Hardware accelerator for skin color detection technique

A very important research area for automated video surveillance systems is skin color detection. Major issues for these systems are real time requirements. In this paper, a block based skin color detection technique is proposed. Results prove that it is a robust and efficient technique. Skin color regions in an image can be detected with the proposed technique. A performance gain of 3.58 times is achieved in results. The algorithms is implemented in C language on a WindowXP machine usind Dev-C++ compiler (Version 4.9.9.2). The experimental results have established the effectiveness and accuracy of the approach. The proposed algorithm is then implemented on Virtex-4 FPGA platform.