基于重建的车牌汉字识别研究  

Research on the Recognition of License Plate Chinese Characters Based on Reconstruction

作　　者：刘龙凤 周欣[1] LIU Long-feng;ZHOU Xin(College of Computer Science, Sichuan University, Chengdu 610065)

机构地区：[1]四川大学计算机学院,成都610065

出　　处：《现代计算机》2019年第10期71-74,共4页Modern Computer

摘　　要：随着社会经济的发展,汽车成为家家户户的常用代步工具。汽车数量的增加,也给公共交通的管理带来一定压力,在传统的人力监督管理下,已远远不能满足现下的需求。于是,智能交通的建设便引起了人们的重视,而车牌识别是其中的重要组成部分。前人对汽车车牌的识别做大量研究,其中对于车牌中数字和字母的识别实现较高的精确识别度,但是,在对车牌中的汉字识别部分精度仍有待提高。针对现有技术中存在的对字符图像形变敏感、特征敏感、特征选取敏感以及计算复杂度高等缺陷,提供一种新的计算简单、鲁棒性高、基于主成分特征重建图像的车牌汉字识别方法。With the development of social economy, cars have become a common transportation tool for every family. The increasing cars make the management of public transportation harder. Under the traditional human supervision and management, it is far from enough to meet the current demand. Therefore, the construction of intelligent transportation has attracted people's attention, and the license plate recognition is an important part of it. Predecessors have done a lot of research on the recognition of vehicle license plate, among which, the recognition of numbers and letters in the license plate has achieved a high degree of accuracy. However, the accuracy of Chinese character recognition in the license plate still needs to be improved. In view of the defects of existing technologies, such as the sensitivity to character image deformation, feature sensitivity, feature selection sensitivity and high computational complexity, proposes a new license plate Chinese character recognition method based on image reconstruction by principal component characteristics, which is simple in calculation and highly robust.

关 键 词：车牌汉字识别 主成分特征 重建图像 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] U495[自动化与计算机技术—计算机科学与技术]