People tracking in a building using color histogram classifiers and Gaussian weighted individual separation approaches

Authors:
Che-Hung Lin;Sheng-Luen Chung;Jing-Ming Guo
Affiliations:
Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan;Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan;Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan
Venue:
MMM'11 Proceedings of the 17th international conference on Advances in multimedia modeling - Volume Part II
Year:
2011

Citing 4
Cited 0

Artificial Intelligence: A Guide to Intelligent Systems

Artificial Intelligence: A Guide to Intelligent Systems
Multi-Camera Multi-Person Tracking for EasyLiving

VS '00 Proceedings of the Third IEEE International Workshop on Visual Surveillance (VS'2000)
Color-Based Video Stabilization for Real-Time On-Board Object Detection on High-Speed Trains

AVSS '03 Proceedings of the IEEE Conference on Advanced Video and Signal Based Surveillance
Real-time people localization and tracking through fixed stereo vision

Applied Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper focuses on people tracking within a building with the installment of distributed overhead cameras. Our primary concern is to keep tracks of: number of people entrance to a particular area; and the whereabouts trajectory of a particular person within the monitored building. With image taken from ceiling mounted camera, pedestrian's physiologic contour is analyzed from four different viewing angles to form a person's identity signature. In doing so, techniques to locate a person's head, to predict his/her movement direction, to separate overlapped physiological blobs, and to differentiate different people by color histogram classifier have been proposed. Special attention has been paid for system configurability such that the proposed software architecture can be deployed to different floor plans. We have conducted a continuing surveillance monitoring on the third floor of EE in NTUST, and the result shows moderate surveillance performance: 93% accuracy in entrance counting, and 76% accuracy in identification checking.