摘要
特征提取是模式识别研究领域的一个热点问题。为了更好地解决人脸识别中的特征提取问题,在分布的非参数间距最大准则算法的基础上,根据最近邻线的算法思想,在点-线距离的基础上,提出一种新的算法,即最近邻线非参数鉴别分析算法,并通过在人脸数据库上的识别,验证了该算法的有效性。
Feature extraction is one of the hot topics in the field of pattern recognition. To solve the problem of feature extrac- tion in face recognition, a new algorithm which is called nearest neighbor line nonparametric discriminating analysis （NNL-NDA） algorithm is proposed. It is based on the algorithm of distributed nonparametric maximum spacing criteria, according to nearest neighbor line idea and in comparison with point to line distance. The validity of the algorithm is verified by recognition in face database.
出处
《现代电子技术》
北大核心
2015年第13期145-148,共4页
Modern Electronics Technique
