Research on Stationary Object Detection Technique Based on Dual-Background

Abstract

In this paper, a novel approach for detecting and analyzing stationary objects driven visual events in video surveillance systems is proposed. In recent years, video surveillance systems have become an extremely active research area due to a sharp increasing in the levels of terrorist attacks on crowded public places and left behind object detection is indispensable in public places such as airport security system. A new stationary object detection algorithm is proposed in this paper to save this problem. Firstly, this method uses the Dual-Background subtraction to get the foreground image based on the approximated median filtering using the adaptive threshold method. Secondly, after detecting the stationary object, through morphology processing, the algorithm can accurately detect the stationary object, such as box, bag and so forth. Experimental results show that the proposed algorithm is robust and precise.