历史解译知识引导下组合遥感图谱特征的变化检测方法

doi:10.3724/SP.J.1047.2016.00655

地球信息科学学报 ›› 2016, Vol. 18 ›› Issue (5): 655-663.doi: 10.3724/SP.J.1047.2016.00655

• 遥感大数据协同计算方法 • 上一篇下一篇

历史解译知识引导下组合遥感图谱特征的变化检测方法

吴田军^1,²(), 马江洪¹

1. 长安大学理学院数学与信息科学系,西安 710064
2. 浙江省海洋大数据挖掘与应用重点实验室,舟山 316022

收稿日期:2015-12-15 修回日期:2016-03-11 出版日期:2016-05-10 发布日期:2016-05-10
作者简介:
作者简介：吴田军（1986-）,男,博士,讲师,研究方向为遥感信息计算。E-mail:wutianjun1986@163.com
基金资助:
浙江省海洋大数据挖掘与应用重点实验室开放课题(OBDMA201508);国家高分辨率对地观测系统重大专项(03-Y30B06-9001-13/15-01);中国科学院重点部署项目(KZZD-EW-07-01);长安大学中央高校基本科研业务费专项资金基础研究项目(310812161010);国家自然科学基金项目(11261044)

A Remote Sensing Change Detection Method Combining with Spatial-Spectral Features under the Guidance of Historical Interpretation Knowledge

WU Tianjun^1,^2,^*(), MA Jianghong¹

1. Department of Mathematics and Information Science, College of Science, Chang'an University, Xi'an 710064
2. Key Laboratory of Oceanographic Big Data Mining & Application of Zhejiang Province, Zhoushan 316022, China

Received:2015-12-15 Revised:2016-03-11 Online:2016-05-10 Published:2016-05-10
Contact: WU Tianjun E-mail:wutianjun1986@163.com

摘要/Abstract

摘要：

高空间分辨率遥感影像为地表变化监测提供了大量细节信息,这使得基于高分辨率影像的变化检测技术成为当前遥感领域的研究热点之一。本文提出了一种历史解译知识引导下组合遥感图谱特征的变化检测方法。首先,通过分割前后时相的组合影像构建空间位置一致的对象,并在提取对象光谱和纹理特征后,引入前期土地覆盖专题图指导2类图谱特征相似度的DS证据融合;然后,利用其历史存档图斑所属区域的优势地类标签指示不同特征相似度的证据差异融合;最后,基于GMM（Gaussian Mixture Mode）模型的二值化方法提取最终的变化区域。实验结果表明,该方法能充分利用历史解译知识指导不同时相高分辨率影像对象特征相似度组合,一定程度上提高了变化检测正确率。

关键词: 高空间分辨率遥感, 面向对象变化检测, 图谱特征组合, DS证据理论, 高斯混合模型

Abstract:

Remote sensing images with high-spatial-resolution can provide a great amount of detailed information for land resource monitoring, which makes the change detection using the high-spatial-resolution remote sensing images become the focus of current remote sensing research. This article proposes a remote sensing change detection method through combining with the spatial-spectral features under the guidance of historical interpretation knowledge. Firstly, the objects of consistent spatial position are constructed by segmenting the combined image from different phases. After that, the objects’ spectral and texture features are extracted. Then, we introduce the previous land cover maps to guide the DS evidence fusion for the similarities of two different types of features. The archived class-type label in each polygon from the land cover maps is used to direct the similarity fusion with different evidence confidences. Finally, a binarization method based on the Gaussian Mixture Model is adopted to extract the change regions. Experimental results show that this method can take advantage of the historical interpretation knowledge to guide the fusion of different object-feature-similarities; moreover, to a certain extent, to effectively improve the accuracy of change detection.

Key words: high-spatial-resolution remote sensing, object-based change detection, spatial-spectral features combination, DS Evidence Theory, Gaussian Mixture Model

吴田军, 马江洪. 历史解译知识引导下组合遥感图谱特征的变化检测方法[J]. 地球信息科学学报, 2016, 18(5): 655-663.DOI:10.3724/SP.J.1047.2016.00655

WU Tianjun,MA Jianghong. A Remote Sensing Change Detection Method Combining with Spatial-Spectral Features under the Guidance of Historical Interpretation Knowledge[J]. Journal of Geo-information Science, 2016, 18(5): 655-663.DOI:10.3724/SP.J.1047.2016.00655

图/表 8

图1

表1

图2

图3

表2

表3

表4

表5

参考文献 21

[1]	Chen G, Hay G J, Carvalho L M T, et al. Object-based change detection[J]. International Journal of Remote Sensing, 2012,33(14):4434-4457.
[2]	Listner C, Niemeyer I.Recent advances in object-based change detection[C]. Proceedings of the 2011 IEEE International Geoscience and Remote Sensing Symposium, 2011:110-113.
[3]	劳小敏. 基于对象的高分辨率遥感影像土地利用变化检测技术研究[D].杭州:浙江大学,2013.
	[ Lao X M.The research of object-based high resolution remote sensing land use change detection[D]. Hangzhou: Zhejiang University, 2013. ]
[4]	祝锦霞. 高分辨率遥感影像变化检测的关键技术研究[D].杭州:浙江大学,2011.
	[ Zhu J X.The key technology of change detection using high resolution remotely sensed imagery[D]. Hangzhou: Zhejiang University, 2013. ]
[5]	Durieux L, Lagabrielle E, Nelson A.A method for monitoring building construction in urban sprawl areas using object-based analysis of Spot 5 images and existing GIS data[J]. Journal of Photogrammetry and Remote Sensing, 2008,63(4):399-408. doi: 10.1016/j.isprsjprs.2008.01.005
[6]	Li D.Remotely sensed images and GIS data fusion for automatic change detection[J]. International Journal of Image and Data Fusion, 2010,1(1):99-108. doi: 10.1080/19479830903562074
[7]	Rozenstein O, Karnieli A.Comparison of methods for land-use classification incorporating remote sensing and GIS inputs[J]. Applied Geography Climate Change and Applied Geography: Place, Policy, and Practice, 2011,31(2):533-544. doi: 10.1016/j.apgeog.2010.11.006
[8]	李雪,舒宁,李亮,等.基于GIS辅助数据的样本像斑变化检测方法研究[J].计算机工程与应用,2010,46(14):215-217. doi: 10.3778/j.issn.1002-8331.2010.14.064
	[ Li X, Shu N, Li L, et al.Research of change detection using GIS auxiliary data and samples[J]. Computer Engineering and Applications, 2010,46(14):215-217. ] doi: 10.3778/j.issn.1002-8331.2010.14.064
[9]	刘锋. 基于单时相遥感影像与GIS的土地利用变化检测方法研究[D].北京:中国测绘科学研究院,2006.
	[ Liu F.Research on land use change detection method based on the single-temporal RS imagery and GIS technology[D]. Beijing: Chinese Academy of Surveying and Mapping, 2006. ]
[10]	王琰. 基于像斑统计分析的高分辨率遥感影像土地利用/覆盖变化检测方法研究[D].武汉:武汉大学,2012.
	[ Wang Y.The research of land use/cover change detection using high resolution remote sensing images based on image segments statistical analysis[D]. Wuhan: Wuhan University, 2012. ]
[11]	Dianat R, Kasaei S.Dimension reduction of remote sensing images by incorporating spatial and spectral properties[J]. AEU-International Journal of Electronics and Communications, 2010,64(8):729-732. doi: 10.1016/j.aeue.2009.10.001
[12]	高伟. 基于特征知识库的遥感信息提取技术研究[D].武汉:中国地质大学,2010.
	[ Gao W.The study of information extraction technology for remote sensing based on feature knowledge[D]. Wuhan: China University of Geosciences, 2010. ]
[13]	杜培军,柳思聪.融合多特征的遥感影像变化检测[J].遥感学报,2012,16(4):663-677. doi: 10.11834/jrs.20121168
	[ Du P J, Liu S C.Change detection from multi-temporal remote sensing images by integrating multiple features[J]. Journal of Remote Sensing, 2012,16(4):663-677. ] doi: 10.11834/jrs.20121168
[14]	汤玉奇. 面向对象的高分辨率影像城市多特征变化检测研究[D].武汉:武汉大学, 2013.
	[ Tang Y Q.Object-oriented change detection with multi-feature in urban high-resolution remote sensing imagery[D]. Wuhan: Wuhan University, 2013. ]
[15]	汪闽,张星月.多特征证据融合的遥感图像变化检测[J].遥感学报,2010,14(3):558-570.
	[ Wang M, Zhang X Y.Change detection using high spatial resolution remotely sensed imagery by combining evidence theory and structural similarity[J]. Journal of Remote Sensing, 2010,14(3):558-570. ]
[16]	王文杰,赵忠明,朱海青.面向对象特征融合的高分辨率遥感图像变化检测方法[J].计算机应用研究,2009,26(8):3149-3151. doi: 10.3969/j.issn.1001-3695.2009.08.102
	[ Wang W J, Zhao Z M, Zhu Q H.Object-oriented change detection method based on features fusion for high-resolution remote sensing image[J]. Application Research of Computers, 2009,26(8):3149-3151. ] doi: 10.3969/j.issn.1001-3695.2009.08.102
[17]	沈占锋,骆剑承,胡晓东,等.高分辨率遥感影像多尺度均值漂移分割算法研究[J].武汉大学学报·信息科学版,2010,35(3):313-316.
	[ Shen Z F, Luo J C, Hu X D, et al.A mean shift multi-scale segmentation for high-resolution remote sensing images[J]. Geomatics and Information Science of Wuhan University, 2010,35(3):313-316. ]
[18]	Huang X, Zhang L P.An adaptive mean-shift analysis approach for object extraction and classification from urban hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2008,46(12):4173-4185. doi: 10.1109/TGRS.2008.2002577
[19]	骆剑承,周成虎,梁怡,等.有限混合密度模型及遥感影像EM聚类算法[J].中国图象图形学报,2002,4(4):26-30. doi: 10.11834/jig.200204119
	[ Luo J C, Zhou C H, Leung Y, et al.Finite mixture model and its EM clustering algorithm[J]. Journal of Image and Graphics, 2002,4(4):26-30. ] doi: 10.11834/jig.200204119
[20]	骆剑承,王钦敏,马江洪,等.遥感图像最大似然分类方法的EM改进算法[J].测绘学报,2002,31(3):234-239. doi: 10.3321/j.issn:1001-1595.2002.03.010
	[ Luo J C, Wang Q M, Ma J H, et al.The EM-based maximum likelihood classifier for remotely sensed data[J]. Acta Geodaetica et Cartographica Sinica, 2002,31(3):234-239. ] doi: 10.3321/j.issn:1001-1595.2002.03.010
[21]	Foody G M.Assessing the accuracy of land cover change with imperfect ground reference data[J]. Remote Sensing of Environment, 2010,114(10):2271-2285. doi: 10.1016/j.rse.2010.05.003

类型	特征类型	图谱类型	主要来源	主要表现形式	适用性	复杂度
视觉	光谱特征	谱	栅格影像	反射率、指数	高	简单
	形状特征	图	栅格影像	距离、面积、方位	高
	纹理特征	图	栅格影像	概率	中
空间	空间特征	图	栅格影像	拓扑	中	复杂
模型	波谱特征	谱	波谱库	波段反射率	中
模型	时相特征	谱	时序影像	差值、比值	高
环境	地形特征	图	地形数据	高度、坡度	中
环境	气候特征	谱	统计数据	温度、湿度	低

地类	分割尺度	光谱特征相似度-证据信任度	纹理特征相似度-证据信任度
耕地	100	0.8	0.2
林地	500	0.3	0.7
水域	50	0.8	0.2
建设用地	20	0.3	0.7

使用特征	虚警率/(%)	漏警率/(%)	正确率/(%)
仅光谱特征	47.65	12.28	69.23
仅纹理特征	41.99	13.43	73.62
光谱特征+纹理特征+等信任度融合	40.14	13.29	74.73
光谱特征+纹理特征+差异融合	39.06	12.57	75.45

方法组合	计算用时/s
方法组合	分割	特征计算	相似度计算	二值化	其它	合计
仅光谱特征		7.64	2.19			217.39
仅纹理特征		10.68	2.15			220.39
光谱特征+纹理特征+等信任度融合	123.14	18.32	2.64	2.22	82.2	228.52
光谱特征+纹理特征+差异融合		18.32	3.23			229.11

二值化方法	计算用时/s	虚警率/(%)	漏警率/(%)	正确率/(%)
基于大津法OTSU的阈值法	1.17	44.87	11.13	73.03
基于Sigma迭代逼近的阈值法	0.23	39.82	13.54	74.89
基于EM算法的GMM模型	2.22	39.06	12.57	75.45

历史解译知识引导下组合遥感图谱特征的变化检测方法

A Remote Sensing Change Detection Method Combining with Spatial-Spectral Features under the Guidance of Historical Interpretation Knowledge

RichHTML

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 21

相关文章 5

Metrics

本文评价

推荐阅读 0

[1]	王丽英, 赵元丁. 一种灰度体素结构分割模型下的机载LiDAR 3D滤波算法[J]. 地球信息科学学报, 2020, 22(11): 2118-2127.
[2]	刘浩, 骆剑承, 黄波, 杨海平, 胡晓东, 徐楠, 夏列钢. 基于特征压缩激活Unet网络的建筑物提取[J]. 地球信息科学学报, 2019, 21(11): 1779-1789.
[3]	陈扬洋, 明冬萍, 徐录, 赵璐. 高空间分辨率遥感影像分割定量实验评价方法综述[J]. 地球信息科学学报, 2017, 19(6): 818-830.
[4]	刘建华. 多色彩高空间分辨率遥感影像矢量边缘信息提取算法与应用[J]. 地球信息科学学报, 2010, 12(6): 850-854.
[5]	刘生, 王潇宇, 邱新法. 基于数学形态学的高空间分辨率遥感影像几何特征提取[J]. 地球信息科学学报, 2008, 10(2): 251-256.