基于改进连通域算法的车牌字符分割方法  

License Plate Character Segmentation Method Based on Improved Connected Domain Algorithm

作　　者：朱亚萍[1] 邱锦山 杨成忠[1] 

机构地区：[1]杭州电子科技大学浙江省物联感知与信息融合技术重点实验室,浙江杭州310018

出　　处：《杭州电子科技大学学报（自然科学版）》2016年第2期48-51,共4页Journal of Hangzhou Dianzi University：Natural Sciences

基　　金：国家自然科学基金资助项目(61427808)

摘　　要：为了解决车牌字符分割中不连通汉字分割、粘连字符分割以及竖直边框干扰等问题,提出了一种基于改进连通域算法的车牌字符分割方法.采用改进的连通域算法并结合车牌尺寸的先验知识进行车牌字符的粗分割,运用二分投影法实现了车牌字符的细分割,有效地结合了改进的连通域算法与传统的二分投影法.实验结果证明,改进后的算法能较好地解决上述在字符分割中存在的问题,具有较高的实用性.In order to solve the problems of the disconnected character segmentation,adhesion character segmentation and vertical border interference in license plate character segmentation,a kind of license plate character segmentation method based on the improved connected domain algorithm is proposed in this paper.Coarse-grained segmentation is conducted on license plate characters by combining the improved connected domain algorithm and prior knowledge of license plate size.Besides,the fine-grained segmentation of license plate characters is achieved with the help of binary projection method and the method effectively combines the improved connected domain algorithm and traditional binary projection method.The experimental results show that the improved algorithm can solve the above problems in character segmentation more effectively and it has higher practicability.

关 键 词：车牌 字符分割 连通域 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]