摘要
提出了一种将局部特征识别与全局特征识别相结合的人脸识别方法。该算法首先提取人脸的局部特征进行识别,然后提取未识别图像的全局特征进行识别。基于ORL人脸数据库的实验证明了该算法的识别性能要优于单一特征识别方法。
A face recognition technique based on local feature and global feature was presented. First the original face image＇s local features were extracted. Then utilize nearest neighbor to classify and extract the error classify image＇s global feature, also utilize nearest neighbor to classify. The experiment shows that the performance of the proposed method in this article is superior to that of single feature.
出处
《计算机与数字工程》
2008年第7期152-153,共2页
Computer & Digital Engineering
关键词
人脸识别
主成分分析
局部特征
全局特征
face recognition, principle component analysis, local feature, global feature
