一种基于Harris-SIFT特征点检测的LBP人脸表情识别算法
|
摘要
针对LBP算法在提取全局特征时不具有针对性的缺点,提出一种基于Harris-SIFT特征点检测的LBP人脸表情识别算法。该算法引入Harris得到表情图像的角点,同时运用SIFT检测算法得到图像的局部最大最小值,通过Harris算法对SIFT特征点进行过滤,得到表情图像的准确感兴趣点;设计特征点最大区域选取法和特征点周围邻域选取法,将选取的区域作为LBP特征提取的输入图像;运用SVM多分类器得到每种表情的识别率。实验表明,该算法能更有效地提取表情特征,达到较好的识别效果。
Aiming at the shortcoming of LBP algorithm that it is not targeted at extracting global features,the LBP algorithm for facial expression recognition based on Harris-SIFT feature point detection was proposed.First,Harris corners algorithm was introduced,at the same time,the SIFT detection algorithm was used to calculate local minimum and maximum value of facial expression images. Then SIFT feature point was filtered by Harris corners to obtain accuracy feature points. Feature point largest region selection method and feature point neighborhood selection method were designed to get the selected region as the input image of LBP feature extraction; then each kind of facial expression recognition rate was got by the multi-classifier of SVM. Experiments results show that the proposed algorithm extracts facial features more effectively and achieves better recognition results.
Aiming at the shortcoming of LBP algorithm that it is not targeted at extracting global features,the LBP algorithm for facial expression recognition based on Harris-SIFT feature point detection was proposed.First,Harris corners algorithm was introduced,at the same time,the SIFT detection algorithm was used to calculate local minimum and maximum value of facial expression images. Then SIFT feature point was filtered by Harris corners to obtain accuracy feature points. Feature point largest region selection method and feature point neighborhood selection method were designed to get the selected region as the input image of LBP feature extraction; then each kind of facial expression recognition rate was got by the multi-classifier of SVM. Experiments results show that the proposed algorithm extracts facial features more effectively and achieves better recognition results.
引文
[1]OJALA T,PIETIK INEN M,HARWOOD D.A comparative study of texture measures with classification based on featured distributions[J].Pattern Recognition,1996,29(1):51-59.
[2]TOPI M,MATTI P,TIMO O.Texture classification by multi-predicate local binary pattern operators[C]∥International Conference on Pattern Recognition,2000,15:939-942.
[3]MENPT,PIETIKI NEN M.Texture analysis with local binary patterns[J].Handbook of Pattern Recognition and Computer Vision,2005(3):197-216.
[4]HADID A,PIETIKI NEM M,AHONEN T.A discriminative feature space for detecting and recognizing faces[C]∥Computer Vision and Pattern Recognition,2004.Proceedings of the 2004 IEEE Computer Society Conference on.IEEE,2004,2:II-797-II-804.
[5]TURTINEN M,PIETIKAINEN M,SILVN O.Visual characterization of paper using isomap and local binary patterns[J].IEICE Transactions on Information and Systems,2006,89(7):2076-2083.
[6]SUMER O,GUNES E O.LBP and SIFT based facial expression recognition[C]∥Seventh International Conference on Machine Vision(ICMV 2014).International Society for Optics and Photonics,2015:94450A-94450A-5.
[7]ZHONG L,LIU Q,YANG P,et al.Learning active facial patches for expression analysis[C]∥Computer Vision and Pattern Recognition(CVPR),2012 IEEE Conference on.IEEE,2012:2562-2569.
[8]YANG M,ZHENG S,LI Y.Facial expression recognition based on LBP and SVM decision tree[C]∥2015 International Industrial Informatics and Computer Engineering Conference.Atlantis Press,2015.
[9]SHAN C,GONG S,MCOWAN P W.Facial expression recognition based on local binary patterns:A comprehensive study[J].Image and Vision Computing,2009,27(6):803-816.
[10]LOWE D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[11]姜桂圆,张桂玲,张大坤.SIFT特征分布式并行提取算法[J].计算机研究与发展,2015,49(5):1130-1141.
[12]刘仲,邢彬朝,陈跃跃.一种面向多核处理器的高效并行PCA-SIFT算法[J].国防科技大学学报,2012,34(4):103-107.
[13]朱英宏,李俊山,郭莉莎,等.基于LBP的尺度不变特征的描述和匹配算法[J].计算机辅助设计与图形学学报,2011,23(10):1758-1763.
[14]郑永斌,黄新生,丰松江.SIFT和旋转不变LBP相结合的图像匹配算法[J].计算机辅助设计与图形学学报,2010,22(2):286-292.
[15]陈梦婷,闫冬梅,王刚.基于Harris角点和SIFT描述符的高分辨率遥感影像匹配算法[J].中国图象图形学报,2012,17(11):1453-1459.
[2]TOPI M,MATTI P,TIMO O.Texture classification by multi-predicate local binary pattern operators[C]∥International Conference on Pattern Recognition,2000,15:939-942.
[3]MENPT,PIETIKI NEN M.Texture analysis with local binary patterns[J].Handbook of Pattern Recognition and Computer Vision,2005(3):197-216.
[4]HADID A,PIETIKI NEM M,AHONEN T.A discriminative feature space for detecting and recognizing faces[C]∥Computer Vision and Pattern Recognition,2004.Proceedings of the 2004 IEEE Computer Society Conference on.IEEE,2004,2:II-797-II-804.
[5]TURTINEN M,PIETIKAINEN M,SILVN O.Visual characterization of paper using isomap and local binary patterns[J].IEICE Transactions on Information and Systems,2006,89(7):2076-2083.
[6]SUMER O,GUNES E O.LBP and SIFT based facial expression recognition[C]∥Seventh International Conference on Machine Vision(ICMV 2014).International Society for Optics and Photonics,2015:94450A-94450A-5.
[7]ZHONG L,LIU Q,YANG P,et al.Learning active facial patches for expression analysis[C]∥Computer Vision and Pattern Recognition(CVPR),2012 IEEE Conference on.IEEE,2012:2562-2569.
[8]YANG M,ZHENG S,LI Y.Facial expression recognition based on LBP and SVM decision tree[C]∥2015 International Industrial Informatics and Computer Engineering Conference.Atlantis Press,2015.
[9]SHAN C,GONG S,MCOWAN P W.Facial expression recognition based on local binary patterns:A comprehensive study[J].Image and Vision Computing,2009,27(6):803-816.
[10]LOWE D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[11]姜桂圆,张桂玲,张大坤.SIFT特征分布式并行提取算法[J].计算机研究与发展,2015,49(5):1130-1141.
[12]刘仲,邢彬朝,陈跃跃.一种面向多核处理器的高效并行PCA-SIFT算法[J].国防科技大学学报,2012,34(4):103-107.
[13]朱英宏,李俊山,郭莉莎,等.基于LBP的尺度不变特征的描述和匹配算法[J].计算机辅助设计与图形学学报,2011,23(10):1758-1763.
[14]郑永斌,黄新生,丰松江.SIFT和旋转不变LBP相结合的图像匹配算法[J].计算机辅助设计与图形学学报,2010,22(2):286-292.
[15]陈梦婷,闫冬梅,王刚.基于Harris角点和SIFT描述符的高分辨率遥感影像匹配算法[J].中国图象图形学报,2012,17(11):1453-1459.