Abstract:ICA coefficients are non-Gaussian in independent component analysis model. The high-order statistical features are used in characterization of non-Gaussian feature. The combined moments of variance, skewness and kurtosis are proposed to describe the ICA coefficients probability distributing characteristic. The combined moments are used in texture classification and it can achieve better classification performance than the previously reported ICA features. Furthermore, L-moments are used to improve robustness in moments estimation and to get better performance than the ordinary moments.
徐小红,杨学志,杨德美,高隽. ICA系数的高阶统计特征在纹理分类中的应用*[J]. 模式识别与人工智能, 2009, 22(3): 499-505.
XU Xiao-Hong, YANG Xue-Zhi, YANG De-Mei, Gao Jun. Application of Higher-Order Statistics Features of ICA Coefficients in Texture Classification. , 2009, 22(3): 499-505.
[1] Tuceryan M, Jain A K. Texture Analysis // Chen C H, Pau L F, Wang P S P, eds. Handbook of Pattern Recognition and Computer Vision. 2nd Edition. Singapore, Singapore: World Scientific, 1999: 207-248 [2] Randen T, Husoy J H. Filtering for Texture Classification: A Comparative Study. IEEE Trans on Pattern Analysis and Machine Intelligence, 1999, 21(4): 291-310 [3] Bell A J, Sejnowski T J. The Independent Components of Natural Scenes Are Edge Filters. Vision Research, 1997, 37(23): 3327-3338 [4] Hyvarinen J, Oja K E. Independent Component Analysis. New York, USA: Wiley, 2001 [5] Oppenheim A V, Lim J S. The Importance of Phase in Signals. Proc of the IEEE, 1981, 69(5): 529-541 [6] Chen Yang, Wang Runsheng. A Method for Texture Classification by Integrating Gabor Filters and ICA. Acta Electronica Sinica, 2007, 35(2): 299-303 (in Chinese) (陈 洋,王润生.结合Gabor滤波器和ICA技术的纹理分类方法.电子学报, 2007, 35(2): 299-303) [7] Zeng Xiangyan, Chen Yenwei, van Alphen D, et al. Selection of ICA Features for Texture Classification // Proc of the International Symposium on Neural Network. Chongqing, China, 2005: 262-267 [8] Liu Xuwen, Cheng Lei. Independent Filters for Texture Classification // Proc of the IEEE International Conference on Image Processing. New York, USA, 2002, Ⅲ: 113-116 [9] Liu Xuwen, Cheng Lei. Independent Spectral Representations of Images for Recognition. Journal of the Optical Society of American, 2003, 20(7): 1271-1282 [10] Liu Xuwen, Wang Deliang. Texture Classification Using Spectral Histograms. IEEE Trans on Image Processing, 2003, 12(6): 661-670 [11] Jenssen R, Eltoft T. Independent Components Analysis for Texture Segmentation. Pattern Recognition, 2003, 36(10): 2301-2315 [12] Manduchi R, Portilla J. Independent Component Analysis of Textures // Proc of the 7th IEEE International Conference on Computer Vision. Kerkyra, Greece, 1999, Ⅱ: 1054-1060 [13] Cheng Jian, Liu Qingshan, Lu Hanqing. Texture Classification Using Kernel Independent Component Analysis // Proc of the 17th International Conference on Pattern Recognition. Cambridge, UK, 2004, Ⅰ: 620-623 [14] Nandi A K. Blind Estimation Using Higher-Order Statistics. Dordrecht, Netherlands: Kluwer, 1999 [15] Barlow H B. Unsupervised Learning. Neural Computation, 1989, 3(1): 295-311 [16] Brodatz P. Texture: A Photographic Album for Artists and Designers.New York, USA: Dover, 1966 [17] Hosking J R M. L-Moments: Analysis and Estimation of Distributions Using Linear Combinations of Order Statistics. Journal of the Royal Statistical Society: Series B, 1990, 52(1): 105-124 [18] Hosking J R M, Wallis J R. A Comparison of Unbiased and Plotting-Position Estimators of L-Moments. Water Resources Research, 1995, 31(8): 2019-2025 [19] Elamir E A H, Seheult A H. Trimmed L-Moments. Computational Statistics and Data Analysis, 2003, 43(3): 299-314