Authors: Yan Wang; Ming Yu; Xinliang Tang
Addresses: College of Information Engineering, Tianjin University of Commerce, TianJin 300134, China ' School of Information, Hebei University of Technology, TianJin, China ' Department of Personnel, University of Science and Technology, ShiJia Zhuang, HeBei, China
Abstract: An adaptive licence plate location algorithm is proposed to improve locating accuracy of licence plates in the case of images collected in conditions of complex weather and insufficient light. The algorithm judges the different weather images by colour features and sharpness. The wavelet coefficients are used to adjust the contrast of image. At last, we fuse the vertical protection and template matching algorithms to locate licence plate. Experiments show the algorithm can locate licence plate in varied weathers, such as sunny, rainy, foggy and evening. The average rate of licence plate location can reach 93.4%.
Keywords: colour feature; wavelet coefficients; vertical protection; template matching.
International Journal of Computer Applications in Technology, 2018 Vol.57 No.2, pp.85 - 93
Received: 16 Feb 2017
Accepted: 19 Feb 2017
Published online: 30 Apr 2018 *