A Brief Review of Methods of Moving Object Detection, Challenges and Shortcomings

Abstract

Moving object detection from dynamic scenes has been used in many computer vision applications like face detection, video processing, video surveillance, traffic monitoring etc. Finding the position is much more challenging task than detecting the moving object in a video. Here we are presenting a brief review of various algorithms for moving object detection which are already available. Moving object detection from dynamic scenes using Multiple Color Space Histogram Model is briefly discussed in this paper. In this model, at first, convert each frame from RGB space to other color spaces and calculate the histograms of selected color components, then we can obtain the background histogram model; then, detect the objects using statistical histogram superposition principle; at last, update MCSHM by the result of detection. The experimental results demonstrate that our method can quickly and accurately detect moving objects in dynamic scenes.