Research on pedestrian detection algorithms based on video
Jinhao Deng, Juan Zhu · 2010
On the basic of the video traffic surveillance system, the pedestrian detection is a crucial part. A study on vehicle and pedestrian recognition based on back propagation (BP) neural network is presented in this paper, and the author puts forward an effective algorithm for pedestrian detection. First, extract the moving objects from the image sequence using background subtraction. Second, select part of the objects for further detected. Most of the moving objects, which can be determined as vehicles would be excluded in the pretreatment. Third, extract several significant eigenvalues from the rest objects, which can indicate the differences of contour between pedestrian and vehicle. Finally, eigenvector is formed and used as the input of the back propagation neural network, the output of which is the detecting result. The BP neural network is trained and used to identify pedestrian, experiment results of which are analyzed. It has been proved that the algorithms proposed in this paper have satisfactory real-time performance and accuracy.