Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved Multiresunet
The current imagery of landslides presents multiple challenges, including considerable scale variations among landslides, similarities in spectral characteristics to those of bare ground surfaces, and irregular edges. Despite significant progress in semantic segmentation achieved by convolutional ne...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10820510/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592907882725376 |
|---|---|
| author | Zhenyu Zhao Shucheng Tan Qinghua Zhang Hui Chen |
| author_facet | Zhenyu Zhao Shucheng Tan Qinghua Zhang Hui Chen |
| author_sort | Zhenyu Zhao |
| collection | DOAJ |
| description | The current imagery of landslides presents multiple challenges, including considerable scale variations among landslides, similarities in spectral characteristics to those of bare ground surfaces, and irregular edges. Despite significant progress in semantic segmentation achieved by convolutional neural networks (CNNs), the local sensory field of CNNs poses difficulties in differentiating between landslides and bare surfaces. As a result, efficient and accurate landslide extraction remains a challenging issue within the global research community. To address this problem, a novel automatic landslide hazard remote sensing image identification model, MultiResUNet-BFDC, is proposed. Initially, null convolutions with various null rates are introduced to replace some standard convolutions, thereby expanding the sensory field without increasing the parameter count and making the model more suitable for multi-scale landslide segmentation. Additionally, the Canny operator is employed to design a lightweight boundary-focused attention (BFA) mechanism, enhancing the model’s ability to emphasize landslide edge features. Furthermore, a new hybrid loss function, adaptive focal and Dice loss (AFD loss), is introduced through the adaptive AdaLoss algorithm by combining focal loss and Dice loss, improving the model’s ability to handle unbalanced samples. The experimental results indicate that the MultiResUNet-BFDC model demonstrates enhanced performance in landslide edge detail segmentation, with fewer misidentifications and omissions and superior functionality of the BFA attentional mechanism compared with other attentional mechanisms. |
| format | Article |
| id | doaj-art-15a51d97c2bd444e8e99a7310a04b721 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-15a51d97c2bd444e8e99a7310a04b7212025-01-21T00:01:13ZengIEEEIEEE Access2169-35362025-01-0113106531066210.1109/ACCESS.2024.352506710820510Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved MultiresunetZhenyu Zhao0https://orcid.org/0009-0004-8318-4593Shucheng Tan1Qinghua Zhang2Hui Chen3Institute of International Rivers and Eco-Security, Yunnan University, Kunming, Yunnan, ChinaYunnan International Joint Laboratory of Critical Mineral Resources, Kunming, ChinaInstitute of International Rivers and Eco-Security, Yunnan University, Kunming, Yunnan, ChinaInstitute of International Rivers and Eco-Security, Yunnan University, Kunming, Yunnan, ChinaThe current imagery of landslides presents multiple challenges, including considerable scale variations among landslides, similarities in spectral characteristics to those of bare ground surfaces, and irregular edges. Despite significant progress in semantic segmentation achieved by convolutional neural networks (CNNs), the local sensory field of CNNs poses difficulties in differentiating between landslides and bare surfaces. As a result, efficient and accurate landslide extraction remains a challenging issue within the global research community. To address this problem, a novel automatic landslide hazard remote sensing image identification model, MultiResUNet-BFDC, is proposed. Initially, null convolutions with various null rates are introduced to replace some standard convolutions, thereby expanding the sensory field without increasing the parameter count and making the model more suitable for multi-scale landslide segmentation. Additionally, the Canny operator is employed to design a lightweight boundary-focused attention (BFA) mechanism, enhancing the model’s ability to emphasize landslide edge features. Furthermore, a new hybrid loss function, adaptive focal and Dice loss (AFD loss), is introduced through the adaptive AdaLoss algorithm by combining focal loss and Dice loss, improving the model’s ability to handle unbalanced samples. The experimental results indicate that the MultiResUNet-BFDC model demonstrates enhanced performance in landslide edge detail segmentation, with fewer misidentifications and omissions and superior functionality of the BFA attentional mechanism compared with other attentional mechanisms.https://ieeexplore.ieee.org/document/10820510/Landslide identificationdeep learningCNNattention mechanismremote sensing imagery |
| spellingShingle | Zhenyu Zhao Shucheng Tan Qinghua Zhang Hui Chen Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved Multiresunet IEEE Access Landslide identification deep learning CNN attention mechanism remote sensing imagery |
| title | Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved Multiresunet |
| title_full | Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved Multiresunet |
| title_fullStr | Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved Multiresunet |
| title_full_unstemmed | Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved Multiresunet |
| title_short | Automatic Identification Model for Landslide Disaster Using Remote Sensing Images Based on Improved Multiresunet |
| title_sort | automatic identification model for landslide disaster using remote sensing images based on improved multiresunet |
| topic | Landslide identification deep learning CNN attention mechanism remote sensing imagery |
| url | https://ieeexplore.ieee.org/document/10820510/ |
| work_keys_str_mv | AT zhenyuzhao automaticidentificationmodelforlandslidedisasterusingremotesensingimagesbasedonimprovedmultiresunet AT shuchengtan automaticidentificationmodelforlandslidedisasterusingremotesensingimagesbasedonimprovedmultiresunet AT qinghuazhang automaticidentificationmodelforlandslidedisasterusingremotesensingimagesbasedonimprovedmultiresunet AT huichen automaticidentificationmodelforlandslidedisasterusingremotesensingimagesbasedonimprovedmultiresunet |