locked
with its doors and Technology.
It'll even remind you to lock your doors
when you leaving or tell you
if they were properly closed or not based on the location
of your phone,
you can even create secure virtual or temporary keys
for your neighbors
and guess just in case now
the next iot device is Nest learning thermostat it is one
of the best examples of an iot device today as a part
of your home automation system
this intelligent device Lawns your schedules
and preferences over the first couple of weeks too
smart in the cooling and heating systems in your home
and then regulate your home temperature
automatically to save electricity by 20%
among a ton of other features.
So to reduce your energy bills,
it uses sensors
and your phone's location to realize the whether the time
or the environment inside
and when nobody's home it will shift into energy saving.
Mode apart from the thermostat Nest also
provide smart cameras with smoke
and carbon monoxide alarms
to better manage your home's atmosphere
and Security Plus it lets you manage
and monitor each of them through its dedicated app
or even build your own Revenue channels upon them.
There is also one more device that I think is worth
the mention in the home automation systems of today.
It's mostly just a doorbell for your front doors.
Only a lot smarter the video dogus
by ring is a modest name for all the things
that it can do.
It's like Always home
as it lets you answer your door from just about
anywhere it will in fact,
even a lot you of motion within 30 feet of your door
and stream its live footage onto your phone.
And if somebody rings a bell you can let them know
that you're not home
and even have a word with them
with its right Vision
the wide-angle high-definition recordings get simultaneously
saved were Cloud for your use anytime.
Now before I tell you
what else iot devices can do for you virtual or home assistance
like the Amazon Alexa the Google Assistant Siri
and Jibo are also attending application of the internet
of things and have made it big
in the home automation systems of today.
Now, other than making your Home Smart iot devices are
also capable of intelligent Lee tracking your health and fitness
and as far as Fitness goes the Fitbit tracker
or the Apple watch
and other such smart watches have been a revolutionary
success and along with your Fitness aspects.
They also solve your day to day Health Care needs
like checking your heart rate and perspiration levels
and tracking your body temperature to predict a cold
or a flu on the Are you will be amazed to know
how the other iot devices are saving more lives
and making Health Care better
like the Philips Healthcare is medication dispensing service
or the hero smart pills dispenser are the most
successful use cases today
and exemplify the potential of iot in the healthcare sector.
It is the perfect reminder for your medications
and is a convenient measure for the elderly
and the sick alike
with the option to remotely manage it with your smartphone.
You could also monitor the medication
of your near and dear ones and be notified upon running low
on any Medication at all
plus you could also get info on the right medication
for common health hazards.
And now with
how integral smart phones and internet have become
in our lives Health
and Fitness Solutions using iot will only
get more and more prominent in the recent years to come.
So apart from the personal benefits
that you can get as a consumer or user iot also aims to impact
the greater good the talk
of smart cities is prevalent these days
where the waste management system is more efficient.
You shall in the environment is checked and even
the outdoor lighting
and traffic signals are equipped
with motion sensors to ensure energy conservation
by turning on only upon detecting traffic updates
and its route
or upon sensing pedestrians and Vehicles nearby.
Now, let us look at some of the other sectors of service
and Civic amenities
that these iot devices are powering to up the efficiencies
of our world today.
One of the most visible Smart Systems in place is
in the domain of transport and Automotive today
and along with more and more connected car platforms
today there has Also been a lot of hype
around smart and self-driving Vehicles hitting the lines
of Transport very soon plus with the maps
providing real-time Intel
and the modern vehicles housing numerous sensors in them.
You could always tell which way to go or
what parts need to be looked at.
Like say tracking the Fulton's option altitude and maintenance
issues of flights in real time without waiting
for it to land every time can help in anticipating
problems to schedule maintenance prior to its arrival.
So that deal is and mishaps are minimized
in the aviation industry another major area
of Application for iot devices isn't the field
of Agriculture or sector
that is often neglected.
Despite.
Its utmost importance can now be brought up to speed
with iot with several cheap and minimal sensors
that monitor the best climate and soil quality
for the right kind of crops or other smart devices
that ensure the efficiency of automated irrigation systems,
even the gardens in your home or the trees in smart cities
or even the plant life and vegetation in agroforestry
or wildlife habitats
could be kept in check the most Innovative iot devices
to they include the smart watering system Blossom,
which can create optimal wanting she Duels
for all the plants in your home based on real-time weather data
and forecasts and will regulate all your sprinkles accordingly
and allows you complete control over them through Bluetooth
or the internet
and also the clean grows
carbon nanotube probe is one of the best iot devices for farmers
and gardeners all
around the world today with sensors to monitor the intake
of nutrients in the crops to better manage farming resources
and improve the quality of their fields farmers
can now order Maturity rate
and the color of props for better yields and faster.
It's a production now.
It might be a little ironical of me to talk about one
of the very first Industries to be made smart
on such a late note in today's session,
but I really just wanted to save the best for last
and believe me the sector of retail and Logistics is
where the internal things promises the most
astounding results
with iot devices already being extensively used
in shopping restaurants hospitality industries
and many other businesses to control
the supply chain effectively.
And obtain valuable insights based on them
and manage their logical or merchandising expenses
in the best possible ways.
Now the Q hop is one
of the leading examples of internal things being used
in the retail today typically
designed to bring in seamless autonomous check
out Technologies to all retail verticals
by digitizing the checkouts through RFID tags
that only unlock after its payment is processed.
The sole purpose
behind it was to allow users to sell check out in stores.
Although it is mainly used for security reasons today
to inhibit Petty thefts.
Doors, but guess what that was
before the Inception of Amazon go earlier this year.
Now the technology
inside this convenience store will really seem
like it was pulled out from the future somewhere.
Although it is so new
that there are only three locations
with this feature has six toes so far.
So operate and are managed by the online retailer Amazon.
These stores can give you a shopping experience
unlike any you've ever
had before with the idea essentially being grab-and-go.
These stores will just need your Amazon go apps
for you to enter and then employees.
Eurovision machine learning
and Sensor Fusion to automatically add items
that you pick instantly on
to your virtual card will also remove them
of just as promptly if you keep them back now
once you've grabbed all the things
on your shopping list and are ready to exit all you need to do
is just walk right out of the store.
Yes, no more lining up in busy cues of the usually
limited checkout counters to wait your turn for the purchase.
You can exit the store
without even having to pull out any cash
or a card and you will find the amount for all the things
that Walk out with the beaded simultaneously from the balance
in your Amazon account such a smart system could also check
the inventory regularly to notify retailers
on the need for restocking
and even helps to manage the supply chain
and a better way.
Now, there are also other promising areas in logistic
that iot devices have started to unpack like in the case
of shipping cargo or Fleet Management smart Bluetooth.
Low energy tags are attached to the items being moved
for remotely tracking the exact locations speed of transport
and storage conditions.
For instance.
The things dot IO is a simple iot platform
that provides a dedicated cloud-based dashboard
for better Logistics by enabling access to real-time
and reliable inputs from it's connected smart sensors
and paired location trackers irrespective of wherever
they might be so
although use cases
and the respective iot devices are things.
We just went over our only a select few
of the countless applications of the internet of things
that can drive almost any sector today
and these iot devices could be things themselves
or even attached to someone
or something to make it a thing in iot.
Like even a person
with a heart rate monitor could be a thing in iot
as it collects and provides information
that can act as inputs
to other Smart Systems to operate on so the Baseline
for anything to become an iot device could be laid
out as any object
with a unique IP address for communication over networks
and the ability to gather and transmit data
or receive data
and perform tasks based on it
and the Ameri technology in these devices are
what interact with either the internal States
or the external environment.
To capture all data
and drive decisions made up on them.
Now with the internet of things being such
a large community of different devices the challenges
of iot devices mainly start with communication as the protocols
and languages used by each of them vary.
Hugely due to the lack of common standards
for all of them yet
and this lack of a uniform
and secure standard across all iot devices,
which is great security risks
making them highly unreliable for most important.
Missions are into operations today.
Also without customers being assured of the privacy
and security of the data.
There is just no reason for them to risk using
or adopting such insecure Solutions.
Do you really understand what the role of security
and iot devices is today.
Well, let me tell you about a major Cyber attack
that happened back in October of 2016 a large distributed denial
of service attack dubbed Mirai affected DNS servers
on the east coast of the United States
which disrupted Services all across the world
upon further investigations.
This issue was tracked back to the hackers
that infiltrated smart networks
through the iot devices being used in them
like the routers or the camera.
So that brings about the situation.
We're in our devices and data are all remotely connected
and stowed upon these networks
and its security gets compromised.
We might even not know it over the first few days or weeks
and it might be just too late by the time we
do realize now the way in
for these hackers were undoubtedly a result
of poor practice at some end
like say the use of default passwords rather than changing.
Hence, the adoption of better practices
and the reinforcement
of proper authentication Network segmentations encryption
and cryptography can still make the things
and its systems quite secure given
that we start making sure of building them up securely
from our end as well
plus the issue of overall connectivity is also a feat
that our world is still driving to achieve
but hasn't been able to accomplish it.
That brings us to the last topic for today.
So let me tell you the important things
that you should know to build an iot device.
So building an iot product
or device or solution must be done thinking
about the relevant purposes
that it can serve
and the ways in which they can be prepared to work
for at least the next couple of decades along with the option
and space for quick improvisations
and upgrades and just
like we saw in all our examples today.
You must have understood the two important categories
that iot devices are mainly made up
of the first is the hardware as And the underlying room
is to aggregate the hardware in the most minimal way.
You can without compromising on the primary features
that you want your device to use.
Now.
These features are mostly due to Bluetooth low energy sensors
or beacons connected to the internet
or a customized product with probably a combination of these
on a circuit board made up of a semiconductor like Silicon
and may also house or the components
like transistors resistors receivers transmitters actuators
and integrated circuit or a microchip.
So the most devices like a smartphone
or a result of such combinations on a little more complex level
with proper casing
and with the second most important aspect
of these devices.
So the second important category
in iot devices is the software aspect
on the device level the size of your software will depend
on how minimal
or bulky your device is.
And what are the components
that your housing on it on the most basic note.
The device software
only needs to be enough for handling the operation
of your device driving components to collect data
and converting them into transmitters.
Fun connecting two networks driving the transmitter to send
data and the receiver to receive data to and from the network
and converting it to the machine understandable form for driving
components to perform some task or display.
Some outcomes based on the received instructions.
Now all this will require very minimal software
unlike the software on your phones
that come with entire mobile operating systems
all the other important software for intelligence
and smartness in these devices
would be provided by the underlying Cloud infrastructure
and even by the Mobile apps or web dashboards.
The software aspect of iot devices is in fact
what controls the hardware aspects to sense
some information or perform an instruction
while both aspects are the most integral parts of an iot device.
You might also need to overcome a few more hurdles
before you iot device is truly ready
to implement smart Solutions
and Power Smart Systems of our world today.
These hurdles could be connectivity issues
or compatibility problems or security and privacy concerns,
but don't let the obstacles dishearten you
in Just consult people with knowledge on the same.
While the Internet of Things equips a multitude of domains
and millions of devices with connectivity every day.
There are still a few domains,
which I find a tad more interesting than the others
and they are iot in everyday lives iot in healthcare
in smart cities agriculture Industrial Automation
and finally in disaster management
first, we have Internet of things in our daily lives.
This is probably a Hallmark in the iot industry
as it's also One
of the first industry is to deploy iot at its service.
So let me give you an example of
how iot can service in our daily lives
consider a home appliance such as your AC currently,
what you do is
that you go home you turn on your AC and wait
for it to reach a temperature
that you like say about 25 degrees Celsius.
So does anybody see a problem here?
No one that's probably
because there isn't a problem here.
This is a perfectly functional setting.
But what if it could be better
what if when your car was five minutes away
your AC received a message.
What if it was connected to a cloud
which had a dashboard containing all the relevant information
like the location of your car the outside temperature
and the temperature
at which you liked your room your AC could then turn on
before you arrived and create an Ambience
that you like?
Wouldn't that be something?
Well, you can remove all the waters
from the previous scenario as this no longer.
Must just be a figment of your imagination.
Thanks to the internet of things iot can connect your fitbit's
to your vehicles from your smart phones
to your in-flight services from home appliances like your ACS
to whole entire cities.
Maybe this is
what Kevin Ashton meant when he spoke of this concept
where every area sensor man and machine could be connected
to one another.
So I hope you all are with me till here.
So next we have iot in healthcare.
I'm sure most of you have heard
about the Smart Medicine dispensers by now
as the name suggests.
It's basically a smart appliance
that stores dispenses and manages your medicines for you.
Now.
This is a very small piece
of a very big picture the health care
and the general practice
of medicine majorly faces issues in one
or more of these three things.
We have your research devices and care.
Medical research has to rely upon leftover data
in control settings for medical examination.
It lacks real world data,
which can solve critical conditions iot
could be the answer to all these problems The Internet
of Things opens ways
to a sea of valuable data through analysis
and real-time testing Internet of Things empowers
Healthcare professionals and improves the quality of care.
Finally It ultimately reduces the unsustainable.
A high costs of medical devices.
So here is an example.
This is basically an outline of how a care device
Works a care device has certain parameters
that are considered safe.
Once one of them is breached
the sensor immediately realize this message
via a secure gateway to a cloud now.
It is vital for this gateway to be secure
as it holds all your valuable medical records the cloud then
passes a remote signal to a Mod device
that is monitored either by a nurse
or a caretaker at home.
The beauty of remote patient monitoring is
that patients can now replace a long wait
at the doctor's office with a quick check-in data share
and instructions on how to proceed iot.
Hence Bridges the gap between reading devices
and delivering Healthcare
by creating systems rather than just equipment with that said,
let's move on to one of the most talked
about prospects of iot the smart cities.
Now the thing about a smart city.
Is that a smart City solution is very specific to that one city.
The problems faced in Bombay are very different
than the problems faced in Delhi.
Even Global issues
such as Waste Control traffic management availability
of finite drinking water housing
and pollution impact
different cities with different intensities.
So the only way to really make a city smarter.
Is to cater specifically to its problem one such problem
consistent among most Urban cities is traffic.
So imagine an intelligent device
like a traffic camera a camera that can monitor the road
of traffic jams accidents rains Etc
and communicate that status to a Gateway.
Now this Gateway also
receives data from various other cameras from all
around the city this
in itself could form a Citywide traffic system.
Now where can we use it?
So let's say the municipal Corporation has decided
to repair a road
which connects to a highway for more than one city.
Now this could cause a massive congestion a smart camera
could send this insight
to the Citywide traffic system now considering this is
a smart traffic system it quickly learns
and predicts patterns in traffic.
It will analyze the situation predict its impact
and relay this information.
Two other cities
that connect to the highway via their own monitoring systems.
Now the traffic Management Systems,
then can derive routes for cars around this projects
live instructions could be sent to drivers via radio channels
and their respective smart devices.
What is a bonus is
that if there is a school or an office building
which has no other way
but to use that road under construction,
they can be automatically Ali called and asked to reschedule
this creates a network of self dependent systems
which leverage real-time control
and this is just one example of the potential benefits
of iot applications.
So next we have a sector
which is most neglected despite the importance.
It holds that is agriculture.
However manual handling often results in loss
of energy labor cost and other inaccuracies,
which may call its Is less effective internet
of things here can provide
with the number of solutions Precision
farming smart irrigation
and smart Greenhouse are a few to name.
The first two cases are pretty similar as in both cases.
There are sensors detecting various parameters
at each level of the soil.
We have moisture content temperature
and weather conditions
one can tell the correct depth to so the crops
or the right time to water them,
but One of the more intriguing Solutions is
the Smart Greenhouse a lot of you might be wondering
what is a greenhouse.
So the greenhouse is a farming technique
where you can increase yield by controlling natural parameters.
So if we could use embedded devices in these greenhouses,
we could not only just monitor it regularly,
but we could practically control the climate
inside the greenhouse.
So how the system works is
that the sensor sends various Toes and areas
of issue inside the greenhouse.
They then relay the information via a connection gateway
to the cloud
which then sends remote commands
via the same Gateway back to the sensors.
So once these signals are acknowledged they
are passed on to the switch gear
which then activate the lighting pumps and turbines
inside the greenhouse
which then create artificial sunlight and sometimes
artificial precipitation.
Inside the greenhouse what more from the connection Gateway
as you can see,
there is a monitor
through which 24/7 real time monitoring has come into effect.
Also, you could use any smart device to remove the conditions
inside your Greenhouse.
So with elimination of irregularities
and human errors,
this creates a much more efficient system.
So with that we conclude the role of Internet
of Things in agriculture.
So next we have another domain
where iot could prove to be a game-changer?
This is one of those fields
where both faster developments as well as the quality
of products are
the critical factors for a higher return on investment.
I'm talking about the Industrial Automation sector,
but there are still a few problems that need
to be addressed in this sector.
I think you all might have already guessed.
What I'm going to say.
Well I OT comes to the rescue Again The Internet
of Things improves the line of command immensely.
It optimizes packaging
and makes quality tests so much easier to run
what more you do not even have to worry about the training cost
of too much staff or a lot of stuff going on a holiday
because these factories pretty much run
on their own you can monitor the supply chain in Real time
while keeping an eye on your inventories
with iot applications one could even re-engineer products
and their packaging to deliver better performance in both cost
and customer experience.
So this brings me to the final application for today
that is disaster management.
Now the iot cannot stop disasters from happening,
but it surely can help
in preparedness and resilience during one due to high.
High population density poor evacuation infrastructure
and exposure to severe weather events developing countries are
more exposed to the risks of natural disasters
and often have limited means to sustain the effects
as a consequence
according to a study more than 95 percent
of all deaths caused by a disaster occur
in developing countries iot can compensate
for this by prediction preparedness response.
And Recovery to rescue developing and emerging
countries from their vulnerable positions.
Let me give you an example of a forest fire
where iot could be used as a solution to be prepared
for critical incidents.
Like a forest fire sensors are installed near the perimeters
of the forest.
They contribute to data feeds
about temperature and carbon emissions to a control room,
which is common to the entire town.
This is done via a secure network.
Gateway in real time,
they anticipate problems and put defenses
in place to mitigate that impact namely
the police the nearby hospitals
and most importantly the fire brigade
if such an emergency occurs near urban areas people
and communities are alerted
via social media conventional media channels and SMS.
This brings me to the end of all applications
that I had listed now that we've gone through a few.
Applications of iot let's move on to its future scope
in just one year
internet-connected devices went from five million to billions,
but there are still a few areas that need work.
So the device is forming the base of iot Are Wireless
in nature and recite at very remote places
where energy is a very vital issue.
So with growing functionality of each device,
we need algorithms and Hardware that are energy efficient.
To avoid quick draining of batteries.
We need to make sure
that sensors are active for longer durations,
like any other advancement in technology,
even in iot security is a standout issue.
This issue keeps getting bigger
with more and more devices being connected
to one another we need information seclusion methods
to benefit end clients
and secure that data and privacy.
No, it will be really tough to implement the anytime concept
of iot in reality,
but this needs continuous work
and the closest we can get to this is only
by reducing the complexity of each existing time systems.
We need continuous work to reduce the gap
between near real-time
and actual real-time iot applications promise to bring
immense value into our lives.
In fact the number of Openings for iot professionals is
at an all-time high.
So I think it is the best time to begin
exploring the true potential of this technology from smartphones
to Smart cards to that Fitbit on your wrist iot represents
millions of rupees of potential for an industry
as The Internet of Things becomes more important
for companies of all sizes.
It professionals are beginning to seek out roles.
Related to this growing Niche.
So before going into the session,
let's quickly look at our agenda.
So we'll start off by discussing a little bit
about the current scenario
in iot market trends current salary situation
and projections in iot
and how it compares
to other sectors after which we shall see the salary Trends
in iot by both experienced and functional areas.
Also I shall be talking about the companies
that are hiring for iot as well as the roles
they offer then I shall Discuss the skill set you must acquire
before applying into the sector.
And finally we will see an outline of the training
provided a Teddy Rekha.
So without Much Ado,
let's get straight into the module.
So with the proliferation of connected devices iot
has ventured successfully across all working areas in India
and abroad India is quickly becoming one
of the largest hubs for iot
across the globe companies are viewing iot as a key enabler
in various fields
and are Come to integrating it into existing infrastructure
as the adoption of iot has increased
iot Market is also expected to increase quite exponentially.
So with that let's talk
a little bit about the salary Trends and projections.
Now the median salary
of iot professionals is 15 Point 2 lakhs per annum
which is according to the 2017 study this figure
compared to the 8.65 lakh per annum in the eye.
It sector gives about 76 percent
more to iot professionals iot being a relatively new entrant
Professionals in this area are commanding higher salaries.
This number is predicted to go up as much as Seventeen
point five lakh by 2019.
So let's see
how iot does compared to other contemporaries?
Like I said before I OT pay is 76 percent more
than the it sector interestingly only 33% of iot.
Specials have salaries under 6 lakh per annum as
compared to 58% in the it sector also we can see around 16 to 18%
of iot Professionals in India earn in a salary bracket
of 10 to 25 lakhs per annum almost 5%
of iot professionals occupy.
The largest salary bracket that is 50 plus LP a while.
In other domains barely
two percent of the workforce commands this figure
with That said let's look at the salary trends
for it Professionals in India
and the US according to the current Financial year.
So what I've done is I've divided this
into two parts salary is according to functional areas
and then according to the experience level first up.
We have salary Trends by functional area.
Now as you can see,
I've divided this into four functional areas,
which majorly command the iot industry one
is embedded systems next system programming
then we have the Engineering design and finally
application programming in one end of the spectrum.
We see application programming being the highest paying job
in India with an average of 13 lakhs per annum
whereas its us counterpart
pays about sixty nine thousand dollars per annum
on the other side of the spectrum.
We have embedded Technologies,
which pays an average of eight point two lakh
per annum in India, but in u.s.
It is the highest paid in the sector with an average
of $9,000 per annum the system programming
and engineering design sectors
lie somewhere in between with 11 point 6 lakh per annum
and 10 lakh per annum
respectively as for the experience level more than 90%
of freshers in the iot field fall under the 6 lakh
per annum salary bracket,
which is not a bad number for a fresher
on the other end somebody
who has a decade of work X in this field can quote up
to a 30 + L. Your salary now.
What is interesting here is
that from each experience level bracket to a higher one can lead
to almost 50 percent jump in the salary with this.
I think we've covered the salary aspect
of this industry.
Let's move on to the career aspect of iot.
I'm sure a lot of you must be waiting
for me to disclose the names of companies
and the roles they offer.
So the concept of iot has given rise to a new era
of economic growth
and companies are Looking at it to transform
their businesses now here you can see two types
of companies firstly there are companies like Dell and IBM
which have a vested interest their core products
and services are built around delivering
and facilitating iot,
but there are also companies like Verizon and Accenture
where it is more about preparing for the future.
Although iot hasn't completely changed the face of it.
It has definitely created new opportunities
for job seekers in the markets.
I Be elaborating on this in the following sections
of the module
which brings me to current roles for iot professional.
Now if you were to ask me what field do you need to be
in to land a job
in iot the answer would be many iot works
on many layers software plays a key role in usability
and functionality Network layers
are key to infrastructure and Hardware defines capabilities
and development opportunities involved in an iot solution.
Existing professionals with transferable skills
will definitely find progression opportunities with iot.
The iot is based off of many technologies
that it professionals are already familiar
with IP experts both software
and Hardware engineers.
And even UI designers could find themselves to fit in somewhere
in the industry.
These companies could be ranging from startups to Tech giants
that are driving the industry,
but with various domains
that have adopted the potential usage
of industrial IO It has led to a massive job creation.
But unlike the other fields iot job market is a complex one
given the wide variety of skill sets required
in this place.
Here are some emerging job profiles for iot professionals
a few years down the line first up.
We have solution architects.
Now, what a solution architect does is
that it takes an idea
and converts it into a design and then takes the design
and converts it into code now a solution.
Architect may not be a back-end developer,
but it takes the requirements of the company and then turns it
into solution blueprints
which brings me
to another very important player in this end-to-end execution,
which is the product manager
the responsibilities of a product manager may be many
but mostly it revolves around making sure
that the clients get the solution they asked
for on time moving on now
when we talk about iot we talk about huge amounts of data.
And where there is huge amounts of data,
we have analytics on top of it.
So this is a job profile created specifically
for the iot Industrial Field,
which is the industrial data scientist.
These are data scientists
that basically work with sensor data
and have expertise on it.
Which brings me
to another very important industrial job profile,
which is the industrial Engineers.
Now as I've mentioned
before Hardware plays a key role in the development.
Moment of connected devices for that.
We need material Specialists who can work on semiconductors
that are used to make microcontrollers
and microprocessors in those connected devices.
Next up.
We have another role
that is created specifically for the iot field.
Now iot has pushed the developers to go
beyond a full stack.
We need developers at the back end
who are not only proficient
in software coding but So in Hardware programming, hence,
we have the full stack developer
which brings me to our final emerging role in iot
as we have all known and predicted very soon.
Our shop floors are going to be taken over by robots.
So what do we need?
We need a supervisor
who can take care just in case there is a miscalculation
or a breakdown.
This has created the final job profile that
I'm talking about today the robot coordinator.
With that said let's move on to our next topic.
So while iot is poised to be the next great boom
in the field of jobs lack
of skill is citing the biggest barrier for companies
that are looking to implement this technology to be considered
for any of the rules mentioned earlier the qualifications.
You need are no different
than an IT development roll Proficiency in coding
and fundamental in object design are at least required
for your resume to be looked
But the skills you need here are way more than that businesses
who engage in iot Technologies are businesses
that are invested in the future.
They are seeking forward-thinking professionals
who not only meet standards in academics,
but also have soft skills
and Innovative thinking I am talking about people
who can integrate deep knowledge in embedded Technologies
and concepts of cloud computing and Edge Computing.
These people are tipped most Be in demand next up.
We have networking and communication protocol.
These are very important
as devices need to be connected to one another at all times
in real time.
Finally.
We need professionals
who can convey a complex idea
in simple ways either speech written communication
or other abstract methods.
So the iot job market is the perfect platform
for job Seekers to Showcase their skills
and for companies to establish relationships
with These professionals
so this is the best time to get into this technology
and who better to guide you then the iot trainers at Eddie Rekha.
So Ed Eureka provides an iot certification training
and the outline of a syllabus look something like this.
This training course is tailored specifically
by industry experts
and a takes you through Concepts such as iot framework ecosystem
and solution architecture it
also teaches you networking protocols and application.
Our in-depth you shall be working
with Raspberry Pi and sends hats
and will gain hands-on experience with Azure iot Hub
and Alexa voice.
This course also has a real life demo
and case studies
which will help you master the iot technology.
So concluding I would like to say
that iot is amidst an explosion this definitely suggest
optimistic future for professionals looking to dive
into this industry.
So grab this opportunity while you still have it.
Thank you and have a great day ahead.
I hope you have enjoyed listening to this video.
Please be kind enough to like it
and you can comment any of your doubts and queries
and we will reply them
at the earliest do look out for more videos in our playlist
And subscribe to any Rica channel to learn more.
Happy learning.