基于超像元词包特征和主动学习的高分遥感影像变化检测

doi:10.12082/dqxxkx.2019.190136

地球信息科学学报 ›› 2019, Vol. 21 ›› Issue (10): 1594-1607.doi: 10.12082/dqxxkx.2019.190136

基于超像元词包特征和主动学习的高分遥感影像变化检测

杨进一^1,^2,³,徐伟铭^1,^2,^3,^*(),王成军^1,^2,³,翁谦^1,^2,^3,⁴

^1. 福州大学数字中国研究院（福建）,福州 350108
^2. 福州大学空间数据挖掘与信息共享教育部重点实验室,福州 350002
^3. 福州大学地理空间信息技术国家地方联合工程技术研究中心,福州 350002
^4. 福州大学数学与计算机科学学院,福州 350116

收稿日期:2019-03-25 修回日期:2019-07-13 出版日期:2019-10-25 发布日期:2019-10-29
通讯作者: 徐伟铭 E-mail:xwming2@126.com
作者简介:杨进一（1993-）,男,安徽霍邱人,硕士生,主要从事高分遥感影像信息提取、机器学习研究。E-mail: yangjinyi_gis@163.com
基金资助:
国家自然科学基金项目(41801324);福建省科技厅引导性项目(2017Y0055);“数字福建”重大项目([2016]203号)

High-Resolution Remote Sensing Imagery Change Detection based on Super-Pixel BOW Features and Active Learning

YANG Jinyi^1,^2,³,XU Weiming^1,^2,^3,^*(),WANG Chengjun^1,^2,³,WENG Qian^1,^2,^3,⁴

^1. The Academy of Digital China, Fuzhou University, Fuzhou 350108, China
^2. Key Laboratory of Spatial Data Mining & Information Sharing of Ministry of Education, Fuzhou University, Fuzhou 350002, China;
^3. National Engineering Research Centre of Geospatial Information Technology, Fuzhou University, Fuzhou 350002, China
^4. College of Mathematics and Computer Science, Fuzhou University, Fuzhou 350116, China

Received:2019-03-25 Revised:2019-07-13 Online:2019-10-25 Published:2019-10-29
Contact: XU Weiming E-mail:xwming2@126.com
Supported by:
National Natural Science Foundation of China(41801324);Science and Technology Agency of Fujian Province(2017Y0055);“Digital Fujian” Program([2016]203号)

摘要/Abstract

摘要：

为解决高分辨率遥感影像变化检测中存在底层特征缺乏语义信息、像元级的检测结果存在“椒盐”现象以及监督分类中样本标注自动化程度较低,本文提出一种基于超像元词包特征和主动学习的变化检测方法。首先采用熵率分割算法获取叠加影像的超像元对象;其次提取两期影像像元点对间的邻近相关影像特征（相关度、斜率和截距）和顾及邻域的纹理变化强度特征（均值、方差、同质性和相异性）,经线性组合作为像元点对的底层特征;然后基于像元点对底层特征利用BOW模型构建超像元词包特征,并采用一种改进标注策略的主动学习方法从无标记样本池中优选信息量较大的样本,且自动标注样本类别;最后训练分类器模型完成变化检测。通过选用2组不同地区的GF-2影像和Worldview-Ⅱ影像作为数据源进行实验,实验结果中2组数据集的F1分数分别为0.8714、0.8554,正确率分别为0.9148、0.9022,漏检率分别为0.1681、0.1868,误检率分别为0.0852、0.0978。结果表明,该法能有效识别变化区域、提高变化检测精度。此外,传统主动学习方法与改进标注策略的主动学习方法的学习曲线对比显示,改进的标注策略可在较低精度损失下,有效提高样本标注自动化程度。

关键词: 高分影像, 变化检测, 超像元, 词包特征, 主动学习, 自动标注

Abstract:

The following problems exist in the change detection of high-resolution remote sensing imagery: lack of the semantic information of low-level features, the "salt and pepper" phenomenon in the detection results based on pixel-level methods, and the low degree of sample labeling automation in supervised classification. In this paper, we proposed a change detection method based on super-pixel Bag-of-Words features and active learning. Firstly, we used the entropy rate segmentation algorithm to obtain the segmentation objects of superimposed images. Secondly, we extracted the features of the Neighborhood Correlation Images (correlation, slope, and intercept) and the change intensity features of texture (mean value, variance, homogeneity, and dissimilarity) while considering neighborhood context information between pixel pairs of the studied two phases of images, and then combined them as the low-level features of pixel pairs. Followingly, based on these low-level features, we constructed the expression of Bag-of-Words features in the super-pixel regions by the Bag-of-Words (BOW) model, and we adopted an improved annotation strategy to annotate automatically the samples with large information from the unlabeled sample pool. Finally, we conducted the change detection using the trained classification model. By choosing two groups in different parts of GF-2 imagery and Worldview-Ⅱ imagery as a data source for experiments, the experimental results show that the F1 scores of the two groups of data sets are 0.8714 and 0.8554, the precision is 0.9148 and 0.9022, the missed detection rate is 0.1681 and 0.1868, and the false detection rate is 0.0852 and 0.0978, respectively. The results demonstrate that our proposed method can effectively ide.pngy the variation area, improve the accuracy of change detection. In addition, the comparison of the learning curves between the traditional active learning method and the active learning method with improved annotation strategy shows that the improved annotation strategy can effectively improve the automation degree of sample annotation at a lower precision loss.

Key words: high-resolution remote sensing imagery, change detection, super-pixel, Bag-of-Words features, active learning, automatic annotation

杨进一,徐伟铭,王成军,翁谦. 基于超像元词包特征和主动学习的高分遥感影像变化检测[J]. 地球信息科学学报, 2019, 21(10): 1594-1607.DOI:10.12082/dqxxkx.2019.190136

YANG Jinyi,XU Weiming,WANG Chengjun,WENG Qian. High-Resolution Remote Sensing Imagery Change Detection based on Super-Pixel BOW Features and Active Learning[J]. Journal of Geo-information Science, 2019, 21(10): 1594-1607.DOI:10.12082/dqxxkx.2019.190136

图/表 13

图1

图2

图3

图4

表1

图5

图6

表2

图7

图8

图9

表3

表4

参考文献 33

[1]	张良培, 武辰 . 多时相遥感影像变化检测的现状与展望[J]. 测绘学报, 2017,46(10):1447-1459.
	[ Zhang L P, Wu C . Advance and future development of change detection for multi-temporal remote sensing imagery[J]. Acta Geodaetica et Cartographica Sinica, 2017,46(10):1447-1459. ]
[2]	RJ Radke, S Andra, O AlKofahi , et al. Image change detection algorithms: A systematic survey[J]. IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society, 2005,14(3):294.
[3]	Wen D, Huang X, Zhang L , et al. A novel automatic change detection method for urban high-resolution remotely sensed imagery based on multiindex scene representation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015,54(1):609-625.
[4]	Singh A . Change detection in the tropical forest environment of northeastern India using Landsat[J]. Remote sensing and tropical land management, 1986: 237-254.
[5]	Luczkovich J, Wagner T, Michalek J , et al. Discrimination of coral reefs, seagrass meadows, and sand bottom types from space: A dominican republic case study[J]. Photogrammetric Engineering and Remote Sensing, 1993,59(3):385-389.
[6]	Yang G, Li H C, Liu C. Unsupervised change detection of remote sensing images using superpixel segmentation and variational Gaussian mixture model[C]// Analysis of Multitemporal Remote Sensing Images. IEEE, 2017: 1-4.
[7]	Angluin D . Queries and concept learning[J]. Machine Learning, 1988,2(4):319-342.
[8]	Yu H, Yang W, Hua G , et al. Change detection using high resolution remote sensing images based on active learning and Markov Random Fields[J]. Remote Sensing, 2017,9(12):1233.
[9]	王成军, 毛政元, 徐伟铭 , 等. 超像素与主动学习相结合的遥感影像变化检测方法[J]. 地球信息科学学报, 2018,20(2):235-245.
	[ Wang C J, Mao Z Y, Xu W M , et al. Change detection approach for high resolution remotely sensed images based on superpixel and active learning[J]. Journal of Geo-information Science, 2018,20(2):235-245. ]
[10]	Yu H, Sun C, Yang W , et al. AL-ELM: One uncertainty-based active learning algorithm using extreme learning machine[J]. Neurocomputing, 2015,166:140-150.
[11]	翟俊海, 臧立光, 张素芳 . 在线序列主动学习方法[J]. 计算机科学, 2017,44(1):37-41.
	[ Zhai J H, Zang L G, Zhang S F . Online sequential active learning approach[J]. Computer Science, 2017,44(1):37-41. ]
[12]	Samat A, Gamba P, Du P , et al. Active extreme learning machines for quad-polarimetric SAR imagery classification[J]. International Journal of Applied Earth Observation and Geoinformation, 2015,35:305-319.
[13]	Liu J, Yu H, Yang W, et al. Combining active learning and semi-supervised learning based on extreme learning machine for multi-class image classification[C]. International Conference on Intelligent Science, 2015: 163-175.
[14]	赵理君, 唐娉, 霍连志 , 等. 图像场景分类中视觉词包模型方法综述[J]. 中国图象图形学报, 2014,19(3):333-343. doi: 10.11834/jig.20140301
	[ Zhao L J, Tang P, Huo L Z , et al. Review of the bag-of-visual-words models in image scene classification[J]. Journal of Image and Graphics, 2014,19(3):333-343. ] doi: 10.11834/jig.20140301
[15]	Wu C, Zhang L, Zhang L . A scene change detection framework for multi-temporal very high resolution remote sensing images[J]. Signal Processing, 2016,124(C):184-197.
[16]	Zhu Q, Zhong Y, Zhao B , et al. Bag-of-Visual-Words scene classifier with local and global features for high spatial resolution remote sensing imagery[J]. IEEE Geoscience & Remote Sensing Letters, 2016,13(6):747-751.
[17]	浮瑶瑶, 柳彬, 张增辉 , 等. 基于词包模型的高分辨率SAR 图像变化检测与分析[J]. 雷达学报, 2014,3(1):101-110. doi: 10.3724/SP.J.1300.2014.13134
	[ Fu Y Y, Liu B, Zhang Z H , et al. Change detection and analysis of high resolution synthetic aperture radar images based on bag-of-words Model[J]. Journal of Radars, 2014,3(1):101-110. ] doi: 10.3724/SP.J.1300.2014.13134
[18]	Tu J, Li D, Feng W , et al. Detecting damaged building regions based on semantic scene change from multi-temporal high-resolution remote sensing images[J]. ISPRS International Journal of Geo-information, 2017,6(5):131.
[19]	罗星, 徐伟铭, 王佳 . 基于对象BOW特征的高分辨率遥感影像变化检测方法[J]. 地球信息科学学报, 2018,20(8):1150-1159.
	[ Luo X, Xu W M, Wang J . Change detection method for high resolution remote sensing images based on BOW features representation[J]. Journal of Geo-information Science, 2018,20(8):1150-1159. ]
[20]	Ren X, Malik J. Learning a classification model for segmentation[C]// IEEE International Conference on Computer Vision. IEEE, 2003: 10-17.
[21]	Liu M Y, Tuzel O, Ramalingam S , et al. Entropy rate superpixel segmentation[J]. Computer Vision and Pattern Recognition, 2011,32(14):2097-2104.
[22]	冯文卿, 眭海刚, 涂继辉 , 等. 联合像素级和对象级分析的遥感影像变化检测[J]. 测绘学报, 2017,46(9):1147-1155.
	[ Feng W Q, Sui H G, Tu J H , et al. Remote sensing image change detection based on the combination of pixel-level and object-level analysis[J]. Acta Geodaetica et Cartographica Sinica, 2017,46(9):1147-1155. ]
[23]	Tu B, Li N, Fang L , et al. Classification of hyperspectral images via weighted spatial correlation representation[J]. Journal of Visual Communication and Image Representation, 2018,56:160-166.
[24]	Fang L, Zhuo H, Li S . Super-resolution of hyperspectral image via superpixel-based sparse representation[J]. Neurocomputing, 2018,273:171-177.
[25]	Im J, Jensen J R . A change detection model based on neighborhood correlation image analysis and decision tree classification[J]. Remote Sensing of Environment, 2005,99(3):326-340.
[26]	Thonfeld F, Feilhauer H, Braun M , et al. Robust Change Vector Analysis (RCVA) for multi-sensor very high resolution optical satellite data[J]. International Journal of Applied Earth Observation & Geoinformation, 2016,50(8):131-140.
[27]	Im J . Neighborhood correlation image analysis for change detection using different spatial resolution imagery[J]. Korean Journal of Remote Sensing, 2006,22(5):337-350.
[28]	Im J, Jensen J R, Tullis J A . Object-based change detection using correlation image analysis and image segmentation[J]. International Journal of Remote Sensing, 2008,29(2):399-423.
[29]	Li M, Zhu X, Zhao A, et al. Single late rice information extraction based on change detection method using neighborhood correlation images[C]//2014 The Third International Conference on Agro-Geoinformatics. IEEE, 2014: 1-4.
[30]	Haralick R M, Shanmugam K . Textural features for image classification[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1973(6):610-621.
[31]	徐盛 . 基于主题模型的高空间分辨率遥感影像分类研究[D]. 上海:上海交通大学, 2012.
	[ Xu S . Classification of high spatial resolution remote sensing images based on topic model[D]. Shanghai: Shanghai Jiao Tong University, 2012. ]
[32]	Huang G B, Zhu Q Y, Siew C K . Extreme learning machine: Theory and applications[J]. Neurocomputing, 2006,70(1):489-501.
[33]	陈伟锋, 毛政元, 徐伟铭 , 等. 基于Adaboost的高分遥感影像自动变化检测方法[J]. 地球信息科学学报, 2018,20(12):1756-1767.
	[ Chen W F, Mao Z Y, Xu W M , et al. Automatic change detection approach for high-resolution remotely sensed images based on Adaboost algorithm[J]. Journal of Geo-information Science, 2018,20(12):1756-1767. ]

	真实变化像元数	真实未变化像元数	总和
检测变化像元数	TP	FP	P
检测未变化像元数	FN	TN	N
总和	P'	N'	T

方法	过程描述
BOW-IAL	基于视觉词典构建样本的词包特征在ELM训练中采用主动学习采样,并利用改进标注策略自动标注所采样本,完成训练、变化检测
BOW-RS	构建样本的词包特征（同BOW-IAL）在ELM训练中采用随机方法进行采样,并利用改进的标注策略自动标注所采样本,完成训练、变化检测
LOW-IAL	构建样本的底层特征（前文所述的3种NCIs特征、4种纹理特征,共16维特征）,其中超像元NCIs特征构造方法参考文献[28] 在ELM训练中采用主动学习采样,并利用改进标注策略自动标注所采样本,完成训练、变化检测
LOW-RS	构建样本的底层特征（同LOW-IAL）在ELM训练中采用随机方法进行采样,并利用改进的标注策略自动标注所采样本,完成训练、变化检测
BOW-AL	构建样本的词包特征（同BOW-IAL）在ELM训练中采用主动学习采样,人工标注所采样本,完成训练、变化检测
LOW-AL	构建样本的底层特征（同LOW-IAL）在ELM训练中采用主动学习采样,人工标注所采样本,完成训练、变化检测
OCVA	基于超像元分割结果构建光谱变化强度图利用大津法获取最佳变化阈值,阈值分割得到变化检测结果
SVM	基于超像元分割结果选取光谱和纹理（光谱特征包括光谱均值和方差,纹理特征包括均值、方差、同质性和相异性）构建差异特征向量随机选择总样本的60%作为训练集,剩余40%为测试集,训练SVM分类器,完成变化检测

方法	F1分数	正确率	漏检率	误检率
BOW-IAL	0.8714	0.9148	0.1681	0.0852
BOW-RS	0.7623	0.8646	0.3183	0.1354
LOW-IAL	0.8143	0.8369	0.2072	0.1631
LOW-RS	0.6774	0.7652	0.3924	0.2348
OCVA	0.6849	0.6332	0.2541	0.3668
SVM	0.8261	0.8455	0.1924	0.1545

方法	F1分数	正确率	漏检率	误检率
BOW-IAL	0.8554	0.9022	0.1868	0.0978
BOW-RS	0.7383	0.8312	0.3359	0.1688
LOW-IAL	0.7847	0.7969	0.2272	0.2031
LOW-RS	0.6316	0.7121	0.4326	0.2879
OCVA	0.6668	0.6230	0.2829	0.3770
SVM	0.7919	0.8009	0.2169	0.1991

基于超像元词包特征和主动学习的高分遥感影像变化检测

High-Resolution Remote Sensing Imagery Change Detection based on Super-Pixel BOW Features and Active Learning

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 33

相关文章 14

Metrics

本文评价

推荐阅读 0

[1]	杜培军, 王欣, 蒙亚平, 林聪, 张鹏, 卢刚. 面向地理国情监测的变化检测与地表覆盖信息更新方法[J]. 地球信息科学学报, 2020, 22(4): 857-866.
[2]	吴瑞娟, 何秀凤, 王静. 结合像元级与对象级的滨海湿地变化检测方法[J]. 地球信息科学学报, 2020, 22(10): 2078-2087.
[3]	罗星, 徐伟铭, 王佳. 基于对象BOW特征的高分辨率遥感影像变化检测方法[J]. 地球信息科学学报, 2018, 20(8): 1150-1159.
[4]	王成军, 毛政元, 徐伟铭, 翁谦. 超像素与主动学习相结合的遥感影像变化检测方法[J]. 地球信息科学学报, 2018, 20(2): 235-245.
[5]	陈伟锋, 毛政元, 徐伟铭, 许锐. 基于Adaboost的高分遥感影像自动变化检测方法[J]. 地球信息科学学报, 2018, 20(12): 1756-1767.
[6]	汤冬梅, 樊辉, 张瑶. Landsat时序变化检测综述[J]. 地球信息科学学报, 2017, 19(8): 1069-1079.
[7]	吴田军, 夏列钢, 吴炜, 马江洪. 土地执法监察中的高分辨率遥感及变化检测技术[J]. 地球信息科学学报, 2016, 18(7): 962-968.
[8]	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.
[9]	施文灶, 毛政元. 基于图分割的高分辨率遥感影像建筑物变化检测研究[J]. 地球信息科学学报, 2016, 18(3): 423-432.
[10]	李杨, 江南, 侍昊, 邵华. Landsat-8影像的LDA模型变化检测[J]. 地球信息科学学报, 2015, 17(3): 353-360.
[11]	高锡章, 刘海江, 李宝林, 袁烨城. 一种基于面积对称模型的土地利用/覆被真伪变化图斑检测方法[J]. 地球信息科学学报, 2014, 16(5): 784-789.
[12]	王丽云, 李艳, 汪禹芹. 基于对象变化矢量分析的土地利用变化检测方法研究[J]. 地球信息科学学报, 2014, 16(2): 307-313.
[13]	龙玄耀, 李培军. 基于图像分割的城市变化检测[J]. 地球信息科学学报, 2008, 10(1): 121-127.
[14]	王琳, 徐涵秋, 李胜. 福州城市扩展的遥感动态监测[J]. 地球信息科学学报, 2006, 8(4): 129-135,151.