Abstract
To solve the problem of over-reliance on a priori assumptions of the parametric methods for finite mixture models and the problem that monic Chebyshev orthogonal polynomials can only process the gray images, a segmentation method of mixture models of multivariate Chebyshev orthogonal polynomials for color image was proposed in this paper. First, the multivariate Chebyshev orthogonal polynomials are derived by the Fourier analysis and tensor product theory, and the nonparametric mixture models of multivariate orthogonal polynomials are proposed. And the mean integrated squared error is used to estimate the smoothing parameter for each model. Second, to resolve the problem of the estimation of the number of density mixture components, the stochastic nonparametric expectation maximum algorithm is used to estimate the orthogonal polynomial coefficient and weight of each model. This method does not require any prior assumptions on the models, and it can effectively overcome the problem of model mismatch. Experimental performance on real benchmark images shows that the proposed method performs well in a wide variety of empirical situations.
Similar content being viewed by others
References
Cheng H, Chen C, Chiu H et al (1998) Fuzzy homogeneity approach to multilevel thresholding. IEEE Trans Image Process 7(7):1084–1086
Cheng HD, Jiang XH, Wang J (2002) Colour image segmentation based on homogram thresholding and region merging. Pattern Recognit 35(2):373–393
Otsu N (1979) A threshold selection method from grey-level histograms. IEEE Trans Syst Man Cybern 9(1):62–66
Saha P, Udupa J (2002) Optimum image thresholding via class uncertainty and region homogeneity. IEEE Trans Pattern Anal Mach Intell 23(5):689–706
Basak J, Chanda B, Majumder DD (1994) On edge and line linking with connectionist models. IEEE Trans Syst Man Cybern 24(3):413–428
Shih FY, Cheng S (2004) Adaptive mathematical morphology for edge linking. Inf Sci 167(1–4):9–21
Gonzalez RC, Woods RE (2002) Digital image processing. Prentice-Hall, Englewood Cliffs
Hojjatoleslami SA, Kittler J (1998) Region growing: a new approach. IEEE Trans Image Process 7(7):1079–1084
Tremeau A, Borel N (1997) A region growing and merging algorithm to colour segmentation. Pattern Recognit 30(7):1191–1203
Ikonomakis N, Plataniotis KN, Zervakis M et al (1997) Region growing and region merging image segmentation. Proc IEEE Conf Digit Signal Process 1:299–302
Murtagh F, Raftery AE, Starck JL (2005) Bayesian inference for multiband image segmentation via model-based cluster trees. Image Vis Comput 23:596–597
Fraley C, Raftery AE (2002) Model-based clustering, discriminant analysis, and density estimation. J Am Stat Assoc 97:611–631
Titterington DM (1985) Statistical analysis of finite mixture distributions. Wiley, New York
Marin JM, Mengersen K, Robert CP (2005) Bayesian modelling and inference on mixtures of distributions. In: Handbook of statistics, vol 25, pp 459–507
McLachlan GJ, Peel D (2000) Finite mixture models. Wiley, New York
da Silva F (2009) Bayesian mixture models of variable dimension for image segmentation. Comput Methods Programs Biomed 94(1):1–14
Veeberk JJ, Vlassis N, Klose B (2001) Greedy gaussian mixture model learning for texture image segmentation. In: Proceedings of the workshop on kernel and subspace methods for computer vision, pp 37–46
Zhang K, Kwok JT (2010) Simplifying mixture models through function approximation. IEEE Trans Neural Netw 21(4):644–658
Joshi N, Brady M (2010) Non-parametric mixture model based evolution of level sets and application to medical images. Int J Comput Vis 88(1):52–68
Peel D, McLachlan GJ (2000) Robust mixture modelling using the t distribution. Stat Comput 10(4):339–348
Tsung IL, Jack CL (2007) Robust mixture modeling using the skew t distribution. Stat Comput 17(2):81–92
Bouguila N, Ziou D, Vaillancourt J (2004) Unsupervised learning of a finite mixture model based on the Dirichlet distribution and its application. IEEE Trans Image Process 13(11):1533–1543
Fan S (2008) Image thresholding using a novel estimation method in generalized Gaussian distribution mixture modeling. Neurocomputing 72(1–3):500–512
Zribi M, Ghorbel F (2003) An unsupervised and non-parametric Bayesian classifier. Pattern Recognit Lett 24:97–112
Zribi M, Ghorbel F (2007) Unsupervised Bayesian image segmentation using orthogonal series. J Vis Commun Image Represent 18(6):496–503
Koornwinder T (1974) Orthogonal polynomials in two variables which are eigenfunctions of the two algebraically independent partial differential operators. Nederl Acad Wetensch Proc Ser A77, 36:357–381
Koornwinder T (1975) Two-variable analogues of the classical orthogonal polynomials. In: Askey RA (ed) Theory and applications of special functions. Academic Press, New York, pp 435–495
Dunkl CF, Xu Y (2001) Orthogonal polynomials of several variables. Cambridge University Press, Cambridge
Suetin PK (1999) Orthogonal polynomials in two variables (trans: the 1988 Russian original by Panklatiev EV). Gordon and Breach, Amsterdam
Caillol H, Pieczynski W, Hillion A (1997) Estimation of fuzzy Gaussian mixture and unsupervised statistical image segmentation. IEEE Trans Image Process 6:425–440
Redner RA, Walker HF (1984) Mixture densities, maximum likelihood and the EM algorithm. SIAM Rev 26:195–239
Masson P, Pieczynski W (1993) SEM algorithm and unsupervised statistical segmentation of satellite images. IEEE Trans Geosci Remote Sens 31:618–633
Braathen B, Pieczynski W, Masson P (1993) Global and local methods of unsupervised Bayesian segmentation of images. Mach Graph Vis 2:39–52
Zhang YJ (1997) Evaluation and comparison of different segmentation algorithms. Pattern Recognit Lett 18:963–974
Dempester AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc B39:1–38
Tang H, Dillenseger JL, Bao XD (2009) A vectorial image soft segmentation method based on neighborhood weighted Gaussian mixture model. Comput Med Imaging Graph 33(8):644–650
Kanungo T, Mount DM, Netanyahu NS, Piatko CD, Silverman R, Wu YA (2002) An efficient k-means clustering algorithm: analysis and implementation. IEEE Trans Pattern Anal Mach Intell 24(7):881–892
Silverman BW (1986) Density estimation for statistics and data analysis. Chapman and Hall, London
Scott DW (1992) Multivariate density estimation: theory, practice, visualization. Wiley, New York
Gehringer K (1990) Nonparametric probability density estimation using normalized b-splines. Master’s Thesis, The University of Tulsa
Cencov NN (1962) Evaluation of an unknown distribution density from observations. Sov Math 3:1559–1562
Koornwinder T (1975) Two-variable analogues of the classical orthogonal polynomials. In: Askey RA (ed) Theory and applications of special functions. Academic Press, New York
Sun JC (2008) A new class of three-variable orthogonal polynomials and their recurrences relations. Sci China Ser A Math 51(6):1071–1092
Sun JC, Li H (2005) Generalized Fourier transform on an arbitrary triangular doman. Adv Comput Math 22:223–248
Sun JC (2006) Multivariate Fourier transform methods over simplex and super-simplex domains. J Comput Math 24:55–66
Sun JC (2006) Approximate eigen-decomposition preconditioners for solving numerical PDE problems. Appl Math Comput 172:772–787
Sun JC (2002) Generalized Fourier transforms over parallel dodecahedron domains. Ann RDCPS 04-01:18–25
Asmar NH (2004) Partial differential equations with Fourier series and boundary value problems, 2nd edn. Prentice Hall, Upper Saddle River
Fowlkes C, Martin D, Malik J The Berkeley segmentation dataset and benchmark (BSDB). www.cs.berkeley.edu/projects/vision/grouping/segbench/
Comaniciu D, Meer P (2002) Mean shift: a robust approach toward feature space analysis. IEEE Trans Pattern Anal Mach Intell 24(5):603–619
Martin DR, Fowlkes CC, Malik J (2004) Learning to detect natural image boundaries using local brightness, color, and texture cues. IEEE Trans Pattern Anal Mach Intell 26(5):530–549
Zhang H, Jason E, Fritts B, Sally A (2008) GoldmanImage segmentation evaluation: a survey of unsupervised methods. Comput Vis Image Underst 110(2):260–280
Acknowledgments
The authors would like to thank Prof. Sun Jiachang and the referees for their helpful comments and suggestions to improve the presentation of this paper. This work was supported by the National Science Foundation of China under Grant No. 60841003, the Special Fund of Software and Integrated Circuit Foundation of Jiangsu Information Industry of China (2009[100]) and the Innovation Fund of PhD candidate in Jiangsu Province of China(CX10B_274Z).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Liu, Z., Song, YQ., Chen, JM. et al. Color image segmentation using nonparametric mixture models with multivariate orthogonal polynomials. Neural Comput & Applic 21, 801–811 (2012). https://doi.org/10.1007/s00521-011-0538-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-011-0538-1