Please wait a minute...
img

官方微信

遥感技术与应用  2023, Vol. 38 Issue (5): 1215-1225    DOI: 10.11873/j.issn.1004-0323.2023.5.1215
遥感应用     
一种图像回归与关联关系特征融合的遥感影像变化检测方法
马宗方(),郝凡(),宋琳,麻瑞
西安建筑科技大学信息与控制工程学院,陕西 西安 710055
Image Regression and Association-based Feature Fusion for Remote Sensing Image Change Detection
Zongfang MA(),Fan HAO(),Lin SONG,Rui MA
College of Information and Control Engineering,Xi’an University of Architecture and Technology,Xi’an 710055,China
 全文: PDF(5951 KB)   HTML
摘要:

异质遥感影像变化检测是一个重要且具有挑战性的研究课题。针对直接比较异质数据进行变化检测导致检测精度低的问题,提出了一种图像回归与关联关系特征融合(Image Regression and Association-based Feature, IRAF)的异质遥感影像变化检测方法。首先基于信息熵理论量化异质数据的信息量差异并确定回归方向,采用多输出多层感知器图像回归得到与原始影像特征空间分布相近的回归图像;其次,得到差异图像并基于模糊局部信息C均值(Fuzzy Local Information C-Means,FLICM)算法找到部分显著样本对用于后续检测。为了考虑不同特征间的关联关系并充分利用数据中潜在的高阶信息,采用基于关联关系特征的融合算法(Association-based Fusion,AF)对原始遥感数据进行增强,最后利用融合后的特征训练分类模型得到最终的变化二值图。为验证该方法的有效性,采用Sardinia、Yellow River和Texas这3组真实数据集进行实验,Ka 分别达到了0.796 1、0.827 1、0.958 1。与相关方法进行对比的实验结果表明该方法在不同数据集上均得到了最优的检测结果,能够抑制噪声的影响且有效提升变化检测精度。

关键词: 变化检测图像回归异质数据关联关系特征    
Abstract:

Change detection from heterogeneous remote sensing images is an important and challenging research topic with a wide range of applications in disaster assessment, urban planning and environmental monitoring. However, the direct comparison of heterogeneous data for change detection always has a poor detection accuracy. To address this issue, a multioutput adaptive regression and association-based feature fusion method for heterogeneous remote sensing change detection is proposed. Firstly, the proposed method determines the adaptive regression direction according to the information entropy, which utilizes the difference of information between heterogeneous data. To transform heterogeneous data into a common feature space, the regression image will be obtained via a multioutput multilayer perceptron image regression algorithm. Then, the fuzzy local information C-means algorithm is used to identify the fuzzy region in the difference image, which further ensures the reliability of significant sample pairs. Finally, an association-based fusion method was applied to the heterogeneous remote sensing change detection dataset by simultaneously exploiting the high-order information of heterogeneous data and the association information between features. The binary change map is obtained via training a classification model with the boosting dataset. Experiments conducted on three real datasets (Sardinia, Yellow River and Texas) show the effectiveness of the proposed method by comparing it with seven related change detection methods. Experimental results indicate that the proposed method owns the best change detection results on both three datasets, which proves its effectiveness, and it can suppress the influence of noise and improve the accuracy of change detection.

Key words: Change detection    Image regression    Heterogeneous data    Association-based feature
收稿日期: 2022-05-10 出版日期: 2023-11-07
ZTFLH:  TP751  
基金资助: 陕西省重点研发计划(2020GY?186);西安建筑科技大学科技基金(ZR21034)
通讯作者: 郝凡     E-mail: zongfangma@xauat.edu.cn;haofan@xauat.edu.cn
作者简介: 马宗方(1980-),男,安徽临泉人,博士,教授,主要从事智能信息处理、机器视觉工业应用研究。E?mail: zongfangma@xauat.edu.cn
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  
马宗方
郝凡
宋琳
麻瑞

引用本文:

马宗方,郝凡,宋琳,麻瑞. 一种图像回归与关联关系特征融合的遥感影像变化检测方法[J]. 遥感技术与应用, 2023, 38(5): 1215-1225.

Zongfang MA,Fan HAO,Lin SONG,Rui MA. Image Regression and Association-based Feature Fusion for Remote Sensing Image Change Detection. Remote Sensing Technology and Application, 2023, 38(5): 1215-1225.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2023.5.1215        http://www.rsta.ac.cn/CN/Y2023/V38/I5/1215

图1  基于图像回归和关联关系特征融合的变化检测方法IRAF
图2  基于多输出多层感知器的图像回归
数据集传感器采集日期采集地点尺寸空间分辨率
SardiniaLandsat-5 TM/ Google Earth1995.9/1996.7意大利撒丁岛412×30030 m
Yellow RiverRadarsat-2/ Google Earth2008.6/2010.9中国黄河入海口291×3438 m
TexasLandsat-5 TM/ EO-1 ALI2011.9/2011.10德克萨斯州巴斯多普县1534×80830 m
表1  3组真实数据集的描述
图3  Sardinia数据集及IRAF在该数据集上的实验结果
图4  Yellow River数据集及IRAF在该数据集上的实验结果
图5  Texas 数据集及IRAF在该数据集上的实验结果
方法FNFPOERaRpRmRfKa
PCC3 32329 55032 8730.7340.127 10.435 70.872 90.118 7
CVA2 23443 27845 5120.631 80.110 80.292 90.889 20.095
IR-MAD1 74026 88428 6240.768 40.179 60.228 20.820 40.212 6
SCCN6 2202 0788 2980.932 90.727 50.528 60.272 50.537 5
FPMSMCD1 5223 3904 9120.960 30.642 90.199 60.357 10.692
SCASC3 1273 4206 5470.9470.568 10.410.431 90.550 6
GIR-MRF1 8523 2915 1430.958 40.6370.242 90.3630.669 7
IRAF1 5781 2892 8670.976 80.824 30.206 90.175 70.796 1
表2  不同方法在Sardinia数据集的变化检测精度
图6  Sardinia数据集上不同方法的变化检测结果
图7  Yellow River数据集上不同方法的变化检测结果
方法FNFPOERaRpRmRfKa
PCC97725 72926 7060.732 40.084 70.290 90.915 30.097 1
CVA1 36624 32125 6870.742 60.075 70.406 70.924 30.079 4
IR-MAD1 00328 98929 9920.699 50.075 20.298 60.924 80.079 8
SCCN8775091 3860.986 10.829 80.261 10.170 20.774 6
FPMSMCD31212 39212 7040.872 70.197 40.092 90.802 60.284 6
SCASC4691 7192 1880.978 10.6270.139 60.3730.714 3
GIR-MRF4781 4081 8860.981 10.671 70.142 30.328 30.743 7
IRAF5256111 1360.988 60.822 60.156 30.177 40.827 1
表3  不同方法在Yellow River数据集的变化检测精度
方法FNFPOERaRpRmRfKa
PCC32 61074 573107 1830.913 50.5710.247 30.4290.601 1
CVA30 074149 178179 2520.855 40.405 60.228 10.594 40.455 9
IR-MAD83 18456 027139 2110.887 70.464 90.630 80.535 10.350 4
SCCN8 339180 476188 8150.847 70.406 30.063 20.593 70.491 3
FPMSMCD21 50251 86273 3640.940 80.680 30.163 10.319 70.717 4
SCASC86 7059 94196 6460.9220.819 60.657 50.180 40.448 5
GIR-MRF1 93523 81425 7490.979 20.845 10.014 70.154 90.898 2
IRAF4 3225 5889 9100.9920.9580.032 80.0420.958 1
表4  不同方法在Texas数据集的变化检测精度
图8  Texas数据集上不同方法的变化检测结果
IRAFFNFPOERaRpRmRfKa
规则(1): RX4 17416 07420 2480.836 20.176 80.547 30.823 20.181 7
规则(2): RY1 5781 2892 8670.976 80.824 30.206 90.175 70.796 1
规则(3): RX + RY3 8932134 1060.966 80.9460.510 50.0540.629 6
表5  IRAF在不同图像回归方向选取规则下的变化检测精度
图9  特征增强率L对本文方法检测结果的影响
消融说明FNFPOERaRpRmRfKa
无回归+无AF1 25340 13141 3840.665 20.1370.164 30.8630.144 8
无回归+AF2 87829 77932 6570.735 80.137 50.377 40.862 50.138 2
回归+无AF1 4272 1703 5970.970 90.740 70.187 10.259 30.759 6
IRAF1 5781 2892 8670.976 80.824 30.206 90.175 70.796 1
表6  IRAF在Sardinia数据集上的消融实验
1 HAO F, MA Z F, TIAN H P, et al. Semi-supervised label propagation for multi-source remote sensing image change detection[J]. Computers & Geosciences, 2023, 170: 105249. DOI: 10.1016/j.cageo.2022.105249
doi: 10.1016/j.cageo.2022.105249
2 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.
2 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.
3 WANG Yuan, CHEN Fulong, HU Qi, et al. Urban change detection based on COSMO-SkyMed multi-temporal images in Nanjing city, China[J]. Remote Sensing Technology and Application, 2019, 34(5): 1054-1063.
3 王源,陈富龙,胡祺,等.COSMO-SkyMed时序影像南京城市变化检测研究[J]. 遥感技术与应用, 2019, 34(5) :1054-1063.
4 WANG Chongyang, TIAN Xin. Forest cover change detection based on GF-1 PMS data[J]. Remote Sensing Technology and Application,2021,36(1):208-216.
4 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 .
5 SUN Y, LEI L, TAN X, et al. Structured graph based image regression for unsupervised multimodal change detection[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2022, 185: 16-31.
6 ZHANG Mingzhe, ZHANG Hong, WANG Chao, et al. Combining super-pixel segmentation and multiple difference maps for SAR change detection[J]. Remote Sensing Technology and Application, 2016,31(3):481 -487.
6 张明哲,张红,王超,等.基于超像素分割和多方法融合的SAR图像变化检测方法[J]. 遥感技术与应用, 2016,31(3) : 481-487.
7 MOSER G, ANFINSEN S N, LUPPINO L T, et al. Change detection with heterogeneous remote sensing data: From semi-parametric regression to deep learning[C]∥IGARSS 2020-2020 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2020: 3892-3895.
8 SUN Y, LEI L, LI X, et al. Patch similarity graph matrix-based unsupervised remote sensing change detection with homogeneous and heterogeneous sensors[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(6): 4841-4861.
9 ZHOU W, TROY A, GROVE M. Object-based land cover classification and change analysis in the Baltimore metropolitan area using multitemporal high resolution remote sensing data[J]. Sensors, 2008, 8(3): 1613-1636. DOI:10.3390/s8031613
doi: 10.3390/s8031613
10 WAN L, XIANG Y, YOU H. An object-based hierarchical compound classification method for change detection in heterogeneous optical and SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(12): 9941-9959.
11 LIU Z, LI G, MERCIER G, et al. Change detection in heterogenous remote sensing images via homogeneous pixel transformation[J]. IEEE Transactions on Image Processing, 2018, 27(4): 1822-1834.
12 LUPPINO L T, BIANCHI F M, MOSER G, et al. Unsupervised image regression for heterogeneous change detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(12): 9960-9975.
13 LIU J, GONG M, QIN K, 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, 2018, 29(3): 545-559.
14 GONG M, NIU X, ZHAN T,et al. A coupling translation network for change detection in heterogeneous images[J]. International Journal of Remote Sensing,2019,40(9):3647-3672.
15 LIU Z G, ZHANG Z W, PAN Q, et al. Unsupervised change detection from heterogeneous data based on image translation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 99: 1-13.
16 WANG X, CHENG W, FENG Y, et al. TSCNet: topological structure coupling network for change detection of heterogeneous remote sensing images[J]. Remote Sensing, 2023, 15(3): 621. DOI: 10.3390/rs15030621
doi: 10.3390/rs15030621
17 LIANG X Y, QIAN Y H, GUO Q, et al. AF: An association-based fusion method for multi-modal classification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021(1): 1-1. DOI:10.1109/tpami.2021.3125995
doi: 10.1109/tpami.2021.3125995
18 CHENG G, WANG G, HAN J. ISNet: Towards improving separability for remote sensing image change detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60:1-11.
19 ZHANG M, LI Q, MIAO Y, et al. Difference-guided aggregation network with multi-image pixel contrast for change detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61:1-14.
20 BERGAMASCO L, SAHA S, BOVOLO F, et al. Unsupervised change detection using convolutional-autoencoder multiresolution features[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-19.
21 SUN Y, LEI L, GUAN D, et al. Sparse-constrained adaptive structure consistency-based unsupervised image regression for heterogeneous remote-sensing change detection [J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 60: 1-14.
22 KRINIDIS S, CHATZIS V. A robust fuzzy local information C-means clustering algorithm[J]. IEEE Transactions on Image Processing, 2010, 19(5): 1328-1337.
23 WU Y, LI J, YUAN Y, et al. Commonality autoencoder: Learning common features for change detection from heterogeneous images[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 33(9):4257-4270.
24 ZHANG Y, ZHANG J X. Measure of information content of remotely sensed images accounting for spatial correlation[J]. Acta Geodaetica et Cartographica Sinica, 2015,44(10):1117-1124.
24 张盈,张景雄. 顾及空间相关性的遥感影像信息量的度量方法[J].测绘学报, 2015, 44(10): 1117-1124.
25 ZHAN T, GONG M, JIANG X, et al. Unsupervised scale-driven change detection with deep spatial-spectral features for VHR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(8): 5653-5665.
26 ZANETTI M, BRUZZONE L. A theoretical framework for change detection based on a compound multiclass statistical model of the difference image[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(2): 1129-1143.
27 VOLPI M, CAMPS-VALLS G, TUIA D. Spectral alignment of multi-temporal cross-sensor images with automated kernel canonical correlation analysis[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 107: 50-63.
28 SERRA P, PONS X, SAURI D. Post-classification change detection with data from different sensors:Some accuracy considerations[J]. International Journal of Remote Sensing, 2003, 24(16): 3311-3340.
29 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, 1(45): 218-236.
30 NIELSEN A A. The regularized iteratively reweighted MAD method for change detection in multi-and hyperspectral data[J]. IEEE Transactions on Image Processing, 2007, 16(2): 463-478.
31 MIGNOTTE M. A fractal projection and Markovian segmentation-based approach for multimodal change detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(11): 8046-8058.
[1] 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.
[2] 高梦洁,陈方,王雷,杨阿强,于博. 基于马尔科夫随机场的PLANET高分影像滑坡提取研究[J]. 遥感技术与应用, 2023, 38(5): 1180-1191.
[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] 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.
[5] 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.
[6] 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.
[7] 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.
[8] 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.
[9] 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.
[10] 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.
[11] 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.
[12] 罗家顺,邱建秀,赵天杰,王大刚. 基于Sentinel-1数据的黑河中游土壤水分反演[J]. 遥感技术与应用, 2020, 35(1): 23-32.
[13] 何浩,刘修国,沈永林. 基于视差的高分辨率遥感影像建筑物变化检测[J]. 遥感技术与应用, 2019, 34(6): 1315-1323.
[14] 王源,陈富龙,胡祺,唐攀攀. COSMO-SkyMed时序影像南京城市变化检测研究[J]. 遥感技术与应用, 2019, 34(5): 1054-1063.
[15] 张因果,陈芸芝. 一种基于改进土地覆盖更新方法的新增建设用地自动提取[J]. 遥感技术与应用, 2019, 34(5): 1073-1081.