ABSTRACT
This project is about an anti-theft security system using face recognition based on
Raspberry pi. Anti-theft security system has been launched for many years but most of
them are just a CCTV, IP camera or door sensor alert system. It could be more efficient
with uses of face recognition. The design of anti-theft security system is based on human
face recognition and remote monitoring technology. It will verify the person identity that
goes near to the camera within certain distance. Only the people who match the identity
with database and key in the correct password have the right to entry. A mobile
application telegram and image processing technique LBP based have been involved in
this system. Electric door lock solenoid, keypad and LCD display will combined operate
the accessibility of door. An ultrasonic sensor is used to detect the person distance
between the door. Face recognition will only start to function when someone goes near
the door within certain distance. An alert message and photo taken will send to owner
mobile phone via telegram with WIFI when the face recognition algorithms detect a
stranger comes near the door and also a buzzer will be trigger to act as an alarm system.
Owner can control the door locker and check the camera with the telegram application.
The security system is designed for actual use in home condition with the uses of a
microcontroller ARM Quad Cortex-A53.
