In this paper, a modular, configurable and versatile hardware platform for real-time video and image processing is presented. The hardware platform is based on the Altera DE2 development board which is completed with a Camera interface for video acquisition and a VGA interface for image restitution. A facial recognition system is a computer application for automatically identifying or verifying a person from a digital image or a video frame from a video source. A number of defense, security and commercial applications demand real time face recognition systems, especially when other biometric techniques are not feasible. Using it in our project fulfils the need of car security so as to prevent car thefts which is easier in case of car locking systems. An FPGA (Field Programmable Gated Array) based novel design has been developed which provides real time face tracking of the respective person and if found the respective persons open the lock of the car. The whole system is developed on Altera DE2 board using Verilog HDL (Hardware Description Language) as a coding language. We have not used any of the basic algorithms for the project; we developed our own algorithm by using DE2_TV core provided along with the Altera DE2 board. In our project a fixed background of black color is used and when the pixels of background are distorted by incoming person the corresponding pixels are saved in the SRAM of the board and then later retrieved for face detection and recognition. If the distorting pixels are same with the already saved then the lock of the car opens.
展开▼
