The Research Of Pedestrian Flow In Public Transport System Based On Image Processing

With the rapid development of computer technology, the computer vision technology is gradually mature, and its applications are increasingly widespread. In this paper, computer vision technology is applied on Intelligent Transportation System, providing real-time data base for intelligent bus dispatch system. That is to provide the number of on-off passengers at each stop, thus reasonable arrangement for departure interval and efficient use of existing resources. In this context, the application of several algorithms in image processing is studied.The thesis firstly introduces the image preprocessing technology and several moving object detection methods, and use MATLAB software for image processing. Then, because the commanding view of passenger’s head is like a circle, so I use the Hough transform to detect circles. Combined with the practical application environment, in order to improve the detection efficiency and reduce the amount of computation, the random Hough transform algorithm is proposed to recognize the passenger’s head. Finally, using Mean-Shift algorithm to track passengers, taking into account the inevitable occlusion problem between the passengers, an algorithm that combines the Mean-Shift algorithm and least square prediction algorithm for passengers to track is proposed. And using MATLAB software to process the image sequence, which can effectively continuous target tracking, and mark the trajectory of the passengers.