A multiclass classifier to detect pedestrians and acquire their moving styles

Authors:
D. Chen;X. B. Cao;H. Qiao;F. Y. Wang
Affiliations:
Department of Computer Science and Technology, University of Science and Technology of China, Hefei, P.R. China;Department of Computer Science and Technology, University of Science and Technology of China, Hefei, P.R. China;Institute of Automation, Chinese Academy of Sciences, Beijing, P.R. China;Institute of Automation, Chinese Academy of Sciences, Beijing, P.R. China
Venue:
ISI'06 Proceedings of the 4th IEEE international conference on Intelligence and Security Informatics
Year:
2006

Citing 2
Cited 0

Making large-scale support vector machine learning practical

Advances in kernel methods
Pedestrian detection and tracking with night vision

IEEE Transactions on Intelligent Transportation Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In a pedestrian detection system, to discover the intention of a pedestrian and warn the driver, it is necessary to obtain the pedestrian’s main moving style. In this paper, an efficient multiclass classifier is presented to detect pedestrians and classify their moving style simultaneously. The multiclass classifier composes of three two-class classifiers and each of them is trained with a SVM algorithm. Experiments based on a single camera pedestrian detection system show that the multiclass classifier has an acceptable detection rate; at the same time, it can judge whether a pedestrian is walking along the road or across the road.