A car detection system based on hierarchical visual features

Fok Hing Chi Tivive, Bouzerdoum Abdesselam

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, we address the problem of detecting and localizing cars in still images. The proposed car detection system is based on a hierarchical feature detector in which the processing units are shunting inhibitory neurons. To reduce the training time and complexity of the network, the shunting inhibitory neurons in the first layer are implemented as directional nonlinear filters, whereas the neurons in the second layer have trainable parameters. A multi-resolution processing scheme is implemented so as to detect cars of different sizes, and to reduce the number of false positives during the detection stage, an adaptive thres holding strategy is developed. Tested on the UIUC car database, the proposed method achieves better classification results than some of the existing car detection approaches.

Original languageEnglish
Title of host publication2009 IEEE Symposium Computational Intelligence for Multimedia Signal and Vision Processing, CIMSVP 2009 - Proceedings
PublisherIEEE Computer Society
Pages35-40
Number of pages6
ISBN (Print)9781424427710
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event2009 IEEE Symposium Computational Intelligence for Multimedia Signal and Vision Processing, CIMSVP 2009 - Nashville, TN, United States
Duration: 30 Mar 20092 Apr 2009

Publication series

Name2009 IEEE Symposium Computational Intelligence for Multimedia Signal and Vision Processing, CIMSVP 2009 - Proceedings

Conference

Conference2009 IEEE Symposium Computational Intelligence for Multimedia Signal and Vision Processing, CIMSVP 2009
Country/TerritoryUnited States
CityNashville, TN
Period30/03/092/04/09

Fingerprint

Dive into the research topics of 'A car detection system based on hierarchical visual features'. Together they form a unique fingerprint.

Cite this