The Statistics Of Multiple Lanes Traffic Flow

With the rapid development of Intelligent Transportation, video-based vehicle detection technology becomes the focus of research in the world. Video-base vehicle detection technology has been developing on a high-speed track due to its advantages such as large detection range and provision of comprehensive traffic information.Based on research into the existing video-based vehicle detection algorithm and experience as well as drawbacks of the previous studies, this paper proposes a video-vehicle detection algorithm based on energy. The algorithm mainly adopts some common video detection methods, like virtual detection line, the frame subtracted, background updating approach etc, to extract the energy information of detection area as the vehicle's test features. Intelligent Transportation System requires real-time detection. In order to reduce the calculation of the detection process and improve the speed of the system, many algorithms set detection area. Only the part within the area is processed in the image. This will not only ensure the accuracy of detection but also increase the processing speed. Frame difference method, which is simple and practical, is widely used in the current algorithms. There are two forms of the method, the between-frame difference method and the background difference method.In this paper the background difference method is used, the background frame update rate and accuracy of which are the basis of algorithm accuracy.Video-based vehicle detection algorithm extracts the energy width of the vehicle as the detection feature, for the energy information of vehicle can inhibit the influence of the vehicle shadow effectively. In the energy detection algorithm of vehicle, the video captured by camera is first pre-processed through image graying and noise removal. Then the extreme images and contrast images can be obtained, which are basic to calculation of energy within the detection area. In order to get the energy width in detection area of the image, energy width of current frame subtracts the energy width of background frame. Energy width is the ultimate extracted parameter of detection in this algorithm, based on which the algorithm determines whether there are cars passing the detection area and then completes the traffic flow statistics.