A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing Images
Change detection (CD) is a crucial task in remote sensing (RS) to distinguish surface changes from bitemporal images. Recently, deep learning (DL) based methods have achieved remarkable success for CD. However, the existing methods lack robustness to various kinds of changes in RS images, which suff...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-12-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/15/1/45 |
_version_ | 1797431283494682624 |
---|---|
author | Jinming Wu Chunhui Xie Zuxi Zhang Yongxin Zhu |
author_facet | Jinming Wu Chunhui Xie Zuxi Zhang Yongxin Zhu |
author_sort | Jinming Wu |
collection | DOAJ |
description | Change detection (CD) is a crucial task in remote sensing (RS) to distinguish surface changes from bitemporal images. Recently, deep learning (DL) based methods have achieved remarkable success for CD. However, the existing methods lack robustness to various kinds of changes in RS images, which suffered from problems of feature misalignment and inefficient supervision. In this paper, a deeply supervised attentive high-resolution network (DSAHRNet) is proposed for remote sensing image change detection. First, we design a spatial-channel attention module to decode change information from bitemporal features. The attention module is able to model spatial-wise and channel-wise contexts. Second, to reduce feature misalignment, the extracted features are refined by stacked convolutional blocks in parallel. Finally, a novel deeply supervised module is introduced to generate more discriminative features. Extensive experimental results on three challenging benchmark datasets demonstrate that the proposed DSAHRNet outperforms other state-of-the-art methods, and achieves a great trade-off between performance and complexity. |
first_indexed | 2024-03-09T09:41:42Z |
format | Article |
id | doaj.art-6c556be1051a464584b3a2857986d558 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-09T09:41:42Z |
publishDate | 2022-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-6c556be1051a464584b3a2857986d5582023-12-02T00:50:39ZengMDPI AGRemote Sensing2072-42922022-12-011514510.3390/rs15010045A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing ImagesJinming Wu0Chunhui Xie1Zuxi Zhang2Yongxin Zhu3Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaChange detection (CD) is a crucial task in remote sensing (RS) to distinguish surface changes from bitemporal images. Recently, deep learning (DL) based methods have achieved remarkable success for CD. However, the existing methods lack robustness to various kinds of changes in RS images, which suffered from problems of feature misalignment and inefficient supervision. In this paper, a deeply supervised attentive high-resolution network (DSAHRNet) is proposed for remote sensing image change detection. First, we design a spatial-channel attention module to decode change information from bitemporal features. The attention module is able to model spatial-wise and channel-wise contexts. Second, to reduce feature misalignment, the extracted features are refined by stacked convolutional blocks in parallel. Finally, a novel deeply supervised module is introduced to generate more discriminative features. Extensive experimental results on three challenging benchmark datasets demonstrate that the proposed DSAHRNet outperforms other state-of-the-art methods, and achieves a great trade-off between performance and complexity.https://www.mdpi.com/2072-4292/15/1/45change detectionconvolutional neural networkfeature fusionmetric learningattention mechanism |
spellingShingle | Jinming Wu Chunhui Xie Zuxi Zhang Yongxin Zhu A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing Images Remote Sensing change detection convolutional neural network feature fusion metric learning attention mechanism |
title | A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing Images |
title_full | A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing Images |
title_fullStr | A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing Images |
title_full_unstemmed | A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing Images |
title_short | A Deeply Supervised Attentive High-Resolution Network for Change Detection in Remote Sensing Images |
title_sort | deeply supervised attentive high resolution network for change detection in remote sensing images |
topic | change detection convolutional neural network feature fusion metric learning attention mechanism |
url | https://www.mdpi.com/2072-4292/15/1/45 |
work_keys_str_mv | AT jinmingwu adeeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages AT chunhuixie adeeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages AT zuxizhang adeeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages AT yongxinzhu adeeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages AT jinmingwu deeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages AT chunhuixie deeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages AT zuxizhang deeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages AT yongxinzhu deeplysupervisedattentivehighresolutionnetworkforchangedetectioninremotesensingimages |