孔玮婷 皋军 丁泽超 史恭波
摘要:针对基于局部纹理特征的人脸表情识别算法不能有效表达不同表情状态下人脸运动单元差异性的问题,提出一种改进的稀疏表示人脸表情识别算法,将人脸纹理特征与全局位置特征用稀疏表示模型相结合,得到人脸表情的稀疏系数矩阵,并作为支持向量机表情识别的输入。人脸表情库BU_3DFE实验结果表明,该算法提高了表情识别的准确率。
关键词:表情识别;稀疏表示;特征融合
DOIDOI:10.11907/rjdk.161671
中图分类号:TP312文献标识码:A文章编号:1672-7800(2016)006-0031-02
参考文献:
[1]JI ZHENG Y,XIA M,MITSURU L,YU-LI X.Facial expression recognition based on feature point vector and texture deformation energy parameters[J].Journal of Electronics and Information Technology,2013,35(10):2403-2410.
[2]ZHAN YONGZHAO,YE JINGFU,NIU DEJIAO.Facial expression recognition based on gabor wavelet transformation and elastictemplates matching[J].International Journal of Image and Graphics,2006,6(1):125-138.
[3]L YIN,X WEI,J WANG,et al.A 3D facial expression database for facial behavior research[C].IEEE International Conference on Automatic Face and Gesture Recognition.(FG2006)[C],IEEE Computer Society TC PAMI.Southampton,UK,2006.
[4]MOORE S,BOWDEN R.Local binary patterns for multi-view facial expression recognition[J].Computer Vision and Image Understanding,2011,115(4):541-558.
[5]BU FA H,YIN-CHENG H,BING-XING C.A novel facial expression recognition method based on semantic knowledge of analytical hierarchy process[J].Journal of Image and Graphics,2011.
[6]RUDOVIC,OGNJEN,PATRAS.Coupled Gaussian process regression for pose-invariant facial expression recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(6):1357-1369.