Share Email Print

Optical Engineering

Clustering method for counting passengers getting in a bus with single camera
Author(s): Tao Yang; Yanning Zhang; Dapei Shao; Ying Li
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Automatic counting of passengers is very important for both business and security applications. We present a single-camera-based vision system that is able to count passengers in a highly crowded situation at the entrance of a traffic bus. The unique characteristics of the proposed system include, First, a novel feature-point-tracking- and online clustering-based passenger counting framework, which performs much better than those of background-modeling-and foreground-blob-tracking-based methods. Second, a simple and highly accurate clustering algorithm is developed that projects the high-dimensional feature point trajectories into a 2-D feature space by their appearance and disappearance times and counts the number of people through online clustering. Finally, all test video sequences in the experiment are captured from a real traffic bus in Shanghai, China. The results show that the system can process two 320×240 video sequences at a frame rate of 25 fps simultaneously, and can count passengers reliably in various difficult scenarios with complex interaction and occlusion among people. The method achieves high accuracy rates up to 96.5%.

Paper Details

Date Published: 1 March 2010
PDF: 10 pages
Opt. Eng. 49(3) 037203 doi: 10.1117/1.3374439
Published in: Optical Engineering Volume 49, Issue 3
Show Author Affiliations
Tao Yang, Northwestern Polytechnical Univ. (China)
Yanning Zhang, Northwestern Polytechnical Univ. (China)
Dapei Shao, Northwestern Polytechnical Univ. (China)
Ying Li, Northwestern Polytechnical Univ. (China)

© SPIE. Terms of Use
Back to Top