Speeding Up Eedge Segment Based Moving Object Detection Using Background Subtraction in Video Surveillance System

Automatic real time video monitoring and object detection is indeed a challenge since there are many criteria that should be taken In mind in designing and implementing algorithms for this sake. The criteria that should be considered for example are processing speed, scene illumination variation and dynamic outdoor environment. In this study we propose a fast, flexible and immune against illumination variation approach for moving object detection based on the combination of edge segment based background modeling and background subtraction techniques. The first technique is used for building robust and flexible statistical background model, while the other technique is used for the prime detection of moving object to be compared later with the flexible background. Thus this combination leads to computational reduction due to the second technique, and then flexible matching and precise detection due to the first technique.