TY - GEN
T1 - Vision-based vehicle event detection through visual rhythm analysis
AU - Yeh, Chia Hung
AU - Bai, Jia Chi
AU - Wang, Sun Chen
AU - Sung, Po Yi
AU - Yeh, Ruey Nan
AU - Shih, Maverick
PY - 2008
Y1 - 2008
N2 - In this paper, a simple and reliable on-road vehicle event detection algorithm is proposed to identify events for vehicle. A virtual line at the same position of a frame is employed to extract visual rhythm. The visual rhythm is a compact representation of a video that captures the temporal information of vehicle status of a coarsely spatially sampled video sequence. By analyzing statistical characteristics of the visual rhythm, the events such as safe distance, passing and lane changing can be effectively detected. The proposed techniques can prevent accidents and improve traffic safety by monitoring the alertness of drivers, augmenting vision fields to prevent collision. The proposed system is efficient both in terms of computational complexity and memory requirements. Experimental results show the efficiency and effectiveness of the proposed system for intelligent transport system.
AB - In this paper, a simple and reliable on-road vehicle event detection algorithm is proposed to identify events for vehicle. A virtual line at the same position of a frame is employed to extract visual rhythm. The visual rhythm is a compact representation of a video that captures the temporal information of vehicle status of a coarsely spatially sampled video sequence. By analyzing statistical characteristics of the visual rhythm, the events such as safe distance, passing and lane changing can be effectively detected. The proposed techniques can prevent accidents and improve traffic safety by monitoring the alertness of drivers, augmenting vision fields to prevent collision. The proposed system is efficient both in terms of computational complexity and memory requirements. Experimental results show the efficiency and effectiveness of the proposed system for intelligent transport system.
KW - Intelligent transport system
KW - Vehicle event detection
KW - Visual rhythm
UR - http://www.scopus.com/inward/record.url?scp=54049112299&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=54049112299&partnerID=8YFLogxK
U2 - 10.1109/ICME.2008.4607433
DO - 10.1109/ICME.2008.4607433
M3 - Conference contribution
AN - SCOPUS:54049112299
SN - 9781424425716
T3 - 2008 IEEE International Conference on Multimedia and Expo, ICME 2008 - Proceedings
SP - 309
EP - 312
BT - 2008 IEEE International Conference on Multimedia and Expo, ICME 2008 - Proceedings
T2 - 2008 IEEE International Conference on Multimedia and Expo, ICME 2008
Y2 - 23 June 2008 through 26 June 2008
ER -