Paper Title
Improving The Performance Of Video Surveillance Using Decision Making Framework

Abstract— In the era of image processing, security through video surveillance plays on important role. The recent trends in video surveillance are focused for machine learning by using artificial intelligence, because the capital assets of country should be protected without huge man power. In the video surveillance scenario, energy constrained sensors, huge amount of video stream, network coverage, decision making are the challenging component for the performance. This paper proposed the frame work with robotic sensor networks to overcome the factors to guard the region under surveillance with limited man power. Nowadays machine learning robots are fine preference for automatic decision making. Robotic Sensors are deployed in the field of surveillance area with video sensing and communication capabilities. Various key challenges are focused in robots during monitoring and some of the sensors has equipped with machine vision processing and directed for granting the command to adjacent sensors to track the particular episode when the moment of out of coverage and occlusion. The processed video stream of each robotic sensor transmitted to neighboring node with the details of tracking and occlusion. We present a decision making framework in robotic sensor network for video surveillance to increase the performance of surveillance system.