Research On Vehicle Anti-theft and Alarm System Using Facing Recognition

In this paper, a vehicle anti-theft system has been developed. When the car is start, this system begins to work, it captures the driver's video images using an active IR illuminator, then locates the driver's eyes and recognizes the driver's face using PCA algorithm. If the current driver is found to be unauthorized by compared his face image with the face images prestored, this system will alarm and send the unauthorized driver's image to car owner or police through CDMA or GPRS networks. This system was tested in a simulating environment with subjects of different ages, with/without glasses, and under different mumination conditions, and it was found very robust, reliable.