基于Rough集理论的车牌汉字特征提取  

Car Plate Chinese Character Feature Extraction Based on Rough Set Theory

作　　者：王希雷[1] 

机构地区：[1]天津科技大学计算机学院,天津300222

出　　处：《计算机技术与发展》2007年第6期26-28,227,共4页Computer Technology and Development

基　　金：天津科技大学科学研究基金项目(20050226);天津科技发展计划项目(0431095R)

摘　　要：车牌首位汉字特征提取和识别是一个难点。传统的车牌汉字的特征提取方法是在具有先验知识的情况下进行的,先验知识的好坏对结果有着非常重要的影响。Rough集,理论上可以从数据集中直接提取特征,不依靠先验知识。先用Rough集理论提取待识别汉字的特征,再用这些特征进行模板匹配。实验结果表明该方法有比较好的识别效果。In car plate recognition system,one of the most difficult problems is the feature extraction and recognition of car plate Chinese character. Presently traditional method of car plate Chinese character feature extraction is based on prior knowledge which is important for the result. Rough set theory can get features from data sets without prior knowledge in theory. In this paper, extract the features of Chinese characters based on rough sets, finally the Chinese characters are recognized by using matching recognition. Compared with some other feature extraction methods, the proposed method can achieve better recognition performance.

关 键 词：ROUGH集 车牌汉字 特征选择 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP18[自动化与计算机技术—计算机科学与技术]