Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection Network
Landslide mapping (LM) from bitemporal remote sensing images is essential for disaster prevention and mitigation. Although bitemporal change detection technology has been applied for LM, there remains room for improvement in its accuracy and automation. In this article, a multilevel feature enhancem...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10044973/ |
| _version_ | 1797845214917820416 |
|---|---|
| author | Lukang Wang Min Zhang Xiaoqi Shen Wenzhong Shi |
| author_facet | Lukang Wang Min Zhang Xiaoqi Shen Wenzhong Shi |
| author_sort | Lukang Wang |
| collection | DOAJ |
| description | Landslide mapping (LM) from bitemporal remote sensing images is essential for disaster prevention and mitigation. Although bitemporal change detection technology has been applied for LM, there remains room for improvement in its accuracy and automation. In this article, a multilevel feature enhancement network (MFENet) is proposed for LM based on modules built in convolutional neural networks (CNNs) like CNN-Attention. MFENet mainly consists of three modules: the postevent feature enhancement module (PFEM), the bifeature difference enhancement module (BFDEM), and the flow direction calibration module (FDCM). Specifically, the main role of PFEM is to selectively fuse postevent multilayer features to provide discriminative postevent features. BFDEM fuses the multilayer differences of both pre-event and postevent features to generate high-quality change detection features, which are sufficiently powerful to distinguish foreground from background. FDCM uses a digital elevation model to calibrate the flow direction of each pixel of the landslide detection results to complete the LM task. Experiments were conducted to test the effectiveness of MFENet on two real-world regions, Lantau Island and Sharp Peak, Hong Kong, where landslides occurred after rainstorms. Compared with other state-of-the-art general change detection methods and landslide-specific change detection methods, the proposed method outperforms all metrics, with its intersection over union reaching 87.23%. The availability of additional features and the generalization performance of MFENet are demonstrated experimentally. It is anticipated that the proposed network will further contribute to disaster prevention and mitigation. |
| first_indexed | 2024-04-09T17:36:31Z |
| format | Article |
| id | doaj.art-ad8684d3259a40ba8bb320a4de0c1002 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-04-09T17:36:31Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-ad8684d3259a40ba8bb320a4de0c10022023-04-17T23:00:12ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352023-01-01163599361010.1109/JSTARS.2023.324506210044973Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection NetworkLukang Wang0https://orcid.org/0000-0001-8556-6422Min Zhang1https://orcid.org/0000-0003-1643-5271Xiaoqi Shen2https://orcid.org/0000-0002-6156-8906Wenzhong Shi3https://orcid.org/0000-0002-3886-7027School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou, ChinaSmart Cities Research Institute and the Department of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Hong KongSchool of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou, ChinaSmart Cities Research Institute and the Department of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Hong KongLandslide mapping (LM) from bitemporal remote sensing images is essential for disaster prevention and mitigation. Although bitemporal change detection technology has been applied for LM, there remains room for improvement in its accuracy and automation. In this article, a multilevel feature enhancement network (MFENet) is proposed for LM based on modules built in convolutional neural networks (CNNs) like CNN-Attention. MFENet mainly consists of three modules: the postevent feature enhancement module (PFEM), the bifeature difference enhancement module (BFDEM), and the flow direction calibration module (FDCM). Specifically, the main role of PFEM is to selectively fuse postevent multilayer features to provide discriminative postevent features. BFDEM fuses the multilayer differences of both pre-event and postevent features to generate high-quality change detection features, which are sufficiently powerful to distinguish foreground from background. FDCM uses a digital elevation model to calibrate the flow direction of each pixel of the landslide detection results to complete the LM task. Experiments were conducted to test the effectiveness of MFENet on two real-world regions, Lantau Island and Sharp Peak, Hong Kong, where landslides occurred after rainstorms. Compared with other state-of-the-art general change detection methods and landslide-specific change detection methods, the proposed method outperforms all metrics, with its intersection over union reaching 87.23%. The availability of additional features and the generalization performance of MFENet are demonstrated experimentally. It is anticipated that the proposed network will further contribute to disaster prevention and mitigation.https://ieeexplore.ieee.org/document/10044973/Change detectionconvolutional neural network (CNN)flow directionlandslide mapping (LM)remote sensing images |
| spellingShingle | Lukang Wang Min Zhang Xiaoqi Shen Wenzhong Shi Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection Network IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Change detection convolutional neural network (CNN) flow direction landslide mapping (LM) remote sensing images |
| title | Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection Network |
| title_full | Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection Network |
| title_fullStr | Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection Network |
| title_full_unstemmed | Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection Network |
| title_short | Landslide Mapping Using Multilevel-Feature-Enhancement Change Detection Network |
| title_sort | landslide mapping using multilevel feature enhancement change detection network |
| topic | Change detection convolutional neural network (CNN) flow direction landslide mapping (LM) remote sensing images |
| url | https://ieeexplore.ieee.org/document/10044973/ |
| work_keys_str_mv | AT lukangwang landslidemappingusingmultilevelfeatureenhancementchangedetectionnetwork AT minzhang landslidemappingusingmultilevelfeatureenhancementchangedetectionnetwork AT xiaoqishen landslidemappingusingmultilevelfeatureenhancementchangedetectionnetwork AT wenzhongshi landslidemappingusingmultilevelfeatureenhancementchangedetectionnetwork |