基于非负矩阵分解的面部表情识别研究

Facial Expression Recognition Research Based on Non-negative Matrix Factorization

提出了一种基于非负矩阵分解与支持向量机相结合的面部表情识别方法.使用直方图均衡化等方法对人脸图像进行预处理,使用非负矩阵分解算法进行表情特征提取,采用支持向量机对面部表情进行分类.以Matlab为仿真工具,在日本女性人脸表情数据库上测试,取得了66.19%的识别率.

A novel approach to facial expression recognition based on the combination of Non-negative Matrix Factorization （NMF） and support vector machine （SVM） was proposed. First, the algorithm processed facial expression images with Histogram equalization operator. Then NMF method was used for feature dimension reduction and SVM for classification. Finally, the algorithm was implemented with Matlab and experimented in Japanese female facial expression database （JAFFE database）. A recognition rate of 66.1% was obtained and the effectiveness of the proposed algorithm is proved.