CAJ | 학술논문

介绍了一种基于稀疏编码的人脸识别算法。先对10副自然图像应用稀疏编码，学习到基函数和图像稀疏表示的拟合分布的参数。在人脸识别中，用稀疏编码和已得到的基函数表示图像的稀疏，再经过拟合分布函数得到人脸图像的最终表示，然后应用多分类线性支持向量机（SVM）来完成识别算法。通过在人脸数据库上的实验表明，该算法具有很高的识别正确率。
개소료일충기우희소편마적인검식별산법。선대10부자연도상응용희소편마，학습도기함수화도상희소표시적의합분포적삼수。재인검식별중，용희소편마화이득도적기함수표시도상적희소，재경과의합분포함수득도인검도상적최종표시，연후응용다분류선성지지향량궤（SVM）래완성식별산법。통과재인검수거고상적실험표명，해산법구유흔고적식별정학솔。
This paper mainly presents an approach to face recognition via sparse coding. Learn the basis function and parame-ters of the sparse representation's fitted distribution after applying sparse coding to ten natural images. In the process of face recognition, the basis function is used to get the images' sparse representation via sparse coding, followed by the fitted distribution function to get final image presentation. Multi-class linear SVM is chosen as the classifier to finish the recognition. The algorithm is applied to face datasets and the results show that it has high recognition correct rate.