基于字符边缘颜色多分量信息的车牌定位方法  

Location Method for Vehicle License Plate Based on Multiple Components' Information of Character Edge Color

作　　者：李刚[1,2] 黄席樾[1] 刘政[2] 周勇 朱彬 白正文 

机构地区：[1]重庆大学自动化学院,重庆400044 [2]重庆理工大学计算机学院,重庆400050 [3]重庆水轮机厂,重庆400054

出　　处：《计算机工程》2009年第18期176-178,共3页Computer Engineering

基　　金：重庆市教委科技计划基金资助项目"公路交通安全与重大事故救援关键技术研究"(KJ090619);重庆市九龙坡区科技计划基金资助项目"汽车牌照自动定位方法的研究与实现"(2007Q76)

摘　　要：针对复杂环境下的车牌定位,提出利用车牌字符边缘颜色多分量信息特征的定位算法,分别提取车牌底色像素点和字符色像素点,排除没有字符色像素点相邻的底色纹理像素点。由于字符色与底色的相互影响,字符边缘处色度、饱和度分量都有较窄的范围,可缩小车牌的搜索区域。结果证明该特征有助于车牌的准确定位,可确定车牌颜色,同时运算多在二值图上进行,速度快。Aiming at the location of vehicle license plate in complex circumstance, this paper presents a plate location method which utilizes the features of multiple components＇ information of plate character edge color. The plate background color pixels and the character color pixels are extracted respectively, those background color texture pixels which do not neighbor the character color ones are discarded. Because the color of plate character and the color of plate background are influenced by each other, both components of the hue and saturation of character edge color have a relative narrow range. It can reduce the plate search area. It is obviously that this trait is helpful to locate plate accurately and it can ascertain plate color at the same time. The algorithm is carried through binary image mainly, so higher location speed can be easily acquired.

关 键 词：车牌定位 字符边缘颜色 颜色多分量信息 纹理 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]