Video surveillance based on energy feature
Zongbo Hao, Nan Sang, Li Chang-Lin, Xin Xu · 2010
Video surveillance is proposed in the paper. The three key features of abnormal behavior are represented by energy. In which the velocity of the moving target is reflected by optical flow, and the disorder feature of the motion is expressed by entropy. For the last feature, energy is used to express the relative positions of the moving targets. The experiments show the good effects of the method in fighting surveillance.