Dynamic contour matching approach in vision-based passenger flow detection
Liu Ji-lin · Journal of Zhejiang University(Engineering Science) · 2008
A method based on contour matching to extract the head disparity was presented,in order to improve the accuracy rate of vision-based passenger flow estimation system by using the depth information.The perspective constraint was applied to reduce the search space,and the head's contour similarity measure was employed to find the optimal matching,and then the head disparity was obtained in the optimal matching path which was searched by using the dynamic programming approach in the multistage graph.The method is very suitable to be applied to the real-time embedded system because it can extract the heads' contours in stereo pairs in parallel and avoid the complex dense disparity computation.The experimental results show that the accuracy rate achieves a great improvement when applying the method to the vision-based passenger flow estimation system.