DocumentCode :
3528008
Title :
A new vision system for traffic sign recognition
Author :
Gu, Yanlei ; Yendo, Tomohiro ; Tehrani, Mehrdad Panahpour ; Fujii, Toshiaki ; Tanimoto, Masayuki
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Nagoya Univ., Nagoya, Japan
fYear :
2010
fDate :
21-24 June 2010
Firstpage :
7
Lastpage :
12
Abstract :
In this paper, we introduce a new vision system for traffic sign recognition. The new vision system is a hybrid camera system, including a wide angle camera, a narrow angle camera and control part. The traffic sign candidates are detected in the image of the wide angle camera, based on shape and local feature. The control part of the vision system is adjusted to the correct positions, which makes the narrow angle camera of this system to get a high resolution image of each detected candidate. This high resolution image is used for traffic sign classification. In the new vision system, the traffic sign detection and classification are processed in different resolution images separately. This advantage of the system makes traffic sign recognition at long distance possible.
Keywords :
cameras; driver information systems; feature extraction; image recognition; road traffic; shape recognition; driver assistance system; hybrid camera system; image detection; narrow angle camera; traffic sign recognition; vision system; wide angle camera; Cameras; Control systems; Image resolution; Machine vision; Mirrors; Roads; Robustness; Shape; Telecommunication traffic; Vehicles;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Vehicles Symposium (IV), 2010 IEEE
Conference_Location :
San Diego, CA
ISSN :
1931-0587
Print_ISBN :
978-1-4244-7866-8
Type :
conf
DOI :
10.1109/IVS.2010.5548005
Filename :
5548005
Link To Document :
بازگشت