Background Update Method Applied in Moving Vehicle Detection
Pan Xiao-lei · Jisuanji gongcheng · 2012
The block-based selective background update method exists weaknesses that it is unnatural between the block and the adjacent block and the anti-interference is weak.Thus,the block-based selective background update method is combined with the filtering method and the gradually correction method in order to refresh the background.The current frame and the previous one are divided into small blocks.According to the mean and variance of the current background,it is established that the foreground blocks,background blocks.The background blocks and the foreground blocks are separately updated.Test results demonstrate that when the method is applied to moving vehicle detection,the accuracy is improved by about nine percent and time efficiency also improved slightly.