Block Based Skin Color Detection for Automated Video Surveillance System

Automated video surveillance systems are an important component for Infocom Technology (IT) Industry. These systems play an important role in a wide range of image processing applications ranging from face detection, face tracking, gesture analysis, and to various human computer interaction domains. Skin color detection provides a useful and robust cue to all these applications. Therefore, skin color detection is one of the important research areas for automated video surveillance systems. Real-time requirements are the major issue for these systems. In this paper, a robust and efficient block based skin color detection technique is proposed. It is capable of detecting skin color regions in an image in real-time and a performance improvement of 3.58 times is achieved without loss of accuracy in results. The presented algorithm is implemented in C language on a standard WindowXP machine using Dev-C++ (Version 4.9.9.2) compiler. For reading and displaying of image/video Intel OpenCV library functions are used. The algorithm yield good results. The experimental results have established the effectiveness of the approach.