Based On The Video Detection And Classification Of Traffic Flow

In recent years, the video-based on traffic parameters detection has been a new research direction in fields of computer vision. The traffic parameters detection is a sub-topic in intelligent transport system, which involves integrated subjects and leading research areas such as image processing, computer vision and artificial intelligence. It not only has a high theoretical study value, but also has a broad market value and application prospects.The background of this paper is a funded project which belongs to Guiyang Science and Technology Bureau. In the case of fixed camera, we built the framework of traffic flow detection system and get the research elements, and design the video traffic parameters detection system.The main research directions in this paper are speed detection, vehicle classification, traffic flow statistics and according to the analysis come out the traffic congestion. In the preparation of this paper, we use the creation methods of background model which base on the gray classification method, the test results show that the method can effectively reduce noise. In the part of speed detection, firstly, we set the video images of the virtual loops and make sure the calculated area, and calculate the moving object’s minimum area bounding rectangle. According to calculate the centroid of minimum area bounding rectangle to come out the vehicle speed and traffic flow statistics. In the part of the vehicle classification, we give a method base on the characteristics of vehicle profile. We used MATLAB to simulation and found our method has a high real-time and good effect, at last we use VC + + and OPENCV to programming the system.