Human Facial Expression Recognition Using Wavelet Transform and Hidden Markov Model

The accuracy of the Facial Expression Recognition (FER) system is completely reliant on the extraction of the informative features. In this work, a new feature extraction method is proposed that has the capability to extract the most prominent features from the human face. The proposed technique has been tested and validated in order to achieve the best accuracy for FER systems. There are some regions in the face that have much contribution in achieving the best accuracy. Therefore, in this work, the human face is divided into number of regions and in each region the movement of pixels have been traced. For this purpose, one of the wavelet families named symlet wavelet is used and individual facial frame is decomposed up to 2 levels. In each decomposition level, the distances between the pixels is found by using the distance formula and by this way some of the informative coefficients are extracted and hence the feature vector has been created. Moreover, the dimension of the feature space is reduced by employing a well-known statistical technique such as Linear Discriminant Analysis (LDA). Finally, Hidden Markov Model (HMM) is exploited for training and testing the system in order to label the expressions. The proposed FER system has been tested and validated on Cohn-Kanade dataset. The resulting recognition accuracy of 94% illustrates the success of employing the proposed technique for FER.

[1]  A. Mehrabian.  Communication without words , 1968 .

[2]  L. Baum,et al.  An inequality and associated maximization technique in statistical estimation of probabilistic functions of a Markov process , 1972 .

[3]  Ferdinando Silvestro Samaria,et al.  Face recognition using Hidden Markov Models , 1995 .

[4]  Penio S. Penev,et al.  Local feature analysis: A general statistical theory for object representation , 1996 .

[5]  David J. Kriegman,et al.  Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection , 1996, ECCV.

[6]  T. Sejnowski,et al.  Face image analysis for expression measurement and detection of deceit , 1999 .

[7]  Marian Stewart Bartlett,et al.  Classifying Facial Actions , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Pong C. Yuen,et al.  Human face recognition using PCA on wavelet subband , 2000, J. Electronic Imaging.

[9]  Takeo Kanade,et al.  Comprehensive database for facial expression analysis , 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580).

[10]  Stan Z. Li,et al.  Learning spatially localized, parts-based representation , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[11]  Beat Fasel,et al.  Automati Fa ial Expression Analysis: A Survey , 1999 .

[12]  Jongsun Kim,et al.  Effective representation using ICA for face recognition robust to local distortion and partial occlusion , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Daw-Tung Lin.  Facial Expression Classification Using PCA and Hierarchical Radial Basis Function Network , 2006, J. Inf. Sci. Eng..

[14]  Daw-Tung Lin,et al.  Facial Expression Classification Using PCA and Hierarchical Radial Basis Function Network , 2006, J. Inf. Sci. Eng..

[15]  Frank Y. Shih,et al.  Recognizing facial action units using independent component analysis and support vector machine , 2006, Pattern Recognit..

[16]  Marc Toussaint,et al.  Extracting Motion Primitives from Natural Handwriting Data , 2006, ICANN.

[17]  Xufa Wang,et al.  Infrared Facial Expression Recognition Using Wavelet Transform , 2008, 2008 International Symposium on Computer Science and Computational Technology.

[18]  Kaur Mandeep,et al.  Comparative Study of Facial Expression Recognition Techniques , 2011 .

[19]  Jukka Turunen,et al.  A wavelet-based method for estimating damping in power systems , 2011 .