Please wait a minute...
img

官方微信

遥感技术与应用  2023, Vol. 38 Issue (5): 1107-1117    DOI: 10.11873/j.issn.1004-0323.2023.5.1107
数据与图像处理     
基于编码对齐生成对抗网络的异源遥感影像地表覆被变化检测方法
张成才1(),刘威1,杨峰2,彭凯1(),周雪丽3
1.黄河实验室(郑州大学),河南 郑州 450001
2.河南省出山店水库建设管理局,河南 信阳 464000
3.河南省地图院,河南 郑州 450003
Change Detection Method for Surface Cover of Heterogeneous Remote Sensing Image based on Code-Aligned Generative Adversarial Network
Chengcai ZHANG1(),Wei LIU1,Feng YANG2,Kai PENG1(),Xueli ZHOU3
1.Yellow River Laboratory (Zhengzhou University),Zhengzhou 450001,China
2.Chushandian Reservoir Management Bureau,Henan Province,Xinyang 464000,China
3.Henan Map Institution,Henan Province,Zhengzhou 450003,China
 全文: PDF(5781 KB)   HTML
摘要:

相较于同源遥感影像地表覆被变化检测,异源影像能综合不同卫星传感器间数据特征和现势性等优势,更好满足应用需求。针对异源遥感影像变化检测中存在的光谱差异和特征空间不一致问题,研究提出编码对齐生成对抗网络实现异源影像的高精度变化检测。考虑到异源影像间通道和数据类型上存在差异,难保持重构前后影像空间结构的一致性,研究通过添加自编码器和构造编码对齐损失,约束编码器输出特征的空间结构变化,使重构前后影像空间结构一致,有效减少信息丢失;在跨域映射过程中为减少源域与目标域间影像的色彩差异,采用循环一致对抗生成网络在无成对影像情况下进行色彩迁移,实现两时相异源影像的相互跨域映射,生成能与原始影像直接对比的无色偏重构影像;利用设计的变化概率权重使网络在训练过程中自动选择样本,有效提取覆被变化信息。实验结果表明:该方法与CGAN、SCCN等方法相比能更充分提取影像特征,降低跨域映射函数的随机性;在4组公开数据集的检测精度分别达到0.93、0.96、0.97、0.88,精度最高;变化检测结果与参考图的一致性、检测差异图质量均最优。因此,该方法在异源遥感影像中能够进行高精度变化检测。

关键词: 异源遥感影像变化检测自适应跨域映射    
Abstract:

Compared to the change detection of homologous remote sensing images, heterogeneous images can integrate the advantages of different satellite sensor data features and temporal relevance, better satisfying application requirements. To address the issues of spectral differences and inconsistent feature spaces in change detection of heterogeneous remote sensing images, this study proposes an aligned generative adversarial network for high-precision change detection of heterogeneous images. Considering the differences in channels and data types between heterogeneous images, it is difficult to maintain the consistency of spatial structures before and after reconstruction. The study incorporates autoencoders and constructs alignment loss to constrain the spatial structure changes of encoder output features, ensuring consistency in spatial structures between the reconstructed images and reducing information loss effectively. In the cross-domain mapping process, to minimize the color differences between source and target domain images, a cycle-consistent adversarial generative network is used for color transfer in the absence of paired images, enabling mutual cross-domain mapping between two temporally distinct heterogeneous images, generating color-preserving reconstructed images that can be directly compared with the original images. By utilizing designed change probability weights, the network automatically selects samples during the training process, effectively extracting land cover change information. Experimental results demonstrate that compared to methods such as CGAN and SCCN, the proposed method can more fully extract image features and reduce the randomness of cross-domain mapping functions. The detection accuracies on four publicly available datasets reach 0.93, 0.96, 0.97, and 0.88, with the highest accuracy achieved. The consistency between the change detection results and the reference maps, as well as the quality of the difference maps, is optimal. This method enables high-precision change detection in heterogeneous remote sensing images.

Key words: Heterogeneous remote sensing images    Change detection    Adaptive    Cross-domain mapping
收稿日期: 2022-08-27 出版日期: 2023-11-07
ZTFLH:  P237  
基金资助: 河南省自然科学基金项目“联合主动微波遥感和光学遥感数据的大型灌区土壤水分反演方法研究”(222300420539);河南省水利科技攻关项目“河南省出山店水库流域覆被遥感监测方法”(GG201902);国家自然科学基金重点项目“基于大数据的城市洪涝灾害预报预警理论与方法研究”(51739009)
通讯作者: 彭凯     E-mail: zhangcc2000@163.com;1029127803@qq.com
作者简介: 张成才(1964-),男,河南郸城人,博士,教授,主要从事水利信息技术研究。E?mail: zhangcc2000@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
张成才
刘威
杨峰
彭凯
周雪丽

引用本文:

张成才,刘威,杨峰,彭凯,周雪丽. 基于编码对齐生成对抗网络的异源遥感影像地表覆被变化检测方法[J]. 遥感技术与应用, 2023, 38(5): 1107-1117.

Chengcai ZHANG,Wei LIU,Feng YANG,Kai PENG,Xueli ZHOU. Change Detection Method for Surface Cover of Heterogeneous Remote Sensing Image based on Code-Aligned Generative Adversarial Network. Remote Sensing Technology and Application, 2023, 38(5): 1107-1117.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2023.5.1107        http://www.rsta.ac.cn/CN/Y2023/V38/I5/1107

图1  变化检测流程图
图2  CAGAN结构(a)网络结构 (b)影像跨域回构过程
名称模块参数设置

编码器

生成器

卷积层conv3×3,stride=1,padding=1LeakyReLUslope=0.3,Dropoutp=0.2×4
conv3×3,stride=1,padding=1tanh×1
判别器卷积层conv3×3,stride=2,padding=0LeakyReLUslope=0.3,Dropoutp=0.2×3
全连接层Input=25,Output=1Sigmoid×1
表1  CAGAN 的参数设置
图3  数据集
图4  数据集变化结果混淆图
图5  数据集差异图
方法AUCOAKC
CAGAN0.870.930.40
CycleGAN0.820.890.31
CGAN0.710.880.13
SCCN0.880.890.39
DNN---0.860.27
Luppino等(2021)0.870.920.41
表2  数据集1变化检测精度指标
方法AUCOAKC
CAGAN0.890.960.71
CycleGAN0.850.930.55
CGAN0.800.920.50
SCCN0.810.930.51
DNN---0.850.17
Touati等(2019)---0.96---
Touati等(2017)---0.94--
表3  数据集2变化检测精度指标
方法AUCOAKC
CAGAN0.920.970.75
CycleGAN0.850.920.54
CGAN0.660.920.29
SCCN0.690.860.21
DNN---0.950.23
表4  数据集3变化检测精度指标
方法AUCOAKC
CAGAN0.790.880.42
CycleGAN0.770.850.37
CGAN0.640.820.10
SCCN0.750.820.38
DNN---0.810.30
Touati等(2017)---0.85---
表5  数据集4变化检测精度指标
图6  高斯滤波和阈值分割法评价结果图
方法AUCOAKC
CAGAN_Gaussian_Ostu0.890.960.71
CAGAN_Gaussian_Kmeans0.890.950.73
CAGAN_unGaussian_Ostu0.840.920.61
表6  高斯滤波和阈值分割精度指标 (impact evaluation accuracy)
1 ZHANG Liangpei, WU Chen.Advance and future development of change detection for multi-temporal remote sensing imagery[J].Acta Geodaetica et Cartographica Sinica,2017,46(10):1447-1459.
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.
2 LI H, XIAO P, FENG X, et al. Using land long-term data records to map land cover changes in China over 1981-2010[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2017,PP(4):1-18. DOI: 10.1109/JSTARS.2016.2645203
doi: 10.1109/JSTARS.2016.2645203
3 ZELINSKI M E, HENDERSON J, SMITH M. Use of Landsat 5 for Change detection at 1998 indian and pakistani nuclear test sites[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2014, 7(8):3453-3460. DOI: 10.1109/JSTARS.2013.2294322
doi: 10.1109/JSTARS.2013.2294322
4 SHI W, ZHANG M, ZHANG R, et al. Change detection based on artificial intelligence: State-of-the-Art and challenges[J].Remote Sensing,2020,12(10):1688. DOI:10.3390/ rs12101688
doi: 10.3390/ rs12101688
5 STORVIK B, STORVIK G, FJORTOFT R.On the combination of multisensor data using meta-gaussian distributions[J]. IEEE Transactions on Geoscience & Remote Sensing,2009,47(7):2372-2379. DOI:10.1109/TGRS.2009. 2012699
doi: 10.1109/TGRS.2009. 2012699
6 ZHANG P, GONG M, SU L, et al. Change detection based on deep feature representation and mapping transformation for multi-spatial-resolution remote sensing images[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 116:24-41. DOI: 10.1016/j.isprsjprs.2016.02.013
doi: 10.1016/j.isprsjprs.2016.02.013
7 TAO Z, GONG M, JIA L, et al. Iterative feature mapping network for detecting multiple changes in multi- source remote sensing images[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 146:38-51. DOI: 10.1016/j.isprsjprs.2018.09.002
doi: 10.1016/j.isprsjprs.2018.09.002
8 YOU Y, CAO J, ZHOU W. A survey of change detection methods based on remote sensing images for multi-source and multi-objective scenarios[J]. Remote Sensing,2020,12(15):2460. DOI: 10.3390/rs12152460
doi: 10.3390/rs12152460
9 MERCIER G, MOSER G, SERPICO S B. Conditional copulas for change detection in heterogeneous remote sensing images[J]. IEEE Transactions on Geoence & Remote Sensing,2008,46(5):1428-1441. DOI:10.1109/TGRS. 2008. 916476
doi: 10.1109/TGRS. 2008. 916476
10 HUSSAIN M, CHEN D M, CHENG A, et al. Change detection from remotely sensed images: From pixel-based to object-based approaches[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2013, 80: 91-106. DOI: 10.1016/j.isprsjprs.2013.03.006
doi: 10.1016/j.isprsjprs.2013.03.006
11 BOVOLO F, BRUZZONE L. A theoretical framework for unsupervised change detection based on change vector analysis in the polar domain[J]. IEEE Transactions on Geoscience and Remote Sensing,2007,45(1):218-36. DOI:10.1109/TGRS. 2006.885408
doi: 10.1109/TGRS. 2006.885408
12 CELIK T. Unsupervised change detection in satellite images using principal component analysis and k-Means clustering[J]. IEEE Geoscience and Remote Sensing Letters, 2009, 6(4): 772-6. DOI: 10.1109/LGRS.2009.2025059
doi: 10.1109/LGRS.2009.2025059
13 CANTY M J, NIELSEN A A. Automatic radiometric normalization of multitemporal satellite imagery with the iteratively re-weighted MAD transformation [J]. Remote Sensing of Environment, 2008, 112(3): 1025-36. DOI: 10.1016/j.rse.2007.07.013
doi: 10.1016/j.rse.2007.07.013
14 BONTEMPS S, BOGAERT P, TITEUX N, et al. An object-based change detection method accounting for temporal dependences in time series with medium to coarse spatial resolution[J]. Remote Sensing of Environment, 2008, 112(6):3181-3191.DOI: 10.1016/j.rse.2008.03.013
doi: 10.1016/j.rse.2008.03.013
15 CHEN H, YOKOYA N, CHINI M. Fourier domain structural relationship analysis for unsupervised multimodal change detection[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 198: 99-114.
16 GUO Qing, ZHU Liya, LI An, et al. Landslide identification method based on NDVI change detection[J]. Remote Sensing Technology and Application, 2022, 37(1): 17-23.
16 郭擎,朱丽娅,李安,等. 基于NDVI变化检测的滑坡遥感精细识别[J]. 遥感技术与应用, 2022, 37(1): 17-23.
17 SHELHAMER E, LONG J, DARRELL T.Fully convolutional networks for semantic segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651.DOI:10.1109/TPAMI.2016.2572683
doi: 10.1109/TPAMI.2016.2572683
18 DAUDT R C, LE SAUX B, BOULCH A, et al. Fully Convolutional Siamese Networks for change detection; Proceedings of the 25th IEEE International Conference on Image Processing (ICIP), Athens, GREECE, F Oct 07-10, 2018 [C]∥ 2018. DOI: 10.1109/ICIP.2018.8451652
doi: 10.1109/ICIP.2018.8451652
19 SONG A, CHOI J. Fully convolutional networks with multiscale 3D filters and transfer learning for change detection in high spatial resolution satellite images[J]. Remote Sensing, 2020, 12(5): 799. DOI: 10.3390/rs12050799
doi: 10.3390/rs12050799
20 JIA L, GONG M, KAI Q, et al. A deep convolutional coupling network for change detection based on heterogeneous optical and radar images[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, PP(99):1-15. DOI: 10.1109/TNNLS.2016.2636227
doi: 10.1109/TNNLS.2016.2636227
21 GONG M, ZHAO J, LIU J, et al. Change detection in synthetic aperture radar images based on Deep Neural Networks[J]. IEEE Transactions on Neural Networks & Learning Systems,2017,27(1):125-138. DOI:10.1109/TNNLS. 2015. 2435783
doi: 10.1109/TNNLS. 2015. 2435783
22 NIU X, GONG M, TAO Z, et al. A Conditional Adversarial Network for change detection in heterogeneous images[J]. IEEE Geoscience and Remote Sensing Letters, 2018, PP(1):1-5. DOI: 10.1109/LGRS.2018.2868704
doi: 10.1109/LGRS.2018.2868704
23 HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J] Science. 2006 Jul 28;313(5786):504-7. DOI: 10.1126/science.1127647
doi: 10.1126/science.1127647
24 ZHU J Y, PARK T, ISOLA P, et al. Unpaired image-to-image translation using Cycle-Consistent Adversarial Networks[J]. IEEE International Conference on Computer Vision,2017,pp.2242-2251. DOI:10.48550/arXiv.1703.10593
doi: 10.48550/arXiv.1703.10593
25 KRHENBÜH L, PHILIPP, KOLTUN V. Efficient inference in fully connected CRFs with Gaussian Edge potentials[J]. Curran Associates Inc, 2012:109-117. DOI: 10.48550/arXiv.1210.5644
doi: 10.48550/arXiv.1210.5644
26 OTSU N. A Threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems Man & Cybernetics, 1979, 9(1):62-66.
27 LUPPINO L T, HANSEN M A, KAMPFFMEYER M, et al. Code-aligned autoencoders for unsupervised change detection in multimodal remote sensing images[J]. arXiv: Computer Vision and Pattern Recognition,2020. DOI: 10.1109/TNNLS.2022.3172183
doi: 10.1109/TNNLS.2022.3172183
28 MACK Y P, ROSENBLATT M. Multivariate k-nearest neighbor density estimates[J]. Journal of Multivariate Analysis, 1979, 9(1):1-15.
29 LUPPINO L T, KAMPFFMEYER M, BIANCHI F M, et al. Deep image translation with an affinity-based change prior for unsupervised multimodal change detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021(99):1–22. DOI: 10.1109/TGRS.2021.3056196
doi: 10.1109/TGRS.2021.3056196
30 TOUATI R, MIGNOTTE M.An Energy-based model enco-ding nonlocal pairwise pixel interactions for multisensor chan-ge detection[J]. IEEE Transactions on Geoence & Remote Sensing,2017:1-13. DOI:10.1109/TGRS.2017. 2758359
doi: 10.1109/TGRS.2017. 2758359
31 TOUATI R, MIGNOTTE M, DAHMANE M. Multimodal change detection in remote sensing images using an unsupervised pixel pairwise-based Markov Random Field Model[J]. IEEE Transactions on Image Processing, 2019, PP(99): 757-767. DOI: 10.1109/TIP.2019.2933747
doi: 10.1109/TIP.2019.2933747
[1] 刘鑫,黄进,杨瑛玮,李剑波. 基于CenterNet的改进遥感旋转目标检测[J]. 遥感技术与应用, 2023, 38(5): 1081-1091.
[2] 符龙崇,朱建军,付海强,解清华,韩文涛. 面向双站SAR系统的长波PolInSAR目标自适应分解[J]. 遥感技术与应用, 2023, 38(5): 1017-1027.
[3] 杜绍杰,赵天杰,施建成,马春锋,邹德富,王振,姚盼盼,彭志晴,郑景耀. Sentinel-1和Sentinel-2协同反演地表土壤水分[J]. 遥感技术与应用, 2022, 37(6): 1404-1413.
[4] 张宇卓,李志伟,沈焕锋,彭小元. 激光雷达数据辅助的FY-3D影像阈值自适应云检测方法[J]. 遥感技术与应用, 2022, 37(5): 1109-1118.
[5] 吕开云,侯昭阳,龚循强,杨硕. 一种基于ASR和PAPCNN的NSCT域遥感影像融合方法[J]. 遥感技术与应用, 2022, 37(4): 829-838.
[6] 韦中晖,靳海亮,顾晓鹤,杨英茹,王庚泽,潘瑜春. 基于多时相协同变化检测的耕地撂荒遥感监测[J]. 遥感技术与应用, 2022, 37(3): 539-549.
[7] 马勇,姚武韬,仉淑艳,尚二萍,陈朝霞. 保护性林地砍伐事件空间分布特征及风险评估研究[J]. 遥感技术与应用, 2022, 37(3): 663-671.
[8] 张飞舟,刘华亮,张立福,岑奕,孙雪剑,张红明. 高分辨率光学遥感影像变化检测算法在地震灾情调查中的应用[J]. 遥感技术与应用, 2022, 37(2): 306-318.
[9] 盖一铭,阿里木·赛买提,王伟,吉力力·阿不都外力. 基于样本迁移的干旱区地表覆盖快速更新[J]. 遥感技术与应用, 2022, 37(2): 333-341.
[10] 郭擎,朱丽娅,李安,顾铃燕. 基于NDVI变化检测的滑坡遥感精细识别[J]. 遥感技术与应用, 2022, 37(1): 17-23.
[11] 李晓东,闫守刚,宋开山. 遥感监测东北地区典型湖泊湿地变化的方法研究[J]. 遥感技术与应用, 2021, 36(4): 728-741.
[12] 陈康明,朱旭东. 基于Google Earth Engine的南方滨海盐沼植被时空演变特征分析[J]. 遥感技术与应用, 2021, 36(4): 751-759.
[13] 胡宜娜,安如,艾泽天,都伟冰. 基于无人机高光谱影像的三江源草种精细识别研究[J]. 遥感技术与应用, 2021, 36(4): 926-935.
[14] 王崇阳,田昕. 基于GF⁃1 PMS数据的森林覆盖变化检测[J]. 遥感技术与应用, 2021, 36(1): 208-216.
[15] 雷华锦,李弘毅,王建,郝晓华,赵宏宇,张娟. 基于环境信息和回归模型的青藏高原MODIS积雪面积比例产品制备[J]. 遥感技术与应用, 2020, 35(6): 1303-1311.