Video Background Extraction Based on Improved Mode Algorithm

Background extraction is a fast and efficient moving object segmentation algorithm. This paper presents a new background extraction algorithm based on improved mode algorithm by using frame differential method to obtain the regions of static background. Then, this series of frames containing no moving foreground objects are counted by mode algorithm, thus improving the accuracy of the background extraction. Evaluation of the background extraction is also made by Background Approach Coefficient (BAC) which is presented in this paper. The experimental results show that the proposed method is fast in speed and accurate in background abstract.