An improved motion detection algorithm using ViBe

Detection of moving objects in video sequences is a vital task in many computer vision applications. Many different algorithms have been proposed to detect moving objects in successive frames. In this paper we present a novel algorithm for moving object detection from a video. The improved approach of ViBe follows a new background model employing the frame differencing information which is being used in detection of the foreground objects zone. The new algorithm which is the combination of temporal differencing and background subtraction, can save the processing time and reduce the complexity of ViBe. Simulation results confirm the superiority of proposed algorithm to the original ViBe motion detection algorithm. It can detect moving objects more accurately and efficiently and its output optimization ability is stronger than original ViBe.