An Improved Sakbot Algorithm Based on Histograms Comparison

The previous work proposes a general purposed moving object detection algorithm based on background suppression called Sakbot. This algorithm is highly reliable and adaptive to various environments. Moreover, it can successfully prevent "dead lock" when dealing with stop and start objects. But it could only put the stopped objects into background group because the algorithm cannot really detect them. In order to solve this problem, an improved algorithm based on histograms comparison rather than optical flow calculation is named as Sakbotum and proposed in this paper. This new Sakbotum algorithm is more adaptive to detect objects with complex motions in a faster way.