Detection Of Pedestrian Flow Based On Video

Abstract: Nowadays, how to track and account pedestrian flow accurately in visible light is a very active research new area in compute vision system. Pedestrian detection, flow analysis in the MTR, junctions and the supermarket has a wide range of applications demand; Especially in urban traffic environment, the technology is obvious social and economic benefits. The computer vision technology can create a smart management system over the use of computer vision and image processing methods, thereby providing a more advanced and viable its processing programmes.In a large number of domestic and foreign information and research access, the author found that the model based on visual pedestrian flow detection or research in transportation system is blank, only a very small amount of literature on commercial entrances.At pedestrian detection, the distance between pedestrain and camera more closer, colour information more vague, rotary more intense, pedestrian deformation largely, which decided this research work will be most challenging.In this paper, the author advanced the relevant solutions. First, based on the neighbouring column and row,set different thresholds to binary the frame, so we can get the edge of the object. Then , a new head character curve detection and extraction algorithms is presented aiming at solving...