VEHICLE DETECTION BY USING REAR PARTS AND TRACKING SYSTEM

Abstract

Vision of Indian government; of making 100 smart cities, attracts our attention to intelligent transport system. Traffic flow analysis is a part of intelligent transport system. It mainly contains three parts: vehicle detection, classification and vehicle tracking par t. Recently, there are different detection and tracking methods like computer vision based, magnetic frequency wave based etc. With the rapid development of computer vision techniques, visual detection has become increasingly popular in the transportation field. In urban traffic video monitoring systems, traffic congestion is a common scene that causes vehicle occlusion and is a challenge for current vehicle detection methods.In practical traffic scenarios, occlusion between vehicles often occurs; therefore, it is unreasonable to treat the vehicle as a whole. To overcome this problem we can use part based detection model. In our system the vehicle is treated as an object composed of multiple salient parts, including the license plate and rear lamps. These parts are localized using their distinctive color, texture, and region feature. Furthermore, the detected parts are treated as graph nodes to construct a probabilistic graph using a Markov random field model. After that, the marginal posterior of each part is inferred using loopy belief propagation to get final vehicle detection. Finally, the vehicles’ trajectories are estimated using a Kalman filter and a tracking-based detection technique is realized. This method we can use in daytime as well as night time and in any bad weather condition.