Tracking pedestrians using multiple single‐row laser range scanners and its reliability evaluation
Katsuyuki Nakamura, Hujing Zhao, Ryosuke Shibasaki, Kiyoshi Sakamoto, Tomowo Ohga, 尚毅 鈴川 · Systems and Computers in Japan · 2006
Abstract In this paper the authors propose a method for pedestrian tracking using multiple laser range scanners and describe verification experiments in a railway station. The proposed method involves synchronizing multiple laser scanners using a network and tracking pedestrians based on range data for foot cross‐sections. The tracking algorithm consists of the following functions: detection of foot candidates using range data clustering; detection of pedestrian candidates using grouping of foot candidates; detection of movement vectors for pedestrian candidates; extended processing of existing tracings using an extended Kalman filter based on a walking model. The results of using the proposed method in a station concourse in Tokyo Station found that at a maximum 150 people could be tracked at the same time. The tracking precision exceeded 80% during commuter rush hour. Applications to high‐precision crowd observations in wide areas can be expected. © 2006 Wiley Periodicals, Inc. Syst Comp Jpn, 37(7): 1–11, 2006; Published online in Wiley InterScience ( www.interscience. wiley.com ). DOI 10.1002/scj.20578