International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 05 | May 2023
p-ISSN: 2395-0072
www.irjet.net
AUTOMATIC DOOR UNLOCK WITH FACIAL RECOGNITION USING IoT Birali Prasanthi1, Kakumani Sravya2, Komirishetty Saiteja3 1Senior Assistant Professor, Dept. of Computer Science and Engineering, Mahatma Gandhi Institute of
Technology, Hyderabad, Telangana, India 2Student, Bachelors in Computer Science and Engineering, Mahatma Gandhi Institute of Technology, Hyderabad, Telangana, India 3Student, Bachelors in Computer Science and Engineering, Mahatma Gandhi Institute of Technology, Hyderabad, Telangana, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Security is the primary problem that must be
In this system an interface is created to collect images of registered users comfortably. Here, the faces in dataset are first detected using hog model and these detected faces are encoded. Haar cascade classifier is used for real-time face detection for more accuracy.
dealt with in the current society. With the recent progress in emerging technologies, IoT stands out as a cutting-edge technology capable of resolving a wide range of securityrelated issues. This paper deals with the proposed system for a sophisticated and knowledgeable door that unlocks automatically when a registered face is recognized. The faces in the dataset are first detected using the hog model and then the detected faces are encoded using OpenCV libraries. Whereas, the Haar Cascade classifier is used for real-time facial detection. The objective of this project is to replace traditional passwords and patterns and add more security to the lives of people. This process is carried out by the raspberry pi3B because it has low processing time.
1.2 Hardware requirements Raspberry Pi 3 B: It is a third-generation Raspberry Pi. This powerful credit-card-sized single board computer outperforms the previous models. Its processor is ten times faster than that of the original Raspberry Pi. It also supports wireless LAN and Bluetooth connectivity. Raspberry Pi camera module v1.3: It receives software and hardware support from the developers as it is an official product of the foundation. It has a 5-megapixel OV5647 sensor. It supports video modes of 1080p30, 720p60 and 640X480p60/90 video record. It can be connected to the Raspberry Pi at the CSI port using a flex cable. It provides 2592 X 1944 still picture resolution. It is compatible with both model A and model B and weighs about 3 grams.
Key Words: OpenCv, Haar Cascade classifier, hog model, Raspberry pi 3B. 1. INTRODUCTION Faster and accurate identification and authentication methods are necessary in this modern era. Facial recognition is one such method, but the problem is that the software should be able to able to distinguish spoofing and should be highly accurate. It is thus necessary to use appropriate algorithm. The system collects the images of registered users in the database. The faces in these images are detected and then encoded. The encodings of detected faces during the livestream are then compared with the encodings in the database. If they match then the door will be unlocked. To achieve this Raspberry pi, solenoid lock, relay module and raspberry pi camera are used .Facial recognition is preferred over other biometrics because they can be taken without owner’s consent and thus useful for various security related applications. Furthermore, images can also be taken from distance which is unique compared to other approaches.
HC-05 - Bluetooth Module: It is a popular Bluetooth module that can add full-duplex wireless functionality. It has a range of up to <100m which depends upon transmitter and receiver, atmosphere and geographic conditions. Serial communication is used between two devices. Relay module: A relay is electromechanical device that uses an electric current to open or close switch contacts. Solenoid lock: Solenoids are primarily electromagnets made up of an enormous coil of copper wire with a metal in the centre. When the current flows through the coil, the slug is drawn into the coil's centre. This allows the solenoid to pull from only one end. It's basically an electronic lock, used for a basic cabinet, safe or door.
1.1 Proposed System
1.3 Software requirements
The proposed system unlocks the door if a registered face is recognized and sends a telegram message with the picture of the unknown face and an OTP. The intruder can unlock the door if and only if he enters correct OTP through Bluetooth.
© 2023, IRJET
|
Impact Factor value: 8.226
Raspberry Pi OS: It is previously known as Raspbian and is officially supported OS. It’s based on Debian and is built on
|
ISO 9001:2008 Certified Journal
|
Page 1342