摘要
提出基于稀疏表示的支持向量机(support vector machine,SVM)人脸识别方法,首先将人脸图像通过稀疏表示出来,然后用SVM对稀疏表示的人脸图像进行多分类识别,利用所提出的方法对ORL人脸图像库进行仿真实验.仿真结果表明,该方法优于一般的主成份分析结合SVM人脸识别方法,同时比单纯的稀疏表示编码方法的效果要好.
In this paper, we proposed a new face recognition method based on sparse representation coding and SVMs （support vector machines）. Firstly, the face is represented by sparse coding. Then, sparse representation is classified by SVMs. At last, the proposed method is used to classify the face of ORL. The result showed the proposed algorithm is better than the classic algorithm based on PCA and SVM, and its performance is better than SRC.
出处
《应用数学与计算数学学报》
2016年第3期437-444,共8页
Communication on Applied Mathematics and Computation
基金
淮南联合大学质量工程建设资助项目(00881106)
关键词
稀疏表示
支持向量机
人脸识别
字典学习
sparse representation
SVMs （support vector machines）
face recognition
dictionary learning
