一种基于快速最近特征线的汽车牌照识别方法  

A Method of License Character Recognition Based on Fast Nearest Feature Line

作　　者：谢建平[1] 

机构地区：[1]丽水学院计算机与信息工程学院,浙江丽水323000

出　　处：《电视技术》2006年第4期93-96,共4页Video Engineering

摘　　要：提出一种基于快速最近特征线法(NFL)的车牌识别方法。根据颜色的视觉一致性,采用Munsell颜色空间的NBS颜色距离的概念对色彩进行聚类,再采用不同的结构元素,对于聚类后的图像进行一系列的数学形态学运算,准确定位出车牌的位置后,以NFL为基础,采用一种快速的计算方法,把待识别字符划分到最相匹配的类别中。实验表明,提出的车牌分割与识别新方法快速,准确,能有效地提高汽车牌照的识别效率。A novel license character recognition method based on the nearest feature line （NFL） classifier is presented in this paper. Based on ocular uniformity,we process our color clustering in Munsell color space using the concepts of NBS distance. Then,with different structure elements,we take a series of operations to the image that was processed by color clustering method before, after that the location of the plate is found accurately, we propose a fast algorithm based on the NFL method to improve the performance of the recognition system by using a lot of virtual data generated by the feature line. Experimental results show that the method is effective and gratifying.

关 键 词：车牌识别 颜色聚类 数学形态学 最近特征线 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]