A Bus Passenger Flow Estimation Method Based on Feature Points Trajectory Clustering
Based on the observation that motion of different pixels from the same target has very similar spatialtemporal properties in bus video surveillance images, a feature points trajectory clustering method is proposed to estimate passenger flow in this paper. Firstly, the pyramid-based optical flow algorithm is utilized to tracking the feature points movement in the images; then, their trajectories are preclassified into passenger getting on, off the bus and others according their motion direction histogram; finally, the pre-classified trajectories are clustered by their spatial-temporal similarity and the cluster number is looked as the result of bus passenger flow estimation. Since it neednt to detect the head contour, face or other features of the passenger, our method is simple, fast and strong. The experiment results on multiple real bus surveillance videos show that it has high counting accuracy (>90％) in different illumination, background and even crowded conditions.
bus passenger flow estimation trajectory clustering
Yuan Hejin
Department of Computer North China Electric Power University Baoding, China
国际会议
厦门
英文
426-430
2010-10-29（万方平台首次上网日期，不代表论文的发表时间）