SAR image segmentation algorithm of regionalized fuzzy clustering based on the Gamma mixture model with variable shape parameters

LI Xiao-li1 ZHAO Quan-hua1 LI Yu1

(1.School of Geomatics, Liaoning Technical University, Fuxin, Liaoning Province, China 123000)
【Knowledge Link】speckle noise

【Abstract】With regard to the problem that the traditional fuzzy clustering algorithm cannot precisely describe the distribution characteristics of synthetic aperture radar (SAR) intensity image and overcome the inherent speckle noises, the SAR image segmentation algorithm of regionalized fuzzy clustering based on the Gamma mixture model (GaMM) with variable shape parameters is proposed. First, the image domain is completely divided into several Voronoi polygons by Voronoi tessellation. Assuming that the pixel intensities follow the GaMM with variable shape parameters, the dissimilarity between the intensities of pixels in Voronoi polygons and clusters is described by the negative logarithmic function of the GaMM. Then, the regionalized fuzzy objective function is defined with the combination of the GaMM and regularization term with spatial constraint between neighbor Voronoi polygons. In the parameter estimation procedure, the moving-updating operation is designed to solve the implicit parameters according to the criterion of minimizing the objective function. The qualitative and quantitative analyses for the segmentation results of real and simulated SAR images effectively prove the fitting ability of the regionalized GaMM with variable shape parameters to SAR data and the noise immunity of the proposed algorithm.

【Keywords】 variable shape parameter; Gamma mixture model; Voronoi tessellation; spatial constraint; fuzzy clustering; SAR image segmentation;


【Funds】 National Natural Science Foundation of China (41271435, 41301479) Natural Science Foundation of Liaoning Province, China (2015020090) Postgraduate Education Innovation Program of Liaoning Technical University, China (YB201605)

Download this article


    [1] Rodrigues F A, Neto J F, Marques R C, et al. SAR image segmentation using the roughness information [J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13 (2): 132–136.

    [2] Li Z, Yang Z, Xiong H. Homogeneous region segmentation for SAR images based on two steps segmentation algorithm [C]. Proceedings of 2015 International Conference on Computers, Communications, and Systems (ICCCS). New York: IEEE, 2015: 2–3.

    [3] Lee J, Jurkevich I. Segmentation of SAR images [J]. IEEE Transactions on Geoscience and Remote Sensing, 1989, 27 (6): 674–680.

    [4] Beauchemin M, Thomson K P, Edwards G, et al. On nonparametric edge detection in multilook SAR images [J]. IEEE Transactions on Geoscience and Remote Sensing, 1998, 36 (5): 1826–1829.

    [5] Majumdar J, Lekshmi S. SAR image segmentation using statistical techniques [J]. InternationalJournal of Computer Applications, 2011, 20 (8):89–92.

    [6] Wan L, You H J, Cheng Y B, et al. Research progress of synthetic aperture radar image segmentation [J]. Remote Sensing Technology and Application, 2018, 33 (1): 10–24 (in Chinese).

    [7] Otsu N. A threshold selection method from gray-level histograms [J]. IEEE Transactions on Systems, Man, and Cybernetics, 1979, 9 (1): 62–66.

    [8] Yin K Y, Liu H W, Jin L. Fast SAR image segmentation method based on Otsu adaptive double threshold [J]. Journal of Jilin University: Engineering and Technology Edition, 2011, 41 (6):760–765 (in Chinese).

    [9] Fjortoft R, Lopes A, Marthon P, et al. An optimal multiedge detector for SAR image segmentation [J]. IEEE Transactions on Geoscience and Remote Sensing, 1998, 36 (3): 793–802.

    [10] Zaart A E, Ziou D, Wang S, et al. Segmentation of SAR images [J]. Pattern Recognition. 2002, 35 (3): 713–724.

    [11] Tian X L, Jiao L C, Gou S P. SAR image segmentation using optimized FCM with weighted spatial function [J]. Journal of Xidian University, 2008, 35 (5): 846–852 (in Chinese).

    [12] Chatzis S P, Varvarigou T. A fuzzy clustering approach toward hidden Markov random field models for enhanced spatially constrained image segmentation [J]. IEEE Transactions on FuzzySystem, 2008, 16 (5): 1351–1361.

    [13] Nikou C, Galatsanos N P, Likas A C. A class-adaptive spatially variant mixture model for image segmentation [J]. IEEE Transactions on Image Processing, 2007, 16 (4): 1121–1130.

    [14] Xu F, Wang X C, Zhou M Q, et al. Segmentationalgorithm of brain vessel image based onSEM statistical mixture model [J]. Journal of Computer-Aided Design and Computer Graphics, 2010, 22 (11): 1905–1911 (in Chinese).

    [15] Hou Y, Sun X, Lun X, et al. Gaussian mixture model segmentation algorithm for remote sensing image [C]. Proceedings of 2010 International Conference on Machine Vision and Human-Machine Interface. New York: IEEE, 2010: 275–278.

    [16] Goodman J W. Statistical properties of laser speckle patterns [C]. Laser Speckle and Related Phenomena. Heidelberg: Springer, 1975: 9–75.

    [17] Dong Y, Forster B C, Milne A K. Comparison of radar image degmentation by Gaussian-and Gamma-Markov random field models [J]. International Journal of Remote Sensing, 2003, 24 (4): 711–722.

    [18] Zhao Q H, Li Y, He X J, et al. Multi-look SAR image segmentation based on Voronoi tessellation technique and EM/MPM algorithm [J]. Journal of Remote Sensing, 2013, 17 (4): 841–854 (in Chinese).

    [19] Li Y, Hu H F, Zhao X M, et al. Fuzzy clustering algorithm for multi-look SAR image segmentation based on Gamma distribution with variable shape parameter [J]. Geomatics and Information Science of Wuhan University, 2018, 43 (7): 984–992 (in Chinese).

    [20] Zhao Q, Li Y, Liu Z, et al. SAR image segmentation using Voronoi tessellation and bayesian inference applied to dark spot feature extraction [J]. Sensors, 2013, 13 (11): 14484–14499.

    [21] Nithyakalyani S, Kumar S S. Voronoi fuzzy clustering approach for data processing inWSN [J]. International Journal of Computational Intelligence Systems, 2014, 7 (1): 105–113.

This Article


CN: 21-1124/TP

Vol 35, No. 07, Pages 1639-1644

July 2020


Article Outline



  • 0 Introduction
  • 1 Algorithm description
  • 2 Experiments and discussion
  • 3 Conclusions
  • References