Abstract To improve anti-noise capability of image segmentation and take full advantage of energy functions in feature and label fields, a regional image segmentation based on energy functions for double random fields is proposed. Firstly, an image domain is partitioned into a set of sub-regions by a geometry tessellation technique. Based on the tessellation, negative logarithm of multi-Gaussian probability distribution is employed to define regionalized feature field energy function to describe the homogeneity of statistical distribution for pixel colors in a homogeneous region. The improved Potts model is the extension of the traditional model for the labels of a pixel and its neighbor pixels, and it is used to define regionalized label field energy function to characterize the relativity of the labels for sub-regions. Combine feature field and label field, and Kullback-Leibler divergence is utilized to define the heterogeneous energy function for describing the heterogeneity of color distributions among different homogeneous regions. The unconditional Gibbs function is adopted to transform the defined energy functions into probability functions for image segmentation. Finally, based on the maximization of probability distribution scheme, Metropolis-Hastings sampler is designed to obtain the optimal segmentation. Synthetic, remote sensing and natural texture images are segmented by several algorithms. Segmentation results show the proposed algorithm realizes image segmentation accurately and efficiently.
［1］ 赵雪梅,李 玉,赵泉华.基于隐马尔可夫高斯随机场模型的模糊类高分辨率遥感影像分割算法.电子学报, 2016, 44(3): 679-686.（ZHAO X M, LI Y, ZHAO Q H. Hidden Markov Gaussian Random Field Based Fuzzy Clustering Algorithm for High Resolution Remote Sensing Image Segmentation. Acta Electronica Sinica, 2016, 44(3): 679-686.） ［2］ ZENG S, HUANG R, KANG Z, et al. Image Segmentation Using Spectral Clustering of Gaussian Texture Models. Neurocomputing, 2004, 144(11): 346-356. ［3］ GEMAN S, GEMAN N. Stochastic Relaxation, Gibbs Distributions and the Bayesian Restoration of Images. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1984, 6(6): 721-741. ［4］ FREEDMAN D, ZHANG T. Active Contours for Tracking Distribution. IEEE Transactions on Image Processing, 2004, 13(4): 518-526. ［5］ 赵泉华,李 玉,何晓军.结合EM/MPM算法和Voronoi划分的图像分割方法.信号处理, 2013, 29(4): 503-512. (ZHAO Q H, LI Y, HE X J. Combining the EM/MPM and Voronoi Tessellation for Image Segmentation. Journal of Signal Processing, 2013, 29(4): 503-512.) ［6］ 曹容菲,张美霞,王醒策,等.基于高斯-马尔科夫随机场模型的脑血管分割算法研究.电子与信息学报, 2014, 36(9): 2053-2060. (CAO R F, ZHANG M X, WANG X C, et al . A Novel Cerebrovascular Segmentation Algorithm Based on Gauss Markov Random Field Model. Journal of Electronics & Information Technology, 2014, 36(9): 2053-2060. ［7］ SCIANNA M, PREZIOSO L. A Cellular Potts Model for the MMP Dependent and Independent Cancer Cell Migration in Matrix Microtracks of Different Dimensions. Computational Mechanics, 2014, 53(3): 485-497. ［8］ YAO Z J, LIU W Y. Extracting Robust Distribution Using Adaptive Gaussian Mixture Model and Online Feature Selection. Neurocompu ting, 2013, 101: 258-274. ［9］ KUMAR M P, TURKI H, PRESTON D, et al. Parameter Estimation and Energy Minimization for Region Based Semantic Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(7): 1373-1386. ［10］LI S Z. Markov Random Field Modeling in Image Analysis. London, UK: Springer, 2009. ［11］ LI L H, ZHANG M L, WU Y Z, et al. Application of Adaptive MRF Based on Region in Segmentation of Microscopic Image. Wseas Transactions on Systems, 2013, 12(4): 240-249. ［12］ 王相海,王金玲,方玲玲.灰度差能量函数引导的图像分割自适应C V 模型.模式识别与人工智能, 2015, 28(3)： 214-222. (WANG X H, WANG J L, FANG L L. An Adaptive C V Image Segmentation Model Guided by Gray Difference Energy Function. Pattern Recognition and Artificial Intelligence, 2015, 28(3): 214-222.) ［13］ 孙 巍.基于图割的交互式图像分割技术研究.硕士学位论文.西安:陕西师范大学, 2015. (SUN W. Study on the Technology of Interactive Image Segmentation Based on Graph Cuts. Master Dissertation. Xi′an, China: Shaanxi Normal University, 2015.) ［14］ SCHERRER B, DOJAT M, FORBES F, et al. Agentification of Markov Model Based Segmentation: Application to Magnetic Resonance Brain Scans. Artificial Intelligence in Medicine, 2009, 46(1): 81-95. ［15］ DRYDEN I L, SCARR M R, TAYLOR C C. Bayesian Texture Segmentation of Weed and Crop Image Using Reversible Jump Markov Chain Monte Carlo Methods. Journal of the Royal Statistical Society(Series C), 2003, 52(1): 31-50. ［16］ MOHAMMAD DIAFARI A. Entropy, Information Theory, Information Geometry and Bayesian Inference in Data, Signal and Image Processing and Inverse Problems. Entropy, 2015, 17(6): 3989-4027. ［17］ SHUTIN D, ZLOBINSKAYA O. Application of Information Theore tic Measures to Quantitative Analysis of Immunofluorescent Microscope Imaging. Computer Methods and Programs in Biomedicine, 2010, 97(2): 114-129. ［18］ MACKAY D J C. Information Theory, Inference, and Learning Algorithms . Cambridge, UK: Cambridge University Press, 2003. ［19］ GEMAN D, GEMAN S, GRAFFIGNE C, et al. Boundary Detection by Constrained Optimization. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990, 12(7): 609-628. ［20］ KATO Z. Segmentation of Color Images via Reversible Jump MCMC Sampling. Image and Vision Computing, 2008, 26(3): 361-371. ［21］KREYSZIG E. Advanced Engineering Mathematics. New York, USA: John Wiley & Sons, 1999. ［22］ KRIKPATRICK S, GELLATT C D, VECCHI M P. Optimization by Simulated Annealing. Science, 1983, 220(4598): 671-680. ［23］ 赵泉华,李 玉,何晓军.结合几何划分技术和最大期望值/最大边缘概率算法的彩色图像分割.中国图象图形学报, 2013, 18(10): 1270-1278. (ZHAO Q H, LI Y, HE X J. Color Image Segmentation Using Geometry Tessellation Technique and EM/MPM Algorithm. Journal of Image and Graphics, 2013, 18(10): 1270-1278.)
2017, Vol. 30 
