结合双树复小波纹理特征和MRF模型的遥感图像分割

doi:10.13474/j.cnki.11-2246.2019.0315

测绘通报 ›› 2019, Vol. 0 ›› Issue (10): 40-45.doi: 10.13474/j.cnki.11-2246.2019.0315

结合双树复小波纹理特征和MRF模型的遥感图像分割

韦春桃¹, 赵平^1,2, 肖博林¹, 白风¹, 李小勇¹, 杨晚芸^1,3

1. 重庆交通大学, 重庆 400074;
2. 中国电建集团贵州电力设计研究院有限公司, 贵州贵阳 550002;
3. 中核咨询有限公司四川分公司, 四川乐山 614000

收稿日期:2019-01-23 修回日期:2019-08-06 出版日期:2019-10-25 发布日期:2019-10-26
作者简介:韦春桃(1968-),女,博士,教授,主要从事影像处理与应用研究。E-mail:269276944@qq.com
基金资助:
重庆市基础科学与前沿技术研究专项重点项目（cstc2015jcyjBX0023）

Remote sensing image segmentation method based on dual-tree complex wavelet texture feature and MRF model

WEI Chuntao¹, ZHAO Ping^1,2, XIAO Bolin¹, BAI Feng¹, LI Xiaoyong¹, YANG Wanyun^1,3

1. Chongqing Jiaotong University, Chongqing 400074, China;
2. Guizhou Electric Power Design & Research Institute Co., Ltd., Guiyang 550002, China;
3. Sichuan Branch, China Nuclear Consulting Co., Ltd., Leshan 614000, China

Received:2019-01-23 Revised:2019-08-06 Online:2019-10-25 Published:2019-10-26

摘要/Abstract

摘要： 针对经典的小波纹理不能准确地表达影像纹理特征的问题，以及影像分割结果缺少对像元空间相关性和分布关系的考虑。本文提出了结合双树复小波（DT-CWT）纹理和马尔可夫随机场（MRF）模型的高分辨率遥感影像分割方法。首先，通过双树复小波变换提取影像纹理特征，联合光谱特征形成表达影像信息的混合特征向量；然后，将混合特征向量高斯归一化处理，并用K-means聚类的方法对特征空间中的混合特征向量聚类得到初始分割图；最后，借助马尔可夫随机场模型在初始分割结果中引入上下文信息，基于贝叶斯最大后验概率准则得到最终的分割结果。本文通过双树复小波纹理提高了特征表达的准确度，同时使用马尔可夫随机场模型减弱了分割结果中同质区域的“椒盐噪声”，从而进一步提高了高分辨率遥感影像分割的精度。

关键词: 双树复小波变换, 纹理特征, 高斯归一化, 马尔可夫随机场, 高分辨率遥感影像

Abstract: For the classic wavelet textures, the image texture features cannot be accurately expressed. The image segmentation results lack the consideration of the spatial correlation and distribution relationship of the pixels. In this paper, we proposes a combination of dual-tree complex wavelet (DT-CWT) texture and Markov random field (MRF) model for high resolution remote sensing image segmentation method. Firstly, the image texture feature is extracted by dual-tree complex wavelet transform, and the mixed feature vector of the expression image is formed by combining the texture feature and the spectral feature. Then, Gaussian normalization of the mixed feature vectors in the feature space is performed. The K-means clustering method is used to perform the feature vectors in the feature space. The initial segmentation map is obtained by clustering. Finally, using the Markov random field model to introduce context information to represent the initial segmentation results, the initial segmentation results are optimized based on the Bayesian maximum a posteriori probability criterion, and the final segmentation results are obtained. This paper improves the accuracy of feature expression by using dual-tree complex wavelet textures. At the same time, the Markov random field model is used to weaken the "pepper and salt noise" in the homogenous region of the segmentation results, further improving the segmentation accuracy of the high resolution remote sensing image.

Key words: dual-tree complex wavelet transform(DT-CWT), texture feature, Gaussian normalization, Markov random field, high resolution remote sensing image

中图分类号:

P237

韦春桃, 赵平, 肖博林, 白风, 李小勇, 杨晚芸. 结合双树复小波纹理特征和MRF模型的遥感图像分割[J]. 测绘通报, 2019, 0(10): 40-45.

WEI Chuntao, ZHAO Ping, XIAO Bolin, BAI Feng, LI Xiaoyong, YANG Wanyun. Remote sensing image segmentation method based on dual-tree complex wavelet texture feature and MRF model[J]. 测绘通报, 2019, 0(10): 40-45.

参考文献 15

[1]	张建廷,张立民.结合光谱和纹理的高分辨率遥感图像分水岭分割[J].武汉大学学报(信息科学版),2017,42(4):449-455,467.
[2]	WANG D, CUI X, XIE F, et al. Multi-feature sea-land segmentation based on pixel-wise learning for optical remote-sensing imagery[J]. International Journal of Remote Sensing, 2017, 38(15):4327-4347.
[3]	巫兆聪,胡忠文,张谦,等.结合光谱、纹理与形状结构信息的遥感影像分割方法[J].测绘学报,2013,42(1):44-50.
[4]	LI Zhenxuan, SHI Wenzhong, HAO Ming, et al. Unsuper-vised change detection using spectral features and a texture difference measure for VHR remote-sensing images[J]. International Journal of Remote Sensing, 2017, 38(23):7302-7315.
[5]	MIRZAPOUR F, GHASSEMIAN H. Improving hyperspe-ctral image classification by combining spectral, texture, and shape features[J]. International Journal of Remote Sensing, 2015, 36(4):1070-1096.
[6]	MAJDAR R S, GHASSEMIAN H. A probabilistic SVM approach for hyperspectral image classification using spectral and texture features[J]. International Journal of Remote Sensing, 2017, 38(15):4265-4284.
[7]	ZHANG Y.Optimisation of building detection in satellite images by combining multispectral classification and texture filtering[J]. ISPRS Journal of Photogrammetry and Remote Sensing,1999,54:50-60.
[8]	MYINT S W, LAM N, TYLER J. Wavelets for urban spatial feature discrimination:comparisons with fractal, spatial autocorrelation, and spatial co-occurrence approaches[J].Photogrammetric Engineering and Remote Sensing, 2004,70(7):803-812.
[9]	黄昕,张良培,李平湘.基于小波的高分辨率遥感影像纹理分类方法研究[J].武汉大学学报(信息科学版),2006,31(1):66-69.
[10]	KINGSBURY N G. Complex wavelets for shift invariant analysis and filtering of signals[J].Applied and Computation Harmonic Analysis, 2001, 10(3):234-253.
[11]	KENNEL P, FIORIO C, BORNE F. Supervised image segmentation using q-shift dual-tree complex wavelet transform coefficients with a texton approach[J]. Pattern Analysis & Applications, 2011, 18(2):1-11.
[12]	鲍蕊,薛朝辉,张像源,等.综合聚类和上下文特征的高光谱影像分类[J].武汉大学学报(信息科学版),2017, 42(7):890-896.
[13]	HONG L, GAO Z, PAN X, et al. Segmentation of high resolution remote sensing image based on hierarchically multiscale object-oriented Markov random fields model[C]//IEEE International Conference on Spatial Data Mining and Geographical Knowledge Services.[S.l.]:IEEE, 2011:343-347.
[14]	ZHENG C, ZHANG Y, WANG L. Multilayer semantic segmentation of remote-sensing imagery using a hybrid object-based Markov random field model[J]. International Journal of Remote Sensing, 2016, 37(23):5505-5532.
[15]	洪亮. 基于对象马尔可夫模型的高分辨率遥感影像分割方法研究[D]. 武汉:武汉大学, 2010.

结合双树复小波纹理特征和MRF模型的遥感图像分割

Remote sensing image segmentation method based on dual-tree complex wavelet texture feature and MRF model

可视化

摘要/Abstract

引用本文

使用本文

参考文献 15

相关文章 15

编辑推荐

Metrics

本文评价

[1]	郁宗桥, 王育红, 刘文宋, 左雨芳, 冯锋. 顾及GF-3影像纹理特征的洪涝范围提取方法[J]. 测绘通报, 2022, 0(5): 26-31.
[2]	潘建平, 付占宝, 程卫华, 陆世安, 李鑫, 孙博文, 黄桂萍. 基于地表类型的山地海绵城市生态建设评估[J]. 测绘通报, 2022, 0(5): 84-88.
[3]	陈智朗, 付振华, 朱紫阳, 王慧慧, 刘沁雯, 杨钰灵, 许耿然. 基于HRNet的高分辨率遥感影像建筑物变化信息提取[J]. 测绘通报, 2022, 0(5): 126-132.
[4]	兰峰, 张帆, 高云龙, 黄先锋. 顾及地物类别的倾斜摄影三维模型简化方法[J]. 测绘通报, 2022, 0(4): 44-50.
[5]	杨素妨, 杜林. 多特征流形鉴别嵌入的高分辨率遥感影像分类[J]. 测绘通报, 2021, 0(9): 37-42.
[6]	耿万轩, 周维勋, 金双根. 基于LDCNN特征提取的多核SVM高分辨率遥感影像场景分类[J]. 测绘通报, 2021, 0(8): 14-21.
[7]	聂倩, 七珂珂, 赵艳福. 融入超像素分割的高分辨率影像面向对象分类[J]. 测绘通报, 2021, 0(6): 44-49.
[8]	陈利燕, 林鸿, 吴健华. 融合随机森林和超像素分割的建筑物自动提取[J]. 测绘通报, 2021, 0(2): 49-53.
[9]	刘明轩, 张黎明, 王昊, 马文骏, 李玉. 运用ULBP的高分辨率遥感影像感知哈希完整性认证算法[J]. 测绘通报, 2021, 0(12): 22-27.
[10]	王媛, 杜明义, 杨柳忠, 蔡国印, 张宁. 多源遥感数据支持下的城市生态状况评价[J]. 测绘通报, 2021, 0(11): 16-20.
[11]	迪里胡玛尔·阿汗木江, 玉素甫江·如素力. 新疆焉耆盆地土壤湿度时空分布特征[J]. 测绘通报, 2021, 0(11): 36-41.
[12]	陈根良, 董胜光. 面向地理国情监测的房屋建筑区树冠覆盖多特征协同提取[J]. 测绘通报, 2021, 0(10): 73-77.
[13]	陈世荣, 申佩佩, 包颖. 一种适用于粗匹配的遥感影像快速变化发现方法[J]. 测绘通报, 2021, 0(1): 90-93,98.
[14]	常京新, 高贤君, 杨元维, 王双喜. 融合多时相高分影像的建筑物轮廓优化方法[J]. 测绘通报, 2020, 0(7): 112-115.
[15]	谢梦, 刘伟, 李二珠, 杨梦圆, 王晓檀. 深度卷积神经网络支持下的遥感影像语义分割[J]. 测绘通报, 2020, 0(5): 36-42.