Moving Object Detection Based on Edged Mixture Gaussian Models

Abstract

Learning background statistics is an essential task for several visual surveillance applications such as incident detection and traffic management. An adaptive foreground object extraction algorithm for real-time video surveillance is presented in this paper. The proposed algorithm improves the classic Gaussian mixture background models (GMM) to remove the undesirable subtraction results due to sudden illumination change. This implementation is achieved by replacing the whole image with edge image to build mixture Gaussian models at every frame. Experimental results show that the proposed algorithm possesses higher performance on real surveillance video under a variety of different environments with lighting variations.