Abstract:Aiming at the overflow of the hidden Markov model (HMM) observation probability, a method is proposed, called Normality Processing. Firstly, the chi-square plot is used to test normality of the sample set, the transformation of square root is performed. The feasibility of the proposed method is validated on the expression sequences database of CED-WYU(1.0) and Cohn-Kanade (CMU). The person-independent expression recognition experiment is made with continuous HMM based on the optical flow features and a better result is obtained when the normality processing is used.
[1] Rabiner L R. A Tutorial on Hidden Markov Models and Selected Application in Speech Recognition. Proc of the IEEE, 1989, 77(2): 257-286 [2] Otsuka T, Ohya J. Recognizing Multiple Persons' Facial Expression Using HMM Based on Automatic Extraction of Significant Frames from Image Sequences // Proc of the International Conference on Image Processing. Barbara, USA, 1997, Ⅲ: 546-549 [3] BenYishai A, Brushtein D. A Discriminative Training Algorithm for Hidden Markov Models. IEEE Trans on Speech and Audio Processing, 2004, 12(3): 204-217 [4] He Q H, Kwong S, Hong Q Y. Adaptation of Hidden Markov Models Using Maximum Model Distance Algorithm. IEEE Trans on Systems, Man and Cybernetics, 2004, 34(2): 270-276 [5] Aleksic P S, Katsaggelos A K. Automatic Facial Expression Recognition Using Facial Animation Parameters and Multi Stream HMMs. IEEE Trans on Information Forensics and Security, 2006, 1(1): 3-11 [6] He Qiang, Mao Shiyi, Zhang Youwei. Re-Estimation of Continuous Hidden Markov Model with Multiple Observation without Overflow. Acta Electronica Sinica, 2000, 28(10): 98-101 (in Chinese) (何 强,毛士艺,张有为.多观察序列连续隐含马尔柯夫模型的无溢出参数重估.电子学报, 2000, 28(10): 98-101) [7] Looney S W, Jr Gulledge T R. Use of the Correlation Coefficient with Normal Probability Plots. The American Statistician, 1985, 39(1): 75-79 [8] Johnson R A, Wichern D W. Applied Multivariate Statistical Analysis. Upper Saddle River, USA: Prentice-Hall, 1982 [9] Tang Jinghai, Ying Zilu, Zhang Youwei. The Contrast Analysis of Facial Expression Recognition by Human and Computer // Proc of the International Conference on Signal Processing. Beijing, China, 2006, Ⅲ: 1649-1653 [10] Cohn J F, Zlochower A J, Lien J T, et al. Feature-Point Tracking by Optical Flow Discriminates Subtle Differences in Facial Expression // Proc of the IEEE International Conference on Automatic Face and Gesture Recognition. Nara, Japan, 1998, Ⅲ: 396-401 [11] Gautanm T, van Hulle M M. A Phase-Based Approach to Estimation of the Optical Flow Field Using Spatial Filtering. IEEE Trans on Neural Networks, 2002, 13(5): 1127-1136 [12] Zitnick C W, Jojic N, Kang S B. Consistent Segmentation for Optical Flow Estimation // Proc of the 10th International Conference on Computer Vision. Beijing, China, 2005, Ⅱ: 1308-1315