摘要
提出了一种基于二维非负矩阵因子的人脸表情识别方法。该算法直接将2维人脸表情图像矩阵作为2维矩阵并结合NMF进行表情特征提取,称之为2DNMF。与NMF等不同,2DNMF充分利用表情图像矩阵中的行向量间的信息和列向量间的信息,尽可能地保留了原始的表情信息。基于CED-WYU(1.0)和JAFFE两个表情数据库的识别结果表明,基于2维非负矩阵因子的特征提取方法能有效地提高识别率及效率。
A facial expression recognition method based on 2-Dimensional Non-negative Matrix Factorization(2DNMF) is proposed in this paper.2DNMF,which regards the 2D original expression images as 2D matrices and represents them with a set of 2D bases via appling NMF.Unlike NMF, 2DNMF takes full advantage of the information between rows and columns of the image,thus remains the original expression information as much as possible.Experimental result on CED-WYU(1.0)and JAFFE shows that it is an effective method for improving the recognition accuracy and effectiveness.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第26期182-183,187,共3页
Computer Engineering and Applications
关键词
非负矩阵因子
2维非负矩阵因子
表情识别
Non-negative Matrix Factorization(NMF)
2-Dimensional Non-negative Matrix Factorization(2DNMF)
expression recognition