车牌字符分割的研究现状和展望.pdfVIP

Computer Science and Application 计算机科学与应用, 2013, 3, 49-53 doi:10.12677/csa.2013.31009 Published Online February 2013 (/journal/csa.html) A Survey of the Research on License Plate Character Segmentation Jianchun Li, Xing Yang, Wei Liu Electronic Engineering Institute, Hefei Email: yangxing.1983@163.com Received: Nov. 12th nd th , 2012; revised: Dec. 2 , 2012; accepted: Dec. 16 , 2012 Abstract: License plate recognition based on machine vision, one of the key technologies of intelligent transportation system, has been paid much attention to. Character segmentation plays an important role in license plate recognition and its aim is to divide the character string in license plate into single characters which are easier to be recognized. Gener- ally, character segmentation is composed of binarization and segmentation. After introducing research achievements of related methods for these two parts, we analyze the merits and drawbacks of each method and look into the develop- ment trends of character segmentation in this paper. Keywords: Intelligent Transportation System; License Plate Recognition; Character Segmentation 车牌字符分割的研究现状及展望 李建春，杨 星，刘 伟 电子工程学院，合肥 Email: yangxing.1983@163.com 收稿日期：2012 年 11 月 12 日；修回日期：2012 年 12 月 2 日；录用日期：2012 年 12 月 16 日 摘 要：作为智能交通系统的核心技术之一，基于机器视觉的车牌识别一直受到广泛的关注。字符分割是车牌 识别的重要步骤，其目的是将车牌字符串分割为更易于识别的单个字符。字符分割一般可分为二值化和分割两 个部分。分别针对这两个部分，本文首先介绍了相关方法的研究成果，然后剖析了各自的优缺点，展望了下一 步的研究方向。 关键词：智能交通系统；车牌识别；字符分割 1. 引言 分割成易于识别的单个字符，大幅削弱噪声对识别过 程的影响。字符分割一般可分为二值