Object Detection and Video Surveillance System

Abstract

Now a days, Video Surveillance systems are quickly becoming a key component of the security system. Making an intelligent video surveillance system, which can able to detect and track multiple objects and also deals with illumination problem and whether conditions(fog, rain, snow etc) is quite challenging task. There are many applications and areas in which Video surveillance is necessary. Video Surveillance in dynamic scenes is active research area in computer vision and there are many applications in which video surveillance is necessary for security, monitoring purpose. From the various application proposed system is designed for real time automatic vehicle detection and vehicle counting from stationary background. Nowadays, we see that the traffic is increases day by day on roads, highways due to increase in number of vehicles. Vehicle detection, classification and counting is very important application by which highway monitoring, traffic planning, analysis of the traffic flow, etc can be easily done. In this thesis, computer vision based moving vehicle detection and counting is presented. By the background subtraction method moving vehicle is detected in ongoing video. And further analysis is done to classify the moving vehicle as car, bike, truck etc. The system is implemented using MATLAB and experimental results are demonstrated on the dataset as well as real time video taken from the static camera. The system is also capable for counting the number of vehicles.