Preceding Vehicle Visual Detection and Tracking Based on Template and Shadow Feature

Article Preview

Abstract:

As to the vehicle visual detection and tracking problem in IDASWs (Intelligent Driver-Assistance and Safety Warning System), this paper proposed a way of a good effect. Firstly use vehicle’s bottom shadow feature and identified lane result to determine the searching region. Then use template matching algorithm based on vehicle’s tail shape to realize detecting and tracking. Experimental results show that the algorithm can effectively detect and track the preceding vehicle from the video sequence, reaching robust and real-time requirements

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 490-495)

Pages:

481-485

Citation:

Online since:

March 2012

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2012 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] Yuan WU: The Detection and Tracking of Moving Vehicle, Nanjing University of Aeronautics and Astronautics. (2007).

Google Scholar

[2] Xiaohui ZHANG, Ding LIU: Research on Vision Recognition and Tracking for Corridor of Autonomous Driving Robot, Journal of Xi'an University of Technology, 22(2): 158~162. (2006).

Google Scholar

[3] Prati, A.: Detecting moving shadows: algorithms and evaluation, IEEE Trans. on Pattern Analysis and Machine Intelligence, 25(7): 918-923. (2003).

DOI: 10.1109/tpami.2003.1206520

Google Scholar

[4] Yujin ZHANG: Image ENGineering, Tsinghua University Press, (2008).

Google Scholar

[5] Youchun XU, Rongben WANG and Keqian LI: A Linear Model Based Road Identification Algorithm, Journal of Image and Graphics, 9(7): 858~864. (2004).

Google Scholar

[6] Paragios N, Deriche R: Active contours and level sets for the detection and tracking of moving objects, IEEE Trans Pattern Anlysis and Machine Intelligence, 22(3): 266-280. (2000).

DOI: 10.1109/34.841758

Google Scholar