Title :
A vehicle license plate recognition system based on fixed color collocation
Author :
Yang, Yao-Quan ; Bai, Jie ; Tian, Rui-Li ; Liu, Na
Author_Institution :
Fac. of control Sci. & Eng., North China Electr. Power Univ., Baoding, China
Abstract :
Locating the region of license plate is the key of the vehicle plate recognition system. A new method is adopted in this paper instead of the traditional method based on gray image. The method that sufficiently utilizes the color characteristics of the colored image is based on the color collocation of the plate´s background and characters combined with the plate´s structure and texture to locate the vehicle license plate. Then the plate´s region would be binarized, segmented and recognized. The recognition rate reaches 95% by experiments.
Keywords :
character recognition; edge detection; image colour analysis; image segmentation; image texture; vehicles; colored image; fixed color collocation; fixed color pair; image binarization; image recognition; image segmentation; image texture; license plate region location; plate characters; plate structure; vehicle license plate recognition system; Automotive engineering; Color; Control systems; Image edge detection; Image processing; Image recognition; Image segmentation; Licenses; Pixel; Vehicles; Vehicle license plate location; binarize; fixed color pair; texture;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527897