Cooperative Neural Network Background Model for Multi-Modal Video Surveillance

This paper proposed a new cooperative background model for multi-modal video surveillance based on probability neural network (PNN). Firstly, probability of being foreground was estimated in visible and infrared channel, and post processed separately. Then, every pixel was classified into foreground, background, and change pixels by fusing this information, and foreground pixels were segmented into motion regions. Thirdly, adaptive learning rate was computed for every frame and every pixel based on frame motion difference and pixel classification result, and background model for every channel was updated. Experimental results on well-known benchmark image sequences show that the proposed algorithm can detect motion region more precisely.