摘要
本文介绍了一种基于粗糙集理论的优化车牌字符识别的方法。粗糙集理论是一种继神经元网络和模糊数学之后的新的处理含糊和不确定性知识的数学工具;粗糙集方法是一种具有发展潜力的智能信息处理方法。本文主要思想就是在汉字的网格特征提取过程中保持分类能力不变的前提下,通过知识约简提出了一种车牌字符网格特征选择的改进算法;它不仅找出了对识别最有效的网格特征集,而且可以大大降低图像特征空间的维数,减少工作量和无用特征干涉,从而提高了分类识别率。
This paper introduced a method of character recognition of license plate. Rough sets theory was a new mathematical tool to deal with problems on vagueness and uncertainty following neural network and fuzzy method , and Rough set was a potential intelligent information processing method. Remaining the ability of classification , this paper present a new algorithm to extract the features of license plate on the basis of knowledge reduction. Not only could the method search available feature sets, reduce feature dimension, but also speed the classification, improve the recognition rate.
出处
《九江学院学报》
2008年第6期36-39,共4页
JOurnal of Jiujiang University :Social Science Edition
关键词
粗糙集
字符识别
分类
知识约简
特征
rough sets
character reeognition
classification
knowledge reduction
feature
