基于隐马尔可夫高斯随机场模型的模糊聚类高分辨率遥感影像分割算法

doi:10.3969/j.issn.0372-2112.2016.03.028

电子学报 ›› 2016, Vol. 44 ›› Issue (3): 679-686.DOI: 10.3969/j.issn.0372-2112.2016.03.028

基于隐马尔可夫高斯随机场模型的模糊聚类高分辨率遥感影像分割算法

赵雪梅, 李玉, 赵泉华

辽宁工程技术大学测绘与地理科学学院遥感科学与应用研究院, 辽宁阜新 123000

收稿日期:2014-06-19 修回日期:2014-09-12 出版日期:2016-03-25
- 作者简介:
- 赵雪梅 女,1989年生,辽宁阜新人,辽宁工程技术大学,博士在读.研究方向为空间统计学、模糊数学在遥感图像处理中的应用. E-mail:374010101@qq.com;李玉男,1963年生,吉林长春人,辽宁工程技术大学,博士,教授.研究方向为遥感数据处理. E-mail:liyu@lntu.edu.cn;赵泉华 女,1978年生,河北承德人,辽宁工程技术大学,博士,副教授.研究方向为遥感图像建模与分析. E-mail:zqhlby@163.com
- 基金资助:
- 国家自然科学基金 (No.41271435,No.41301479); 中华环境保护基金会"123工程" (No.CEPF-2013-123-1-3)

Hidden Markov Gaussian Random Field Based Fuzzy Clustering Algorithm for High-Resolution Remote Sensing Image Segmentation

ZHAO Xue-mei, LI Yu, ZHAO Quan-hua

Institute for Remote Sensing Science and Application, School of Geomatics, Liaoning Technical University, Fuxin, Liaoning 123000, China

Received:2014-06-19 Revised:2014-09-12 Online:2016-03-25 Published:2016-03-25

摘要/Abstract

摘要：

本文利用隐马尔可夫随机场和高斯模型分别建立标号场和特征场的邻域关系,提出了基于隐马尔可夫高斯随机场模型的模糊聚类分割算法.该算法用隐马尔可夫随机场模型定义先验概率,并将该先验概率作为尺度控制因子引入到KL(Kullback-Lerbler)信息中,在目标函数的定义中,KL信息作为规则化项,其系数表示算法的模糊程度.在基于高斯模型的后验概率中,像素相关性被定义在空间和谱间,并用该概率的负对数值表征像素点到聚类中心的非相似性测度.通过对合成遥感影像和高分辨率遥感影像进行分割实验,证明了算法的有效性和普适性.

关键词: 遥感影像分割, 隐马尔可夫随机场, 高斯模型, 模糊C均值算法

Abstract:

We establish neighbor relationships on both label field and feature field,and proposed an algorithm called the hidden Markov Gaussian random field fuzzy c-means(HMGRF-FCM) algorithm.In this algorithm,Markov theory is used to define a prior distribution which is introduced to the KL(Kullback-Lerbler) information and acts as a variable which controls the cluster size.KL information is used as a regularization item in the objective function and its coefficients can express the fuzziness of the algorithm.Besides,the posterior distribution is defined by Gaussian model which contains not only neighbor relationships in spatial space but also that in spectral space.Negative log posterior distribution function can express the dissimilarity between pixels and the cluster center,so it is defined as the dissimilarity measure.The algorithm is described in d-dimensional case,so we do some segmentation experiments on 3-dimensional synthetic remote sensing image,3-dimensional high resolution IKONOS pan-sharpen images,and 4-dimensional SPOT-5 images to prove its accuracy and universality.

Key words: remote sensing image segmentation, hidden Markov random field(HMRF), Gaussian model, fuzzy c-means(FCM) algorithm

中图分类号:

TP751

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聚类高分辨率遥感影像分割算法[J]. 电子学报, 2016, 44(3): 679-686.

ZHAO Xue-mei, LI Yu, ZHAO Quan-hua. Hidden Markov Gaussian Random Field Based Fuzzy Clustering Algorithm for High-Resolution Remote Sensing Image Segmentation[J]. Acta Electronica Sinica, 2016, 44(3): 679-686.

参考文献

[1] 林剑,鲍光淑,王润生,王欣.基于模糊密度分解的遥感图像光谱和纹理信息的融合[J]. 电子学报,2004,32(12):2028-2030. Lin J,Bao G S,Wagn R S,Wang X.Fusion spectrum and texture information of RS image based on decomposing fuzzy density[J]. Acta Electronica Sinica,2004,32(12):2028-2030.(in Chinese)
[2] 魏飞鸣,李小文,顾行发,余涛,孙源.基于形状参数的遥感图像"同谱异物"目标区分[A]. 第十四届全国图象图形学学术会议[C]. 福州:中国图象图形学会,2008.466-470.
[3] 赵维兵.中巴地球资源卫星图像中伪"同谱异物"区分方法研究[J]. 遥测遥控,2008,29(2):19-25. Zhao W B.Research on the method of differentiating false"different objects same image" in the CBERS image[J]. Journal of Telemetry,Tracking and Command,2008,29(2):19-25.(in Chinese)
[4] 丁震,胡钟山,杨静宇,唐振民.FCM算法用于灰度图像分割的研究[J]. 电子学报,1997,25(5):39-43. Ding Z,Hu Z S,Yang J Y,Tang Z M.FCM algorithm for the research of intensity image segmentation[J]. Acta Electronica Sinica,1997,25(5):39-43.(in Chinese)
[5] 裴继红,谢维信.直方图模糊约束FCM聚类自适应多阈值图像分割[J]. 电子学报,1999,27(10):38-42. Pei J H,Xie W X.Adaptive multi-thresholds image segmentation based on fuzzy restrainted histogram FCM clustering[J]. Acta Electronica Sinica,1999,27(10):38-42.(in Chinese)
[6] 王向阳,王春花.基于特征散度的自适应FCM图像分割算法[J]. 中国图象图形学报,2008,13(5):906-910. Wang X Y,Wang C H.An adaptive FCM image segmentation algorithm based on the feature divergence[J]. Journal of Image and Graphics,2008,13(5):906-910.(in Chinese)
[7] Bezdek J.Pattern Recognition with Fuzzy Objective Function Algorithms[M]. New York:Plenum,1981.79-88.
[8] Ahmed M,Yamany S,Mohamed N,Farag A,Moriarty T.A modified fuzzy c-means algorithm for bias field estimation and segmentation of MRI data[J]. IEEE Transactions on Medical Imaging,2002,21(3):193-199.
[9] Szilagyi L,Benyo Z,Szilagyii,Adam H.MR brain image segmentation using an enhanced fuzzy C-means algorithm[A]. Proceedings of the 25th Annual International Conference of the IEEE[C]. Cancun,Mexico:IEEE,2003.724-726.
[10] Cai W,Chen S,Zhang D.Fast and robust fuzzy c-means clustering algorithms incorporating local information for image segmentation[J]. Pattern Recognition,2007,40(3):825-838.
[11] Krinidis Sm Chatzis V.A robust fuzzy local information c-means clustering algorithm[J]. IEEE Transactions on Image Processing,2010,19(5):1328-1337.
[12] Chatzis S P,Varvarigou T A.A fuzzy clustering approach toward hidden Markov random field models for enhanced spatially constrained image segmentation[J]. IEEE Transactions on Fuzzy System,2008,16(5):1351-1361.
[13] Miyamoto S,Mukaidono M.Fuzzy c-means as a regularization and maximum entropy approach[A]. Proceedings of the 7th IFSA Conference[C]. Prague:University of Economics,1997.86-92.
[14] Ichihashi H,Miyagishi K,Honda K.Fuzzy c-means clustering with regularization by K-L information[A]. Proceedings of the 10th IEEE International Conference on Fuzzy System[C]. Mrlbourne:IEEE,2001.924-927.
[15] Panjwani D K,Healey G.Markov random field models for unsupervised segmentation of textured color image[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995,17(10):939-954.
[16] 于剑,程乾生.关于FCM算法中的权重指数m的一点注记[J]. 电子学报,2003,31(3):478-480. Yu J,Cheng Q S.A note on the weighting exponent m in FCM algorithm[J]. Acta Electronica Sinica,2003,31(3):478-480.(in Chinese)
[17] 宫改云,高新波,伍忠东.FCM聚类算法中模糊加权指数m的优选方法[J]. 模糊系统与数学,2005,19(1):143-148. Gong G Y,Gao X B,Wu Z D.An optimal choice method of parameter m in FCM clustering algorithm[J]. Fuzzy Systems and Mathematics,2005,19(1):143-148.(in Chinese)
[18] Zhao Y X,Atlas L,Zhuang S.Application of the Gibbs distribution to hidden Markov modeling in isolated word recognition[A]. International Conference on Acoustics,Speech,and Signal Processing[C]. New York:IEEE,1988.28-31.
[19] Clifford P.Markov random fields in statistics[J]. Disorder in Physical Systems,1990,a Volume in Honor of John M Hammersley:19-32.
[20] Zhang J.The mean field theory in EM procedures for Markov random fields[J]. IEEE Transactions on Signal Processing,1992,40(10),2570-2583.
[21] Yuan J Y,Wang D L,Li R.Remote sensing image segmentation by combining spectral and texture features[J]. IEEE Transactions on Geoscience and Remote Sensing,2014,52(1),16-24.
[22] Wang M,Li R.Segmentation of high spatial resolution remote sensing imagery based on hard-boundary constraint and two-stage merging[J]. IEEE Transactions on Geoscience and Remote Sensing,2014,52(9),5712-5725.

基于隐马尔可夫高斯随机场模型的模糊聚类高分辨率遥感影像分割算法

Hidden Markov Gaussian Random Field Based Fuzzy Clustering Algorithm for High-Resolution Remote Sensing Image Segmentation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 12

编辑推荐

Metrics

[1]	陈荣元, 徐雪松, 申立智, 刘跃华, 陈浪. 模糊区域级MRF方法在城镇自动识别中的应用[J]. 电子学报, 2019, 47(2): 489-494.
[2]	詹曙, 方琪, 杨福猛, 常乐乐, 闫婷. 基于耦合特征空间下改进字典学习的图像超分辨率重建[J]. 电子学报, 2016, 44(5): 1189-1195.
[3]	王荔霞, 谢维信, 裴继红. 多高斯模型特征空间覆盖学习的海洋航摄图像分割[J]. 电子学报, 2014, 42(10): 2117-2122.
[4]	刘宗香, 谢维信, 王品, 余友. 一种具有信息保持能力的GM-PHD滤波器[J]. 电子学报, 2013, 41(8): 1603-1608.
[5]	何继爱, 裴承全, 郑玉峰. 稳定分布下基于FAM的低阶循环谱算法研究[J]. 电子学报, 2013, 41(7): 1297-1304.
[6]	王星东, 熊章强, 李新武, 梁雷. 基于改进的小波变换的南极冰盖冻融探测[J]. 电子学报, 2013, 41(2): 402-406.
[7]	贾建;陈莉. 基于正态逆高斯模型的非下采样Contourlet变换图像去噪[J]. 电子学报, 2011, 39(7): 1563-1568.
[8]	凤宏晓;侯彪;焦李成;卜晓明. 基于非下采样Contourlet域局部高斯模型和MAP的SAR图像相干斑抑制[J]. 电子学报, 2010, 38(4): 811-0816.
[9]	陈姝, 邹北骥, 彭小宁, 杨明. 应用粒子滤波及先验概率模型进行图像分割的新算法[J]. 电子学报, 2007, 35(8): 1533-1537.
[10]	董红斌, 黄厚宽, 周成, 何军, 尚文倩. 基于模糊权和有效性函数的演化聚类算法[J]. 电子学报, 2007, 35(5): 964-970.
[11]	李文斌, 刘椿年, 陈嶷瑛. 基于混合高斯模型的电子邮件多过滤器融合方法 [J]. 电子学报, 2006, 34(2): 247-251.
[12]	岳博;焦李成. 混合因子分析的重新抽样方法[J]. 电子学报, 2002, 30(12): 1873-1875.