Unsupervised Texture Classification and Segmentation

Authors: V.P.Subramanyam Rallabandi, S.K.Sett

Abstract:

An unsupervised classification algorithm is derived by modeling observed data as a mixture of several mutually exclusive classes that are each described by linear combinations of independent non-Gaussian densities. The algorithm estimates the data density in each class by using parametric nonlinear functions that fit to the non-Gaussian structure of the data. This improves classification accuracy compared with standard Gaussian mixture models. When applied to textures, the algorithm can learn basis functions for images that capture the statistically significant structure intrinsic in the images. We apply this technique to the problem of unsupervised texture classification and segmentation.

Keywords: Gaussian Mixture Model, Independent Component Analysis, Segmentation, Unsupervised Classification.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1332180

References:

[1] R.W.Buccigrossi, E.P.Simoncelli, "Image compression via joint statistical characterisation in the wavelet domain," IEEE Trans. Image Processing, Vol.8, Dec. 1999, pp.1688-1701.
[2] J.-F. Cardos, B.Laheld, "Equivariant adaptive source separation", IEEE. Trans. Signal Processing", Vol. 45, Feb. 1996, pp.434-444.
[3] R.O.Duda and P.E. Hart, Pattern Classification and Scene Analysis. New York: Wiley, 1973.
[4] Y.-W.Wang, Y.-F.Wang, Y. Xue, W.Gao, "A new algorithm for remotely sensed image classification and segmentation", GeoScience and Remote Sensing Symposium 2003, IGARSS-03 Proceedings, Vol.6, July 2003, pp.3534-3536.
[5] Unser. M, "Texture classification and segmentation using wavelet frames," IEEE Trans. Image Processing., Vol.4, Issue 11, Nov. 1995,pp.1549-1560.
[6] Charalampidis.D, Kasparis.T, "Wavelet-based rotational invariant roughness features for texture classification and segmentation," IEEE Trans.Image Processing., Vol. 11, Issue .8, Aug. 2002, pp.825-837.
[7] Kaplan L.M., "Extended fractal analysis for texture classification and segmentation", IEEE Trans. Image Processing, Vol.8, Issue 11, Nov 1999, pp.1572-1585.
[8] Arof.H, Deravi.F, "Circular neighbourhood and 1-D DFT features for texture classification and segmentation," IEEE Proc. Vision, Image and Signal Processing, Vol.145, Issue 3, June 1998, pp.167-172.
[9] Tan T.N., Constantinides A.G., "Texture analysis based on a human visual system", Acoustics, Speech and Signal Processing, ICASSP 90, 3- 6, April, 1990, Vol.4, pp.2137-2140.
[10] Lueng. M., Peterson A.M., "Multiple channel neural network model for texture classification and segmentation," Acoustics Speech and Signal Processing ICASSP 91, 14-17 April 1991, Vol.4,pp.2677-2680.
[11] Mengyang Liao, J.Qin, Yanni Tan, "Texture classification and segmentation using autoregressive models," Computer-based Medical Systems, Fifth Annual IEEE Symposiums, 14-17 June, 1992, pp.398- 401.
[12] Patel .D, Stonham .T.J.., "Texture image classification and segmentation using rank-order clustering," Pattern Recognition, Vol.3, Eleventh IAPR International Conference, 30 Aug-3 Sep, 1992, pp.92-95.
[13] Jiang Wen, You Zhiseng, Li Hui, "Segment the metallograph images using Gabor filter", Speech, Image Processing and Neural Networks, Proceedings ISSIPNN-94, 1994 International Symposium, 13-16 Apr, 1994, Vol.1, pp.25-28.
[14] Oja.E, Valkealathi .K, "Compressing higher-order co-occurrences for texture analysis using self-organising maps" in International Conference on Neural Networks,,27 Nov-1 Dec, 1995, Vol.2. pp.1160- 1164.
[15] Gambotto.J, Gueguen.C,, "A multidimensional modeling approach to texture classification and segmentation," Acoustics, Speech and Image Processing, IEEE Int. Conf. on ICASSP-79 Vol.4., pp. 962-966.
[16] Arrowsmith.M.J., Varley.M.R., Picton. P.D., Heys. J.D.,, "Hybrid neural network system for texture analysis", Image Processing and its Applications 1999, Seventh International Conference (pub.No.465), 13- 15 July 1999,Vol..1, pp.339-343.
[17] Claude.I, Smolarz.Z.A, "A new textured image segmentation algorithm by autoregressive modeling and multiscale block classification", Image Processing and its Applications 1997, Sixth.International Conference, 14-17 July 1997, Vol.2, pp. 586-590.