IRJET- Smart Switch Board Compatible with Google Assistance along with Face Recognition and Security System
0 likes10 views
This document proposes a smart switch board system that is compatible with Google Assistant and includes face recognition and security features. The system would allow users to control home appliances and devices using voice commands through Google Assistant or face recognition. It would use a Convolutional Neural Network algorithm to process images for face recognition at the front door to identify people. The system is designed to be cost-effective and provide easy and secure control of devices either through voice commands or switches, with priority-based intensity control of devices. It aims to make home automation more convenient, especially for elderly or disabled people.
![International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 746
Smart Switch Board Compatible with Google Assistance Along with
Face Recognition and Security System
Soniya Abhyankar1, Ajayan Saroj2, Shital Patole3
1,2,3B.E Student, Dept. of Electronics Engineering, AET’s ACE, Maharashtra, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - With the advancement of technologies in the
market the standards are needed to revolutionize an attempt
to implement 'Smart Switch Board' to operate over fullyhome
automated system which provides the user vast reliable
features from simple switch control to load current intensity
variable features, face recognition approach to have
predefined designed features for the user to control different
parameters for different household appliances, which can be
achieved by algorithms like CNN as it provides higher
efficiency additional to face recognition and security system
for the home is an optional feature to the user to stay
protected against any theft which may enter the house such
that if the face is recognized with the known database then it
is matched with the desired user parameter else it is identified
as theft and desired actions are performed along with rising
alert.
Key Words: Google Assistance, Convolution Neural
Network, Face Recognition, Node MCU
1. INTRODUCTION
Human life will become more comfortable when it will have
devices which will listen to it. In case of elderly people and
physically impaired people it is very difficult to get up and
operate devices hence in ordertomaketheirlifecomfortable
this paper proposes the system which will take voice
commands and operate devices. The concept behind the
project is to control home devices by sitting at the place, this
can be done by our proposed system which involves google
assistant which will listen to our command and operate
devices accordingly this includes turning devices on and off
also varying their intensity as per our need. This also
provides door lock and unlock system by voice commands,
the person will come to know who is at the door this can be
done by face recognition system at the door, and hence the
system also provides safety to person. Further additional
home security featurescanalsobeincluded.Thiswillprovide
cost efficiency and reliability. This can be achieved by CNN
algorithms and advanced processors.
1.1 Literature Review
According to survey there exists many systems which can
control home appliances using android and other operating
systems each system has its uniquefeatures.Currentlymany
companies official registered are working to provide better
home automation system features. Following systems
describes the work beingperformed byothers.N.Srikanthan
[1] has worked for and also explained modern home
automation systems systems using Bluetooth via PC. It was
working efficiently also but the problem was it was unable
for user to control remotely or using mobile phones. Web
based prototype electrical device control system by
Muhammad Izhar Ramli [2]. They also set the server with
auto restart if the server is currently down. Hasan has
developed a telephone and PIC remote- controlled device to
control the devices pin check algorithm has beenintroduced
where it was with cable network. Hence here the system
worked well but it was wired system. Then an application
was developed by Amul Jadhav [3] ina universal XMLformat
which can be easily ported to any other mobile devices
instead of just targeting a single platform. Each of these
systems has their own unique features and in comparison,
toone another lacks some advancement. Our designed
system has application layer prototype. The application is
able to synthesize the speech data with the help of Google
Voice Reorganization.Thesynthesizeddata areanalysed and
further processing is carried out. In layman words, our
design system provides features of controlling the home
appliances using voice commands. To connect the android
application with the raspberry pi the use of socket
programming is performed. Thisfurtheraddssecurityto our
system. The data are received only by the server at the
specified port and data are further analysed. The project
which is proposed by us is different than the system which
were developed earlier .This project uses android operating
system which is now easily available on almosteverymobile
phone hence easy to operate.
2. PROPOSED SYSTEM
Smart switch board compatible with google assistant and
face recognition system is the modified version of home
automation system or smart home. Proposed system will
provide access to the switches in home, offices without
physically moving or reaching to the switches. We will be
able to handle them by just ordering to them. For eg.by
ordering fan to turn on fan will be turned on.it will be voice
controlled. Intensity can also be varied as required of the
appliances’ only that the system will have priority- based
handing or operation. And all these things will be done in
following mode.
1. Switch Mode 2.Voice Mode
Switch mode: This will be the normal mode of operation,the
way we do in our home, physically going to the switches and
turning them on or off.](https://image.slidesharecdn.com/irjet-v7i2162-201110073207/85/IRJET-Smart-Switch-Board-Compatible-with-Google-Assistance-along-with-Face-Recognition-and-Security-System-1-320.jpg)
![International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 747
Voice Mode: In this mode we will be having facility to
operate switches by voice-controlled system. This will be
done by accessing google assistant in phone. The second
method is based on the principle power electronics devices
the variable output can be achieved by SCR followed by
Bridge circuit you are simple the output is control based on
firing pulse of the SCR which is we are changing the duty
cycle of the input and desired varying output is achieved.
Basic principle used here is pulse width modulation via the
firing angle of SCR Two- Way simultaneous working: Face
recognition and voice control use over here are the two
particular method to operate the switching at load the voice
control which is achieved by using the Google assistant and
the face recognition which is achieved by the use of the
convolution neural network but to achieve the maximum
efficiency the output of both face recognition aswell asvoice
control are passed through the OR logic circuit due to which
unless both the circuits are of the switching be done .
2. Diagram of the proposed system
1. Face Recognition: For face recognition, we are using a
camera as an input source. Even a webcam can be used for
furtheradvancedimageprocessing.WeareusingConvolution
Neural Network algorithm for image processing and
enhancement of the captured images.
2. Convolution Neural Network: Convolution neural
networks (CNNs) are a tool, a means of adaptive image
processing, and a link between general feed-forward neural
networks and adaptive filters. Two-dimensional CNNs
consist of one or more two-dimensional filter layers, with
potential non-linear activation and/or down-sampling
functions. CNNs have important translation invariance
properties and local spatial relations (receptive fields). We
presentadescriptionoftheconvolutionnetworkarchitecture
and an application on a mobile robot for practical image
processing. A CNN is used on an autonomous sewer
inspection system to identify and classify cracks.In all cases,
the filter sizes used were 4x4, with non-linear activations
between each layer. From input to output, the number of
feature maps used in the three hidden layers was 4, 4, 4.The
network was equipped using a dataset of 48x48 sub-regions
from 320x240 pixel frames of 30 still images taken from a
pre-recorded video of sewer pipe inspection. Fifteen frames
have been used for testing and fifteen for network
performance validation.
Features
1. Cost Effective
2. Easily compatible with normal device AC, Fan etc.
3. Highly secured plus Backup
4. Supports face recognition
5. Priority based intensity control of devices
6. Voice controlled operation of devices
3. CONCLUSION
Thus for the home automationwehaveusedtwoapproaches
which is based on face recognition and voice recognition the
utilisation of the above two approaches ensures us
maximum efficiency and working possibilities for all
different conditions and caes. To achieve this particular
working we have implemented an or logic which will ensure
us that even if anyone particular approachfailstheother one
gives us the backup for the failed system and vice versa.
Depending upon user priority the particular parameter of a
load are varied.
REFERENCES
[1] N. Sriskanthan and Tan Karand. “BluetoothBasedHome
Automation System”. Journal of Microprocessors and
Microsystems, Vol. 26, pp.281-289, 2002.
[2] Muhammad Izhar Ramli, Mohd Helmy Abd Wahab,
Nabihah, “TOWARDS SMART HOME: CONTROL
ELECTRICAL DEVICES ONLINE”, Nornabihah Ahmad
International Conference on Science and Technology:
Application in Industry and Education (2006)
[3] E. Yavuz, B. Hasan, I. Serkan and K. Duygu. “Safe and
Secure PIC Based Remote Control Application for
Intelligent Home”. International Journal of Computer
Science and Network Security, Vol. 7, No. 5, May 2007](https://image.slidesharecdn.com/irjet-v7i2162-201110073207/85/IRJET-Smart-Switch-Board-Compatible-with-Google-Assistance-along-with-Face-Recognition-and-Security-System-2-320.jpg)
IRJET- Smart Switch Board Compatible with Google Assistance along with Face Recognition and Security System
- 1. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 746 Smart Switch Board Compatible with Google Assistance Along with Face Recognition and Security System Soniya Abhyankar1, Ajayan Saroj2, Shital Patole3 1,2,3B.E Student, Dept. of Electronics Engineering, AET’s ACE, Maharashtra, India ---------------------------------------------------------------------***---------------------------------------------------------------------- Abstract - With the advancement of technologies in the market the standards are needed to revolutionize an attempt to implement 'Smart Switch Board' to operate over fullyhome automated system which provides the user vast reliable features from simple switch control to load current intensity variable features, face recognition approach to have predefined designed features for the user to control different parameters for different household appliances, which can be achieved by algorithms like CNN as it provides higher efficiency additional to face recognition and security system for the home is an optional feature to the user to stay protected against any theft which may enter the house such that if the face is recognized with the known database then it is matched with the desired user parameter else it is identified as theft and desired actions are performed along with rising alert. Key Words: Google Assistance, Convolution Neural Network, Face Recognition, Node MCU 1. INTRODUCTION Human life will become more comfortable when it will have devices which will listen to it. In case of elderly people and physically impaired people it is very difficult to get up and operate devices hence in ordertomaketheirlifecomfortable this paper proposes the system which will take voice commands and operate devices. The concept behind the project is to control home devices by sitting at the place, this can be done by our proposed system which involves google assistant which will listen to our command and operate devices accordingly this includes turning devices on and off also varying their intensity as per our need. This also provides door lock and unlock system by voice commands, the person will come to know who is at the door this can be done by face recognition system at the door, and hence the system also provides safety to person. Further additional home security featurescanalsobeincluded.Thiswillprovide cost efficiency and reliability. This can be achieved by CNN algorithms and advanced processors. 1.1 Literature Review According to survey there exists many systems which can control home appliances using android and other operating systems each system has its uniquefeatures.Currentlymany companies official registered are working to provide better home automation system features. Following systems describes the work beingperformed byothers.N.Srikanthan [1] has worked for and also explained modern home automation systems systems using Bluetooth via PC. It was working efficiently also but the problem was it was unable for user to control remotely or using mobile phones. Web based prototype electrical device control system by Muhammad Izhar Ramli [2]. They also set the server with auto restart if the server is currently down. Hasan has developed a telephone and PIC remote- controlled device to control the devices pin check algorithm has beenintroduced where it was with cable network. Hence here the system worked well but it was wired system. Then an application was developed by Amul Jadhav [3] ina universal XMLformat which can be easily ported to any other mobile devices instead of just targeting a single platform. Each of these systems has their own unique features and in comparison, toone another lacks some advancement. Our designed system has application layer prototype. The application is able to synthesize the speech data with the help of Google Voice Reorganization.Thesynthesizeddata areanalysed and further processing is carried out. In layman words, our design system provides features of controlling the home appliances using voice commands. To connect the android application with the raspberry pi the use of socket programming is performed. Thisfurtheraddssecurityto our system. The data are received only by the server at the specified port and data are further analysed. The project which is proposed by us is different than the system which were developed earlier .This project uses android operating system which is now easily available on almosteverymobile phone hence easy to operate. 2. PROPOSED SYSTEM Smart switch board compatible with google assistant and face recognition system is the modified version of home automation system or smart home. Proposed system will provide access to the switches in home, offices without physically moving or reaching to the switches. We will be able to handle them by just ordering to them. For eg.by ordering fan to turn on fan will be turned on.it will be voice controlled. Intensity can also be varied as required of the appliances’ only that the system will have priority- based handing or operation. And all these things will be done in following mode. 1. Switch Mode 2.Voice Mode Switch mode: This will be the normal mode of operation,the way we do in our home, physically going to the switches and turning them on or off.
- 2. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 747 Voice Mode: In this mode we will be having facility to operate switches by voice-controlled system. This will be done by accessing google assistant in phone. The second method is based on the principle power electronics devices the variable output can be achieved by SCR followed by Bridge circuit you are simple the output is control based on firing pulse of the SCR which is we are changing the duty cycle of the input and desired varying output is achieved. Basic principle used here is pulse width modulation via the firing angle of SCR Two- Way simultaneous working: Face recognition and voice control use over here are the two particular method to operate the switching at load the voice control which is achieved by using the Google assistant and the face recognition which is achieved by the use of the convolution neural network but to achieve the maximum efficiency the output of both face recognition aswell asvoice control are passed through the OR logic circuit due to which unless both the circuits are of the switching be done . 2. Diagram of the proposed system 1. Face Recognition: For face recognition, we are using a camera as an input source. Even a webcam can be used for furtheradvancedimageprocessing.WeareusingConvolution Neural Network algorithm for image processing and enhancement of the captured images. 2. Convolution Neural Network: Convolution neural networks (CNNs) are a tool, a means of adaptive image processing, and a link between general feed-forward neural networks and adaptive filters. Two-dimensional CNNs consist of one or more two-dimensional filter layers, with potential non-linear activation and/or down-sampling functions. CNNs have important translation invariance properties and local spatial relations (receptive fields). We presentadescriptionoftheconvolutionnetworkarchitecture and an application on a mobile robot for practical image processing. A CNN is used on an autonomous sewer inspection system to identify and classify cracks.In all cases, the filter sizes used were 4x4, with non-linear activations between each layer. From input to output, the number of feature maps used in the three hidden layers was 4, 4, 4.The network was equipped using a dataset of 48x48 sub-regions from 320x240 pixel frames of 30 still images taken from a pre-recorded video of sewer pipe inspection. Fifteen frames have been used for testing and fifteen for network performance validation. Features 1. Cost Effective 2. Easily compatible with normal device AC, Fan etc. 3. Highly secured plus Backup 4. Supports face recognition 5. Priority based intensity control of devices 6. Voice controlled operation of devices 3. CONCLUSION Thus for the home automationwehaveusedtwoapproaches which is based on face recognition and voice recognition the utilisation of the above two approaches ensures us maximum efficiency and working possibilities for all different conditions and caes. To achieve this particular working we have implemented an or logic which will ensure us that even if anyone particular approachfailstheother one gives us the backup for the failed system and vice versa. Depending upon user priority the particular parameter of a load are varied. REFERENCES [1] N. Sriskanthan and Tan Karand. “BluetoothBasedHome Automation System”. Journal of Microprocessors and Microsystems, Vol. 26, pp.281-289, 2002. [2] Muhammad Izhar Ramli, Mohd Helmy Abd Wahab, Nabihah, “TOWARDS SMART HOME: CONTROL ELECTRICAL DEVICES ONLINE”, Nornabihah Ahmad International Conference on Science and Technology: Application in Industry and Education (2006) [3] E. Yavuz, B. Hasan, I. Serkan and K. Duygu. “Safe and Secure PIC Based Remote Control Application for Intelligent Home”. International Journal of Computer Science and Network Security, Vol. 7, No. 5, May 2007