基于底色分辨和分段属性的车牌字符分割  

License Plate Character Segmentation Based on Background Color Distinguishing and Segment Attributes

作　　者：陈箫枫[1] 潘保昌[1] 郑胜林[1] 赵全友[1] 

机构地区：[1]广东工业大学信息工程学院,广东广州510006

出　　处：《电视技术》2008年第z1期137-139,共3页Video Engineering

基　　金：国家自然科学基金项目(60572078)

摘　　要：根据车牌有多种颜色不利分割字符的问题,提出一种分辨底色并用分段属性来进行字符分割的算法。算法中,采用直方图平滑曲线的特征来分辨白底黑字和黑底白字的车牌区域图像,统一为白底黑字,根据二值图像纵向投影平滑曲线,求其分段属性,分析中心线距离、间隙、宽度等,来分割出车牌的字符。将该算法应用到车牌识别系统中,实验效果比较好。According to the problem of character segmentation disadvantage about multicolor of license plate, an algorithm of character segmentation with background color distinguishing and segment attributes are put forward. In the algorithm, histogram smooth curve features are used to distinguish license plate regional images between white background and black background, then the image is unified to white background and black characters. The segment attributes are calculated with the image longitudinal projection smooth curve, and the midline distance, space, width, and so on are analyzed.Then the characters of license plate are segmented. The algorithm is applied in license plate recognition system, and the experimental results are quite perfect.

关 键 词：车牌字符分割 底色分辨 分段属性 直方图 投影曲线 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]