Title :
Outdoor vehicle and shadow segmentation by temporal edge density information of adjacent frames
Author :
Techawatcharapaikul, Chanchai ; Kaewtrakulpong, Pakorn ; Siddhichai, Supakorn
Author_Institution :
Dept. of Comput. Eng., King Mongkut´´s Univ. of Technol. Thonburi, Bangkok
Abstract :
The problem of vehicle detection and segmentation in outdoor scenes is tackled. Vehicle shadows pose problems in vehicle segmentation step. In this paper, it is proposed to employ temporal edge density information as prior knowledge to distinguish moving vehicle from moving shadow and background in a traffic surveillance system. Experimental results showed good moving vehicle segmentation performance. The proposed approach is currently limited to free-flowing traffic scenes.
Keywords :
automated highways; image motion analysis; image segmentation; object detection; video surveillance; free-flowing traffic scenes; outdoor vehicle; shadow segmentation; temporal edge density information; traffic surveillance system; vehicle detection; vehicle segmentation step; Automotive engineering; Cameras; Image segmentation; Layout; Lighting; Robustness; Sensor phenomena and characterization; Surveillance; Traffic control; Vehicle detection;
Conference_Titel :
Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, 2008. ECTI-CON 2008. 5th International Conference on
Conference_Location :
Krabi
Print_ISBN :
978-1-4244-2101-5
Electronic_ISBN :
978-1-4244-2102-2
DOI :
10.1109/ECTICON.2008.4600463
