Please wait a minute...
 
Remote Sensing for Natural Resources    2025, Vol. 37 Issue (2) : 19-29     DOI: 10.6046/zrzyyg.2023333
|
Unsupervised change detection using SAR images based on the broad learning system
SHAO Pan1,2(), GUAN Zongsheng1,2, JIA Fuwen1,2()
1. Hubei Key Laboratory of Intelligent Vision Monitoring for Hydroelectric Engineering, China Three Gorges University, Yichang 443002, China
2. College of Computer and Information Technology, China Three Gorges University, Yichang 443002, China
Download: PDF(6080 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks    
Abstract  

Change detection using synthetic aperture radar (SAR) images based on deep learning has been a significant research topic in the field of remote sensing. However, it is limited by unreliable training samples and highly time-consuming training. Hence, this study proposed a novel unsupervised change detection method using SAR images based on the broad learning system (BLS). First, a reliable pre-classification method is presented by incorporating neighborhood information into similarity operators, adaptive dual-threshold segmentation, superpixel correction, and visual saliency analysis. This pre-classification method generates a pre-classification map and corresponding training samples. Second, the BLS network is trained using the training samples to generate the BLS-based prediction map for change detection. Third, the pre-classification map and the BLS-based prediction map are fused through two-stage voting to generate the final change detection map. The experimental results of five real SAR image datasets show that the proposed method can produce more reliable training samples and achieve higher accuracy in change detection. Moreover, its efficiency is significantly higher than the change detection model using SAR images based on deep learning.
Keywords unsupervised change detection      synthetic aperture radar      broad learning system      adaptive dual threshold      superpixel      visual saliency analysis     
ZTFLH:  TP79  
Issue Date: 09 May 2025
Service
E-mail this article
E-mail Alert
RSS
Articles by authors
Pan SHAO
Zongsheng GUAN
Fuwen JIA
Cite this article:   
Pan SHAO,Zongsheng GUAN,Fuwen JIA. Unsupervised change detection using SAR images based on the broad learning system[J]. Remote Sensing for Natural Resources, 2025, 37(2): 19-29.
URL:  
https://www.gtzyyg.com/EN/10.6046/zrzyyg.2023333     OR     https://www.gtzyyg.com/EN/Y2025/V37/I2/19
Fig.1  Flowchart of the proposed BLS change detection method
Fig.2  Diagram for dividing DI with adaptive dual thresholds
Fig.3  An example of generation process of pre-classification map
Fig.4  The training process of BLS model
Fig.5  Network structure of BLS
Fig.6  Datasets
Fig.7  Change trends of KC-values with K, T0 and k
数据集 PCAKM GaborTLC ELM PCANet CWNN RUSACD BLS 变化参考图
A
B
C
D
E
Tab.1  Change detection results of different methods on five datasets
数据
集		 方法 FA MD OE KC IoU
A PCAKM 27 4 311 4 338 0.780 1 0.675 9
GaborTLC 5 4 881 4 886 0.747 3 0.634 6
ELM 9 4 801 4 810 0.752 0 0.641 9
PCANet 2 4 717 4 719 0.757 3 0.647 0
CWNN 34 3 946 3 980 0.801 1 0.692 2
RUSACD 8 4 806 4 814 0.751 7 0.640 1
BLS 307 2 532 2 839 0.866 1 0.792 4
B PCAKM 1 602 633 2 235 0.754 0 0.616 7
GaborTLC 1 863 704 2 567 0.724 3 0.580 4
ELM 708 1 176 1 884 0.758 2 0.620 3
PCANet 641 1 235 1 876 0.755 6 0.616 8
CWNN 1 163 695 1 858 0.785 6 0.657 0
RUSACD 538 1 164 1 702 0.777 4 0.644 9
BLS 824 706 1 530 0.816 5 0.698 8
C PCAKM 41 573 7 41 580 0.040 7 0.031 2
GaborTLC 22 024 4 22 026 0.090 1 0.045 8
ELM 6 199 25 6 224 0.258 6 0.175 5
PCANet 12 826 12 12 838 0.155 7 0.094 2
CWNN 13 217 50 13 267 0.146 8 0.089 1
RUSACD 239 176 415 0.849 7 0.741 1
BLS 280 182 462 0.832 8 0.716 2
D PCAKM 65 783 41 65 824 0.016 6 0.008 7
GaborTLC 33 520 80 33 600 0.041 6 0.016 2
ELM 36 336 112 36 448 0.035 7 0.024 4
PCANet 45 968 48 46 016 0.028 9 0.020 1
CWNN 17 072 384 17 456 0.055 7 0.034 7
RUSACD 288 256 544 0.714 1 0.557 5
BLS 181 208 389 0.806 5 0.677 4
E PCAKM 430 724 172 693 603 417 0.520 7 0.424 0
GaborTLC 231 003 212 652 443 655 0.590 4 0.476 7
ELM 444 164 175 766 619 930 0.513 4 0.416 1
PCANet 370 534 171 819 542 353 0.553 8 0.450 7
CWNN 498 038 150 816 648 854 0.510 1 0.418 0
RUSACD 214 101 251 318 465 419 0.553 5 0.435 9
BLS 36 699 163 666 200 365 0.794 8 0.693 4
Tab.2  Accuracy indicators of change detetion maps on 5 datasets
方法 数据集
A B C D E
PCAKM 1.1 0.93 0.97 1.03 18.45
GaborTLC 2.38 4.12 6.02 6.72 1.8×102
ELM 9.03 13.66 12.03 15.26 4.54×102
PCANet 5.06×102 7.84×102 1.04×103 1.36×103 1.09×105
CWNN 5.14×102 6.57×102 6.77×102 7.02×102 8.96×103
RUSACD 2.87×102 3.57×102 3.39×102 3.71×102 1.63×104
BLS 14.65 19.77 13.92 16.23 6.15×102
Tab.3  The computation times of different methods on 5 datasets (s)
方法 数据集
A B C D E
ELM 0.801 4 0.862 1 0.205 3 0.045 4 0.617 7
PCANet 0.809 3 0.885 1 0.144 3 0.039 1 0.671 1
CWNN 0.822 2 0.868 4 0.184 1 0.076 2 0.647 9
RUSACD 0.713 1 0.760 7 0.920 0 0.907 1 0.564 8
BLS 0.904 2 0.876 6 0.877 4 0.913 7 0.817 2
Tab.4  KC values of “pseudolabel samples” of different methods
方法 数据集
A B C D E
ELM* 0.873 8 0.677 8 0.819 2 0.580 3 0.656 5
PCANet* 0.675 2 0.792 7 0.823 3 0.813 2 0.633 5
CWNN* 0.780 1 0.809 5 0.841 1 0.647 9 0.723 1
RUSACD* 0.795 4 0.786 9 0.849 7 0.791 6 0.717 3
BLS 0.866 1 0.816 5 0.832 8 0.806 5 0.794 8
Tab.5  KC values of different methods
[1] Gong M G, Zhao J J, Liu J, et al. Change detection in synthetic aperture Radar images based on deep neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, 27(1):125-138.
doi: 10.1109/TNNLS.2015.2435783 pmid: 26068879
[2] Li W C, Ma P F, Wang H P, et al. SAR-TSCC:A novel approach for long time series SAR image change detection and pattern analysis[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61:5203016.
[3] Hamidi E, Peter B G, Muñoz D F, et al. Fast flood extent monitoring with SAR change detection using google earth engine[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61:4201419.
[4] Li S C, Wang Y J, Cai H F, et al. MF-SRCDNet:Multi-feature fusion super-resolution building change detection framework for multi-sensor high-resolution remote sensing imagery[J]. International Journal of Applied Earth Observation and Geoinformation, 2023, 119:103303.
[5] Li Y Y, Peng C, Chen Y Q, et al. A deep learning method for change detection in synthetic aperture Radar images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(8):5751-5763.
[6] Gao F, Dong J Y, Li B, et al. Automatic change detection in synthetic aperture Radar images based on PCANet[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(12):1792-1796.
[7] Zhang X Z, Su H, Zhang C, et al. Robust unsupervised small area change detection from SAR imagery using deep learning[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 173:79-94.
[8] Qu X F, Gao F, Dong J Y, et al. Change detection in synthetic aperture Radar images using a dual-domain network[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 19:4013405.
[9] 王建明, 史文中, 邵攀. 自适应距离和模糊拓扑优化的模糊聚类SAR影像变化检测[J]. 测绘学报, 2018, 47(5):611-619.
doi: 10.11947/j.AGCS.2018.20160607
[9] Wang J M, Shi W Z, Shao P. Change-detection method for SAR image using adaptive distance and fuzzy topology optimization-based fuzzy clustering[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(5):611-619.
doi: 10.11947/j.AGCS.2018.20160607
[10] 公茂果, 苏临之, 李豪, 等. 合成孔径雷达影像变化检测研究进展[J]. 计算机研究与发展, 2016, 53(1):123-137.
[10] Gong M G, Su L Z, Li H, et al. A survey on change detection in synthetic aperture Radar imagery[J]. Journal of Computer Research and Development, 2016, 53(1):123-137.
[11] Ban Y F, Yousif O A. Multitemporal spaceborne SAR data for urban change detection in China[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 5(4):1087-1094.
[12] Ma J J, Li D S, Tang X, et al. Unsupervised SAR image change detection based on feature fusion of information transfer[J]. IEEE Geoscience and Remote Sensing Letters, 2023, 20:4004905.
[13] Li H C, Celik T, Longbotham N, et al. Gabor feature based unsupervised change detection of multitemporal SAR images based on two-level clustering[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(12):2458-2462.
[14] Shao P, Shi W Z, Liu Z W, et al. Unsupervised change detection using fuzzy topology-based majority voting[J]. Remote Sensing, 2021, 13(16):3171.
[15] Gao F, Wang X, Gao Y H, et al. Sea ice change detection in SAR images based on convolutional-wavelet neural networks[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(8):1240-1244.
[16] Vudattu V S, Pati U C. Change detection in synthetic aperture Radar images based on a spatial pyramid pooling attention network (SPPANet)[J]. Remote Sensing Letters, 2023, 14(11):1139-1149.
[17] Chen C L P, Liu Z L. Broad learning system:An effective and efficient incremental learning system without the need for deep architecture[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018, 29(1):10-24.
[18] 邵攀. 非监督遥感变化检测模糊方法研究[D]. 武汉: 武汉大学, 2016.
[18] Shao P. Study on fuzzy unsupervised change detection methods based on remotely sensed image[D]. Wuhan: Wuhan University, 2016.
[19] Perez A, Gonzalez R C. An iterative thresholding algorithm for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1987, 9(6): 742-751.
pmid: 21869436
[20] Ghosh A, Mishra N S, Ghosh S. Fuzzy clustering algorithms for unsupervised change detection in remote sensing images[J]. Information Sciences, 2011, 181(4):699-715.
[21] Shao P, Shi W Z, He P F, et al. Novel approach to unsupervised change detection based on a robust semi-supervised FCM clustering algorithm[J]. Remote Sensing, 2016, 8(3):264.
[22] Bovolo F, Bruzzone L, Marconcini M. A novel approach to unsupervised change detection based on a semisupervised SVM and a similarity measure[J]. IEEE Transactions on Geoscience and Remote Sensing, 2008, 46(7):2070-2082.
[23] 宋熙煜, 周利莉, 李中国, 等. 图像分割中的超像素方法研究综述[J]. 中国图象图形学报, 2015, 20(5):599-608.
[23] Song X Y, Zhou L L, Li Z G, et al. Review on superpixel methods in image segmentation[J]. Journal of Image and Graphics, 2015, 20(5):599-608.
[24] 袁巧, 程艳芬, 陈先桥. 多先验特征与综合对比度的图像显著性检测[J]. 中国图象图形学报, 2018, 23(2):239-248.
[24] Yuan Q, Cheng Y F, Chen X Q. Saliency detection based on multiple priorities and comprehensive contrast[J]. Journal of Image and Graphics, 2018, 23(2):239-248.
[25] 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):772-776.
[26] Gao F, Dong J Y, Li B, et al. Change detection from synthetic aperture Radar images based on neighborhood- based ratio and extreme learning machine[J]. Journal of Applied Remote Sensing, 2016, 10(4):046019.
[27] Gong M G, Zhou Z Q, Ma J J. Change detection in synthetic aperture Radar images based on image fusion and fuzzy clustering[J]. IEEE Transactions on Image Processing:A Publication of the IEEE Signal Processing Society, 2012, 21(4):2141-2151.
[28] Zhao S J, Zhang X L, Xiao P F, et al. Exchanging dual-encoder-decoder:A new strategy for change detection with semantic gui-dance and spatial localization[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61:4508016.
[1] YANG Chen, JIN Yuan, DENG Fei, SHI Xuguo. Detection and monitoring of landslides along the Xuyong-Gulin Expressway using SBAS InSAR[J]. Remote Sensing for Natural Resources, 2025, 37(1): 161-168.
[2] ZHUANG Huifu, WANG Peng, SU Yanan, ZHANG Xiang, FAN Hongdong. Dynamic monitoring of flood inundation in Zhuozhou, Hebei Province based on multi-temporal SAR data[J]. Remote Sensing for Natural Resources, 2024, 36(4): 218-228.
[3] CAI Jian’ao, MING Dongping, ZHAO Wenyi, LING Xiao, ZHANG Yu, ZHANG Xingxing. Integrated remote sensing-based hazard identification and disaster-causing mechanisms of landslides in Zayu County[J]. Remote Sensing for Natural Resources, 2024, 36(1): 128-136.
[4] LI Lei, SUN Xiyan, JI Yuanfa, FU Wentao. Hyperspectral image classification based on superpixel segmentation and extended multi-attribute profiles[J]. Remote Sensing for Natural Resources, 2023, 35(4): 114-121.
[5] DONG Ting, FU Weiqi, SHAO Pan, GAO Lipeng, WU Changdong. Detection of changes in SAR images based on an improved fully-connected conditional random field[J]. Remote Sensing for Natural Resources, 2023, 35(3): 134-144.
[6] SUN Sheng, MENG Zhimin, HU Zhongwen, YU Xu. Application of multi-scale and lightweight CNN in SAR image-based surface feature classification[J]. Remote Sensing for Natural Resources, 2023, 35(1): 27-34.
[7] LI Yuan, WU Lin, QI Wenwen, GUO Zhengwei, LI Ning. A SAR image classification method based on an improved OGMRF-RC model[J]. Remote Sensing for Natural Resources, 2021, 33(4): 98-104.
[8] WANG Hua, LI Weiwei, LI Zhigang, CHEN Xueye, SUN Le. Hyperspectral image classification based on multiscale superpixels[J]. Remote Sensing for Natural Resources, 2021, 33(3): 63-71.
[9] XIA Yan, HUANG Liang, CHEN Pengdi. Tobacco fine extraction from UAV image based on fuzzy-superpixel segmentation algorithm[J]. Remote Sensing for Land & Resources, 2021, 33(1): 115-122.
[10] ZHANG Rui, YOU Shucheng, DU Lei, LU Jing, HE Yun, HU Yong. High-resolution remote sensing image segmentation based on improved superpixel and marker watershed[J]. Remote Sensing for Land & Resources, 2021, 33(1): 86-95.
[11] SUN Ke. Remote sensing image classification based on super pixel and peak density[J]. Remote Sensing for Land & Resources, 2020, 32(4): 41-45.
[12] JIANG Shan, WANG Chun, SONG Hongli, LIU Yufeng. A study of crop planting type recognition based on SAR and optical remote sensing data[J]. Remote Sensing for Land & Resources, 2020, 32(4): 105-110.
[13] Guangyu ZHOU, Bangquan LIU, Dan ZHANG. Target recognition in SAR images based on variational mode decomposition[J]. Remote Sensing for Land & Resources, 2020, 32(2): 33-39.
[14] Bingxiu YAO, Liang HUANG, Yansong XU. A high resolution remote sensing image segmentation method based on superpixel and graph theory[J]. Remote Sensing for Land & Resources, 2019, 31(3): 72-79.
[15] Nianqin WANG, Dejing QIAO, Xiyou FU. An analysis of the influence of filtering parameter on the performance of Goldstein InSAR interfergram filter[J]. Remote Sensing for Land & Resources, 2019, 31(1): 117-124.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
京ICP备05055290号-2
Copyright © 2017 Remote Sensing for Natural Resources
Support by Beijing Magtech