中文摘要：

车牌识别是智能交通系统的重要组成部分,提高车牌字符识别率的关键在于提取字符的特征。主曲线是主成分分析的非线性推广,它是通过数据分布“中间”并满足“自相合”的光滑曲线。通过对现有主曲线算法的分析可知：软K段主曲线算法对提取分布在弯曲度很大或相交曲线周围的数据的主曲线的效果较好。因此,尝试用该主曲线算法来提取车牌字符的结构特征。实验结果表明,利用该主曲线算法来提取车牌识别的结构特征能够取得较好的实验效果。所提方法为提取车牌字符特征的研究提供了一条新途径。

英文摘要：

License plate recognition is an important part of intelligent transportation systems. In order to improve the recognition rate of LPR characters, extraction of features are critical. Principal curves are nonlinear generalizations of principal components analysis. They are smooth self-consistent curves that pass through the ＂middle＂ of the distribu- tion. By analysis of existed principal curves, we learned that a soft K-segments algorithm for principal curves exhibits good performance in such situations in which the data sets are concentrated around a highly curved or self-intersecting curves. Therefore,we attemptd to use the algorithm to extract structural features of LPR characters. Experiment results show that the algorithm is feasible for extraction of structural features of LPR characters. The proposed method can provide a new approach to the research for extraction of LPR characters features.