A video surveillance systemfor people detection and number estimation

This paper proposes a method to detect people and estimate the number of the people in a supermarket for energy-saving. By using background subtraction technique to extract foreground, and utilizing the foreground to match adjustable standard template, the method detects people and eliminates non-human objects. Simultaneously, the people positions can be marked in the image. The experiment results show that the proposed method has good performance on people detection and counting in the surveillance camera of the supermarket.