节点文献
基于正交Gegenbauer矩的车牌汉字字符识别
Licensed plate recognition based on orthogonal Gegenbauer moments
【摘要】 针对已分割完成的车牌字符,提出了一种基于正交Gegenbauer矩特征提取方法,并将其应用于车牌汉字字符识别中.Gegenbauer矩法有别于其他传统的字符特征提取方法,它能获得字符的全局特征信息,因此可以唯一地表征一个字符.通过与其他方法的对比实验,可以看出此方法在低噪环境下具有更好的识别效果.
【Abstract】 A liensed plated recognition method is proposed,based on Gegenbauer moments. Compared with other feature extraction method, it can capture the global properties of characters rather than local ones. Accordingly, this method can represent each character uniquely regardless how close the characters are in terms of local features. The comparatine experimental results show that the method based on Gegenbauer moments performs better in the low noise environment.
【关键词】 车牌识别;
Gegenbauer矩;
字符识别;
特征提取;
低噪环境;
【Key words】 licensed plate recognition; Gegenbauer moments; character recognition; feature extraction; low noise environment;
【Key words】 licensed plate recognition; Gegenbauer moments; character recognition; feature extraction; low noise environment;
- 【文献出处】 武汉大学学报(工学版) ,Engineering Journal of Wuhan University , 编辑部邮箱 ,2008年01期
- 【分类号】TP391.43
- 【被引频次】5
- 【下载频次】276