自适应类别的层次高斯混合模型遥感影像分割

doi:10.3969/j.issn.0372-2112.2020.01.016

电子学报 ›› 2020, Vol. 48 ›› Issue (1): 131-136.DOI: 10.3969/j.issn.0372-2112.2020.01.016

自适应类别的层次高斯混合模型遥感影像分割

石雪, 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所, 辽宁阜新 123000

收稿日期:2018-12-10 修回日期:2019-08-11 出版日期:2020-01-25 发布日期:2020-01-25
作者简介:石雪女.1992年1月出生,辽宁人.2014年毕业于辽宁工程技术大学测绘工程系,现为硕博连读研究生,从事遥感影像处理有关的研究.E-mail:374636252@qq.com;李玉男.1963年3月出生,吉林人.教授、博士生导师.1984年在西北电讯工程学院获得学士学位,1991年在东南大学获得硕士学位,2006年在瑞尔森获得硕士学位,2010年在滑铁卢大学获得博士学位.现为辽宁工程技术大学教授,主要从事遥感数据处理理论与应用研究.E-mail:liyu@lntu.edu.cn
基金资助:
国家自然科学基金（No.41301479，No.41271435）；辽宁省自然科学基金（No.2015020090）

Remote Sensing Image Segmentation Based on Hierarchy Gaussian Mixture Model with Self-adaptive Number of Classes

SHI Xue, LI Yu, ZHAO Quan-hua

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

Received:2018-12-10 Revised:2019-08-11 Online:2020-01-25 Published:2020-01-25

摘要/Abstract

摘要： 为了实现自动确定类别数的高精度遥感影像分割，提出一种自适应类别的层次高斯混合模型（Hierarchical Gaussian Mixture Model，HGMM）遥感影像分割算法.提出算法采用多个高斯混合模型加权和定义HGMM，用于建模具有非对称，重尾和多峰等复杂特性的影像统计模型.采用期望最大化算法（Expectation Maximization，EM）求解模型参数.为了实现自动确定类别数，采用贝叶斯信息准则（Bayesian Information Criterion，BIC）求解最优类别数，其中惩罚项采用加权像素数定义.为了验证提出算法可行性和有效性，对模拟和全色遥感影像进行分割实验，并对分割结果进行定性和定量分析.结果表明HGMM具有准确建模复杂统计分布的能力，提出算法具有高精度和高效率，同时可自动确定最优类别数.

关键词: 高分辨率遥感图像分割, 层次高斯混合模型, 贝叶斯信息准则, 自适应类别, 期望最大化

Abstract: To accurately segment remote sensing image and automatically determine the number of classes,a Hierarchical Gaussian Mixture Model (HGMM) based remote sensing image segmentation algorithm with self-adaptive number of classes is proposed.In the proposed algorithm,HGMM is defined by the weighted sum of several Gaussian Mixture Models (GMM),which is used to model the asymmetric,heavy-tailed and multimodal distributions of image.Expectation Maximization (EM) is used to estimate the model parameters.Bayesian Information Criterion (BIC) is used to solve the optimal number of classes,where penalty term is defined by the weighted number of pixels.To test the feasibility and availability of the proposed algorithm,segmentation experiments are carried out on simulated and panchromatic remote sensing images.Segmentation results are analyzed qualitatively and quantitatively,which show that HGMM can accurately model the complicated statistical distributions.The proposed algorithm can obtain high accuracy and has good efficiency,as well as determine the optimal number of classes.

Key words: high resolution remote sensing image segmentation, hierarchical Gaussian mixture model, Bayesian information criterion, self-adaptive number of classes, expectation maximization

中图分类号:

TP751

石雪, 李玉, 赵泉华. 自适应类别的层次高斯混合模型遥感影像分割[J]. 电子学报, 2020, 48(1): 131-136.

SHI Xue, LI Yu, ZHAO Quan-hua. Remote Sensing Image Segmentation Based on Hierarchy Gaussian Mixture Model with Self-adaptive Number of Classes[J]. Acta Electronica Sinica, 2020, 48(1): 131-136.

参考文献

[1] 林雨准,张保明,王丹菂,等.多特征融合的高分辨率遥感影像建筑物分级提取[J].中国图象图形学报,2017,22(12):1798-1808. LIN Yu-zhun,ZHANG Bao-ming,WANG Dan-di,et al.Hierarchical building extraction from high resolution remote sensing imagery based on multi-feature fusion[J].Journal of Image and Graphics,2017,22(12):1798-1808.(in Chinese)
[2] 赵雪梅,李玉,赵泉华.结合高斯回归模型和隐马尔可夫随机场的模糊聚类图像分割[J].电子与信息学报,2014,36(11):2730-2736. ZHAO Xue-mei,LI Yu,ZHAO Quan-hua.Image segmentation by fuzzy clustering algorithm combining hidden Markov random field and Gaussian regression model[J].Journal of Electronics and Information Technology,2014,36(11):2730-2736.(in Chinese)
[3] MCLACHLAN G J,PEEL D.Finite Mixture Model[M].New York:John Wiley & Sons,2000.1-37.
[4] NGUYEN T M,WU M J.Fast and robust spatially constrained Gaussian mixture model for image segmentation[J].IEEE Transactions on Circuits and Systems for Video Technology,2013,23(4):621-635.
[5] 石雪,李玉,李晓丽,等.融入邻域作用的高斯混合分割模型及简化求解[J].中国图象图形学报,2017,22(12):1758-1768. SHI Xue,LI Yu,LI Xiao-li,et al.Gaussian mixture model with neighbor relationship for image segmentation and simplified solving method[J].Journal of Image and Graphics,2017,22(12):1758-1768.(in Chinese)
[6] 赵泉华,石雪,王玉,等.可变类空间约束高斯混合模型遥感图像分割[J].通信学报,2017,38(2):34-43. ZHAO Quan-hua,SHI Xue,WANG Yu,et al.Remote sensing image segmentation based on spatially constrained Gaussian mixture model with unknown class number[J].Journal on Communications,2017,38(2):34-43.(in Chinese)
[7] NGUYEN T M,WU Q M.Robust student' s-t mixture model with spatial constraints and its application in medical image segmentation[J].IEEE Transactions on Medical Imaging,2012,31(1):103-116.
[8] 赵泉华,李晓丽,赵雪梅,等.基于空间约束Student' s-T混合模型的模糊聚类图像分割[J].控制与决策,2016,31(11):2065-2070. ZHAO Quan-hua,LI Xiao-li,ZHAO Xue-mei,et al.Fuzzy clustering algorithm based on spatially constrained student' s-t mixture model for image segmentation[J].Control and Decision,2016,31(11):2065-2070.(in Chinese)
[9] ZHAO Q H,LI X L,and LI Y.Multilook SAR image segmentation with an unknown number of clusters using a Gamma mixture model and hierarchical clustering[J].Sensors,2017,17(5):1114.
[10] LI X L,ZHAO Q,and LI Y.Multilook SAR intensity image segmentation based on Voronoi tessellation and a Gamma mixture model[J].Remote Sensing Letters,2018,10(3):254-263.
[11] JI Z,HUANG Y,SUN Q,et al.A rough set bounded spatially constrained asymmetric Gaussian mixture model for image segmentation[J].Plos One,2017,12(1):697-708.
[12] NGUYEN T M,WU Q M,ZHANG H.Asymmetric mixture model with simultaneous feature selection and model detection[J].IEEE Transactions on Neural Networks and Learning Systems,2015,26(2):400-408.
[13] WANG Y,LI Y,ZHAO Q H.Segmentation of high resolution SAR image with unknown number of classes based on regular tessellation and RJMCMC algorithm[J].International Journal of Remote Sensing,2015,36(5):17.
[14] 王玉,李玉,赵泉华.利用RJMCMC算法的可变类SAR图像分割[J].信号处理,2014,30(10):1193-1203. Wang Yu,Li Yu,Zhao Quan-hua.SAR image segmentation with variable classes using RJMCMC algorithm[J].Journal of Signal Processing,2014,30(10):1193-1203.(in Chinese)
[15] SCHWARZ G.Estimating the dimension of a model[J].Annals of Statistics,1978,6:461-464.
[16] MEHRJOU A,HOSSEINI R,ARAABI B N.Improved Bayesian information criterion for mixture model selection[J].Pattern Recognition Letters,2016,69:22-27.
[17] BISHOP C M.Pattern Recognition and Machine Learning[M].New York:Springer Science and Business Media,2006.

自适应类别的层次高斯混合模型遥感影像分割

Remote Sensing Image Segmentation Based on Hierarchy Gaussian Mixture Model with Self-adaptive Number of Classes

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]	吴林煌, 苏凯雄, 王琳, 陈志峰, 陈平平. 正交匹配追踪和BIC准则的自适应双频段预失真模型优化算法[J]. 电子学报, 2018, 46(9): 2149-2156.
[2]	谢中华, 马丽红. 采用拉普拉斯尺度混合先验的结构化近似消息传递算法[J]. 电子学报, 2018, 46(3): 520-528.
[3]	张凯, 于宏毅, 胡赟鹏, 沈智翔. 基于EM-SBL迭代的稀疏SIMO信道频域盲均衡算法[J]. 电子学报, 2018, 46(2): 372-380.
[4]	薛高茹, 梁彦, 谯平, 杨柳青. 基于EM的主动段弹道导弹跟踪算法研究[J]. 电子学报, 2017, 45(7): 1770-1774.
[5]	孙国玺, 张清华, 文成林, 段志宏. 基于随机退化数据建模的设备剩余寿命自适应预测方法[J]. 电子学报, 2015, 43(6): 1119-1126.
[6]	王旭, 何凡, 冯辉, 杨涛, 胡波. MU-MIMO-OFDM上行频偏与信道联合跟踪[J]. 电子学报, 2014, 42(7): 1256-1261.
[7]	李晓旭, 李睿凡, 冯方向, 曹洁, 王小捷. 多视图有监督的LDA模型[J]. 电子学报, 2014, 42(10): 2040-2044.
[8]	张润生, 李艳斌, 李啸天. 基于合并分层聚类的网络拓扑推断算法[J]. 电子学报, 2013, 41(12): 2346-2352.
[9]	王小林;郭波;程志君. 融合多源信息的维纳过程性能退化产品的可靠性评估[J]. 电子学报, 2012, 40(5): 977-982.
[10]	程玉虎;仝瑶瑶;王雪松. 类相关性影响可变选择性贝叶斯分类器[J]. 电子学报, 2011, 39(7): 1628-1633.
[11]	李涛;冯大政;夏宇垠. 基于EM算法和GOF的宽带分布式目标检测算法[J]. 电子学报, 2010, 38(10): 2246-2250.
[12]	吴昊, 郁文贤, 匡纲要, 李智勇. 一种模型选择优化准则及其在高光谱图像非监督分类中的应用[J]. 电子学报, 2003, 31(S1): 2154-2157.