Comparative analysis of motion detection methods or video surveillance systems

Abstract

In this paper, a motion detection module is proposed for real time dynamic video frames by comparing the three major classes of methods for motion detection namely Background Subtraction, Temporal differencing and Optical Flow method .A hierarchical background model is proposed based on segmenting the background images. The region model is extracted from the histogram of a specific region which is similar to the kind of a Gaussian mixture model. The pixel model is described by histograms of oriented gradients of pixels in each region based on the co-occurrence of image variations. Silhouette detection algorithm is proposed. The experimental results are carried out with a video database to demonstrate the effectiveness, which is applied to both static and dynamic scenes by comparing it with some well-known motion detection methods namely Temporal differencing and Optical Flow method and based on the results a motion detection module for dynamic video frames can be developed which is cost effective, shows high rate of accuracy, low rate of complexity, and well adapt to different kinds of shadow distribution