See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/340668268

Smart Home Security System using Iot, Face Recognition and Raspberry Pi

Article  in  International Journal of Computer Applications · April 2020

DOI: 10.5120/ijca2020920105

CITATIONS READS

2 2,427

4 authors, including:

Rekha Yashwantrao Biradar

VVP Polytechnic, Solapur.
7 PUBLICATIONS   43 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Task based approach towards Load Balancing in Cloud Environment View project

All content following this page was uploaded by Rekha Yashwantrao Biradar on 11 July 2020.

The user has requested enhancement of the downloaded file.

 International Journal of Computer Applications (0975 – 8887)
Volume 176 – No. 13, April 2020

Smart Home Security System using Iot, Face

Recognition and Raspberry Pi

Manoj R. Dhobale Rekha Y. Biradar Raju R. Pawar

Dept. of Computer Engineering Dept. of Computer Engineering Dept. of Computer Engineering
V.V.P. Polytechnic, Solapur, V.V.P. Polytechnic, Solapur, V.V.P. Polytechnic, Solapur,
Maharashtra, India Maharashtra, India Maharashtra, India

Sharad A. Awatade
Dept. of Computer Engineering
V.V.P. Polytechnic, Solapur,
Maharashtra, India

ABSTRACT recognition. LBP based Face recognition system help us to

In the today’s world the security of the home is the prime detect the person at the door with more accuracy.[Shown in
concern. The traditional methods of securing our home are Fig1.a and Fig1.b][9]The internet of things, or IoT, is a
very easy to break and lead to theft. To protect the home, we system of interrelated computing devices, mechanical and
need to install the costly security system. To overcome this digital machines, objects, animals or people that are provided
problem, we are presenting IoT based solution where we can with unique identifiers (UIDs) and the ability to transfer data
setup a smart home security system. In this paper we are over a network without requiring human-to-human or human-
proposing the system with the help of face recognition to to-computer interaction.[6] Data transmission over a network
develop the automatic door lock and unlock system. It also can be made part of IoT system by embedding them with
gives us the facility to monitor our home remotely and take electronic hardware such as sensors, software and networking
appropriate action if anything goes wrong. The Pi camera will gear.
be attached to the Raspberry pi accompanied with Passive
Infrared and other sensors. Camera captures an image of the
person in front of the door, then real-time face recognition is
done using local binary pattern (LBP) [3]. If person’s image
matches with one of the home members or identified person
then the door will unlock otherwise won’t. Email containing
image of the unknown person will be sent to the homeowner
to his Gmail Id. The proposed system keeps the owner
informed in the real time about the unknown person at the
door of the home.This information will help the user to take
necessary actions.

General Terms
Home Security System

Keywords
Home Security System, Face Recognition, Raspberry Pi

1. INTRODUCTION Fig .1.a LBP feature extraction procedure[9]

A home security is the prime concern to anybody. A home
security system is required to keep our belongings and
2. LITERATURE REVIEW
valuables safe from any kind of robbery or theft. In the United The latest technologies which are available for security
States, there is a home related burglary that takes place every includes RFID card technologies, biometric, OTP based
13 seconds, 4 burglaries a minute, 240 an hour and nearly unlock system and many more. Each of which is effective in
6,000 a day! some of the statistics are 88% of all burglaries some areas but don’t provide the complete security system.
are residential in nature, 77% of all crimes are property Many of the researchers have proposed home security system
crimes, 38% of all robberies are committed with guns, identity but very few used face recognitions for it. Face recognition
theft is the fastest growing crime in the U.S., Canada and UK. has the good potential to offer a strong security system to the
3 out of 4 homes in the U.S. Will be broken within the next 20 home. [7]. A Face Recognition System Based on Local Binary
years [5]. The traditional home security system is easily Patterns and Support Vector Machine for Home Security
breakable and quite outdated. This in turns results in the Service Robot by authors Jia Kailin Wang, Jinjin Zheng,
robbery and also needs installation of the costly security Shiwu Zhang, Jijun He, Xiao Liang and Sui Feng has
system. Our proposed system delivers a cost and energy explained how LBP works in face detection module but lack
efficient solution for home security by using IoT and face the support in case of power failure[8].

45
 International Journal of Computer Applications (0975 – 8887)
Volume 176 – No. 13, April 2020

Fig .1.b LBP extraction procedure [9]

In [2], Face Recognition System Using IoT by authors parts. The system will consist of the identified member’s
Sandesh Kulkarni,Minakshee Bagul, Akansha Dukare, Prof. photos in the database. Any new entry can be made as and
Archana Gaikwad, has the proposed system very well when required in the database. Image will be captured and
presented. The methodology is quite well explained in detail. face recognition will be performed. If face match found the
The working of door lock-unlock on face recognition is very door will open else red color bulb will light up, if the owner is
good, but use of Wi-Fi limits its area of coverage and home at home(specified in android application) then bell will ring
owner cannot access the system remotely. Another else notification will be sent i.e., someone is waiting outside
disadvantages of above systems that they do not inform the your home. An email will be sent to the owner on his Gmail
owner regarding the security breach issue on their Smartphone along with the photo of the unknown
so that he can take the appropriate action on it.
person if the person is not identified. All these activities are
3. METHODOLOGY stored in the database which is in micro SD card in raspberry
pi and maximum size of SD card is 32GB. All the data is
3.1 Proposed System available on the Internet which makes our system more
Our proposed system consists of raspberry pi as a main efficient.
controller which is cost effective and requires very less power
to operate. It has embedded sensor like PIR for motion 3.2 Local Binary Pattern
detection and ultrasonic sensors to measure the distance. The Texture classification is needed for face recognition. For this
stepper motor is used to help the door lock-unlock mechanism Local Binary Pattern (LBP) approach is used. LBP is a simple
which is controlled by Raspberry pi. Block diagram of and very efficient texture operator which labels the pixels of an
proposed system is shown in fig2. image by thresholding the neighborhood of each pixel and
considers the result as a binary number.It was first described in
1994 (LBP) and has since been found to be a powerful feature
for texture classification. It has further been determined that
when LBP is combined with histograms of oriented gradients
(HOG) descriptor, it improves the detection performance
considerably on some datasets. LBP computes local
representation of texture and it is computed by comparing each
pixel with its surrounding neighborhood of pixel. For texture
descriptor, image needs to be converted into gray scale
image.Using the LBP combined with histograms, the face
images are represented with a simple data vector [9].

3.3 Raspberry Pi
The Raspberry Pi is a very cheap computer that runs Linux,
but it also provides a set of GPIO (general purpose
input/output) pins that allow you to control electronic
components for physical computing and explore the Internet
of Things (IoT). Raspberry Pi GPIO is shown in figure 3.All
over the world, people use Raspberry Pi to learn programming
skills,build hardware projects, do home automation, and even
Fig.2 The block diagram of the proposed system: use them in industrial applications. In proposed system, GPIO
pins are controlled using python programming. Any of the
The Raspberry pi consists of a face recognition algorithm, GPIO pin is designed as input or output using available
web application and an Android system. In our proposed software [10]. Raspberry pi operates in the open source
system we have used the cloud to send the data to the remote ecosystem. Its main supported operating system is Raspbian
Smartphone which is at the user side. This is the IoT based and it is open source.
system and can be made automated as and when required. It
has very low power requirement which makes it most
affordable. It requires minimum 5V to work. We can easily
replace any part of this system without disturbing the other

46
 International Journal of Computer Applications (0975 – 8887)
Volume 176 – No. 13, April 2020

6. CONCLUSION
In this paper we have proposed and demonstrated smart home
security approach which requires less cost and provide the
security to our home. We have the system withthe help IoT
and face recognition. The real time face recognition allows us
to detect the stranger atthe door and give notification to the
owner on his Gmail regarding this.

Fig. 5 Working model of the proposed system with

Raspberri Pi 3
Fig. 3 Raspberry pi GPIO layout
7. ACKNOWLEDGMENTS
4. RESULT Authors would like to thank Prof. Shaikh SM., Principal,
After capturing the photo at the door face recognition is done V.V.P. Polytechnic, Solapur, for his constant guidance and
to check whether person is a home member or identified one. family members for their unconditional love and support.
It will determine if the person is stranger or not. If the person
is stranger the led will turn red and the door will not open. If 8. REFERENCES
the person is a stranger then his photo will be sent to the [1] IndiaBurglaryRate,20032016:https://knoema.com/atlas/I
owners Gmail ID, the photo is shown in the figure 4. Working ndia/topics/Crime-Statistics/Burglary-Car-Theft-and-
model of system is shown in figure 5. Housebreaking/Burglary-ra te
[2] Sandesh Kulkarni, Minakshee Bagul, AkanshaDukare,
Prof. Archana Gaikwad. “Face Recognition System
Using IoT”, International Journal ofAdvanced Research
in Computer Engineering&Technology, November-2017.
[3] Zhengzheng Liu, Lianrong Lv, “Development of face
Recognition System Based on PCA and LBP for
Intelligent Anti- Theft Doors”, 2nd IEEE International
Conference on Computer and Communications,2016.
[4] Anjali Patel, “IoT based Facial door access control home
security system”, “International Journal of Computer
Applications, August 2017.
[5] HomeBurglary Statistics 2019: https://reolink.com/home-
burglary-crime-statistics/
[6] Internet of Things-IoT Agenda-TechTarget:
https://internetofthingsagenda.techtarget.com/definition/I
nternet-of-Things-IoT
[7] Dwi Ana Ratna, Wati Dika Abadianto, “Design of Face
Detection and Recognition System for Smart Home
Security Application”, 2017 2nd ICITISEE

Fig. 4 Unknown person identified at the door [8] Jiakailin Wang, Jinjin Zheng, Shiwu Zhang, Jijun He,
Xiao Liang and Sui Feng, “A Face Recognition System
5. FUTURE SCOPE Based on Local Binary Patterns and Support Vector
In the future a more accurate and exact face recognition Machine for Home Security Service Robot”, 2016 9th
algorithm can be presented. We can control the access to the International Symposium on Computational Intelligence
door using our Smartphone remotely by sending the signals to and Design
the system. We can build an Android application to control [9] LocalBinaryPattern:https://towardsdatascience.com/face-
our system more easily and efficiently. We can even provide recognition-how-lbph-works-90ec258c3d6b
live streaming of the video to the owner on his Smartphone.
[10] Raspberry Pi: www.raspberrypi.org

IJCATM : www.ijcaonline.org 47

View publication stats