Abandoned Object Detection Based on Tachograph Videos

In the traditional abandoned object detection(AOD for short) researches, the monitoring equipment is stationary which is limited in practice. In this paper, a novel abandoned object detection based on tachograph is proposed. Firstly, we use the Harris-SIFT features and particle filter to achieve object tracking. After that, the VIBE algorithm is applied to detect the abandoned object with background modeling frame by frame, using both background subtraction method and frame difference method. Experimental results show that the proposed algorithm can detect the abandoned object accurately and robustly in both ideal and practical condition.