SGH-Net: An Efficient Hierarchical Fusion Network with Spectrally Guided Attention for Multi-Modal Landslide Segmentation.
Saved in:
| Title: | SGH-Net: An Efficient Hierarchical Fusion Network with Spectrally Guided Attention for Multi-Modal Landslide Segmentation. |
|---|---|
| Authors: | Wang, Jing1 (AUTHOR), Li, Haiyang2 (AUTHOR), Wu, Shuguang3 (AUTHOR), Yu, Yukui1,4 (AUTHOR), Nie, Guigen1,4 (AUTHOR), Fan, Zhaoquan1,2 (AUTHOR) fanzhaoquan@xidian.edu.cn |
| Source: | Remote Sensing. Apr2026, Vol. 18 Issue 8, p1115. 19p. |
| Subjects: | Multispectral imaging, Digital elevation models, Edge detection (Image processing), Convolutional neural networks |
| Abstract: | Highlights: What are the main findings? A lightweight multi-modal landslide segmentation framework is developed that avoids naive input-level concatenation and instead injects multispectral band and terrain guidance into RGB features across multiple encoder stages. The proposed Guided Attention Block improves boundary delineation and reduces background confusion by selectively recalibrating RGB representations using multispectral and DEM-derived cues. What are the implications of the main findings? This study indicates that modality-guided hierarchical fusion is more effective for landslide segmentation than the straightforward early fusion of heterogeneous inputs. The results indicate that physically grounded modality guidance can systematically mitigate background confusion, improving robustness to spectral confusion and fragmented target structure. Accurate landslide segmentation from remote sensing imagery is important for geohazard assessment and emergency response, yet it remains challenging because landslide regions are often spectrally confused with bare soil, riverbeds, shadows, and disturbed surfaces while also suffering from severe foreground–background imbalance. To address these issues, we propose an Efficient Spectrally Guided Hierarchical Fusion Network (SGH-Net) for multi-modal landslide segmentation. Instead of directly concatenating heterogeneous inputs at the image level, SGH-Net adopts an asymmetric encoder–decoder design in which a pretrained EfficientNet-B4 extracts RGB features, while two lightweight guidance encoders capture complementary multispectral band and DEM-derived terrain cues. These guidance features are progressively injected into the RGB backbone through multi-stage Guided Attention Blocks, enabling selective feature recalibration and reducing cross-modal interference. In addition, a hybrid Dice–Focal loss is used to alleviate class imbalance. Experiments on the Landslide4Sense dataset show that SGH-Net achieves the best overall performance among the compared methods under the adopted evaluation protocol, reaching 81.15% IoU and a 77.86% F1-score. Compared with representative multi-modal baselines, the proposed method delivers more accurate boundary delineation and fewer false alarms while maintaining favorable model complexity. These results indicate that modality-guided hierarchical fusion is an effective and efficient strategy for multi-modal landslide segmentation. [ABSTRACT FROM AUTHOR] |
| Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 193435594 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: SGH-Net: An Efficient Hierarchical Fusion Network with Spectrally Guided Attention for Multi-Modal Landslide Segmentation. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Wang%2C+Jing%22">Wang, Jing</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Haiyang%22">Li, Haiyang</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wu%2C+Shuguang%22">Wu, Shuguang</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Yu%2C+Yukui%22">Yu, Yukui</searchLink><relatesTo>1,4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Nie%2C+Guigen%22">Nie, Guigen</searchLink><relatesTo>1,4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Fan%2C+Zhaoquan%22">Fan, Zhaoquan</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> fanzhaoquan@xidian.edu.cn</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Remote+Sensing%22">Remote Sensing</searchLink>. Apr2026, Vol. 18 Issue 8, p1115. 19p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Multispectral+imaging%22">Multispectral imaging</searchLink><br /><searchLink fieldCode="DE" term="%22Digital+elevation+models%22">Digital elevation models</searchLink><br /><searchLink fieldCode="DE" term="%22Edge+detection+%28Image+processing%29%22">Edge detection (Image processing)</searchLink><br /><searchLink fieldCode="DE" term="%22Convolutional+neural+networks%22">Convolutional neural networks</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Highlights: What are the main findings? A lightweight multi-modal landslide segmentation framework is developed that avoids naive input-level concatenation and instead injects multispectral band and terrain guidance into RGB features across multiple encoder stages. The proposed Guided Attention Block improves boundary delineation and reduces background confusion by selectively recalibrating RGB representations using multispectral and DEM-derived cues. What are the implications of the main findings? This study indicates that modality-guided hierarchical fusion is more effective for landslide segmentation than the straightforward early fusion of heterogeneous inputs. The results indicate that physically grounded modality guidance can systematically mitigate background confusion, improving robustness to spectral confusion and fragmented target structure. Accurate landslide segmentation from remote sensing imagery is important for geohazard assessment and emergency response, yet it remains challenging because landslide regions are often spectrally confused with bare soil, riverbeds, shadows, and disturbed surfaces while also suffering from severe foreground–background imbalance. To address these issues, we propose an Efficient Spectrally Guided Hierarchical Fusion Network (SGH-Net) for multi-modal landslide segmentation. Instead of directly concatenating heterogeneous inputs at the image level, SGH-Net adopts an asymmetric encoder–decoder design in which a pretrained EfficientNet-B4 extracts RGB features, while two lightweight guidance encoders capture complementary multispectral band and DEM-derived terrain cues. These guidance features are progressively injected into the RGB backbone through multi-stage Guided Attention Blocks, enabling selective feature recalibration and reducing cross-modal interference. In addition, a hybrid Dice–Focal loss is used to alleviate class imbalance. Experiments on the Landslide4Sense dataset show that SGH-Net achieves the best overall performance among the compared methods under the adopted evaluation protocol, reaching 81.15% IoU and a 77.86% F1-score. Compared with representative multi-modal baselines, the proposed method delivers more accurate boundary delineation and fewer false alarms while maintaining favorable model complexity. These results indicate that modality-guided hierarchical fusion is an effective and efficient strategy for multi-modal landslide segmentation. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=193435594 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.3390/rs18081115 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 19 StartPage: 1115 Subjects: – SubjectFull: Multispectral imaging Type: general – SubjectFull: Digital elevation models Type: general – SubjectFull: Edge detection (Image processing) Type: general – SubjectFull: Convolutional neural networks Type: general Titles: – TitleFull: SGH-Net: An Efficient Hierarchical Fusion Network with Spectrally Guided Attention for Multi-Modal Landslide Segmentation. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Wang, Jing – PersonEntity: Name: NameFull: Li, Haiyang – PersonEntity: Name: NameFull: Wu, Shuguang – PersonEntity: Name: NameFull: Yu, Yukui – PersonEntity: Name: NameFull: Nie, Guigen – PersonEntity: Name: NameFull: Fan, Zhaoquan IsPartOfRelationships: – BibEntity: Dates: – D: 15 M: 04 Text: Apr2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 20724292 Numbering: – Type: volume Value: 18 – Type: issue Value: 8 Titles: – TitleFull: Remote Sensing Type: main |
| ResultId | 1 |
