Simple and Model-Free Algorithm for Real-Time Pedestrian Detection and Tracking
Yegor Malinovskiy, Jianyang Zheng, Yinhai Wang · Transportation Research Board 86th Annual MeetingTransportation Research Board · 2007
Pedestrian monitoring is quickly becoming an avid area of interest as information regarding pedestrian flow is needed not only for developing competent access in particular urban corridors, but also system optimization scenarios, such as transit systems and intersection control optimization. In this paper, we present a simple, yet effective method for tracking pedestrian objects in a relatively large surveillance area, using ordinary CCTV cameras. Object extraction is accomplished via background subtraction, while tracking is accomplished through an inherent characteristic cost function. Composite objects are used as a means to deal with occlusions. The algorithm is implemented in a Pedestrian Tracking (PedTrack) system using Microsoft Visual C#. Tested with both synthetic and real video data, this system was proved to be reliable and about 80% pedestrian were assigned with correct IDs.