Title :
Real-time vision-based preceding vehicle tracking and recognition
Author :
Chang, Hao-Yuan ; Fu, Chih-Ming ; Huang, Chung-Lin
Author_Institution :
Dept. of Electr. Eng., Nat. Tsing Hua Univ., Hsinchu, Taiwan
Abstract :
This paper proposes a preceding vehicle tracking and detection system which includes road area finding, vehicle footprint extraction, and vehicle bounding box extraction. Our method detects the footprint of the vehicle inside the road area to avoid the false detection of delineators. By extracting the preceding vehicle, we effectively track the vehicles, and detect the passing car and distant car. Then, we also propose the distance measure formula. Through camera calibration process and vanishing point detection process, we can calculate the distance to the preceding vehicle accurately.
Keywords :
cameras; computer vision; distance measurement; image classification; object detection; object recognition; real-time systems; road traffic; road vehicles; tracking; traffic engineering computing; camera calibration process; real-time vision-based preceding vehicle; road area finding; vanishing point detection; vehicle bounding box extraction; vehicle detection system; vehicle footprint extraction; vehicle tracking; Cameras; Charge coupled devices; Charge-coupled image sensors; Clustering algorithms; Image edge detection; Land vehicles; Remotely operated vehicles; Road vehicles; Tracking; Vehicle detection;
Conference_Titel :
Intelligent Vehicles Symposium, 2005. Proceedings. IEEE
Print_ISBN :
0-7803-8961-1
DOI :
10.1109/IVS.2005.1505155