Moving Object Detection Algorithm Based on Variance Analysis

A novel object detection approach based on variance analysis is presented in this paper. Firstly, the variance of the frame difference information is analyzed and the background model is built. Then, the variance is used as threshold to separate the frame difference image into two parts, moving region and background region. Thirdly, moving region is further analyzed by comparing it with background; the moving object and uncovering region are detected. Lastly, the background is updated with uncovering area for detecting the moving object in next video frame. The experiment results show that the proposed method is practical and with fast detection speed, the proposed approach could detect clear motion objects and meet the need of the real time intelligent video surveillance and monitoring system.