Multi-moving target detection based on the combination of three frame difference algorithm and background difference algorithm

Abstract

Aiming at the problem of ghosting detected by two-frame different method and the defects existed in the independent detection by inter-frame difference method and background difference method, three-frame difference method is proposed in this paper. That is, the different result between frame and frame can be calculated. The outline of the moving target is roughly marked to solve the problem of ghosting. At the same time, we combine the three frame difference method and the background difference method to get the algorithm of this paper. The algorithm uses a mixed Gaussian method to establish a background model and modify the variance update so that the background model fits with the real background and performs morphological processing to extract the moving target. The experimental results show that the combination of the three-frame difference method and the background difference method can effectively remove the noise and ghosting, which avoids inaccurate object extraction, caused by the background difference method and avoids the incomplete moving object of the inter-frame difference method.