EGMamba-Net: Edge-Guided Global–Local Mamba Network with Region-Adaptive Routing for Salient Object Detection in Optical Remote Sensing Images.
Saved in:
| Title: | EGMamba-Net: Edge-Guided Global–Local Mamba Network with Region-Adaptive Routing for Salient Object Detection in Optical Remote Sensing Images. |
|---|---|
| Authors: | Zhang, Fubin1 (AUTHOR), Zhang, Zichi1 (AUTHOR), Zhang, Feihu1 (AUTHOR) feihu.zhang@nwpu.edu.cn |
| Source: | Remote Sensing. May2026, Vol. 18 Issue 10, p1568. 37p. |
| Subjects: | Optical remote sensing, Object recognition (Computer vision), Image enhancement (Imaging systems), Optimization algorithms |
| Abstract: | Highlights: What are the main findings? We propose EGMamba-Net, an edge-guided global–local collaborative network for salient object detection in optical remote sensing images, which integrates a hybrid CNN–Mamba backbone, edge prior enhancement, global–local interaction, region-adaptive routing, and difficulty-aware joint optimization. EGMamba-Net achieves strong performance on ORSSD, EORSSD, and ORSI-4199. On the more challenging EORSSD dataset, it reaches an S-measure of 0.9389, a max F-measure of 0.8972, and an MAE of 0.0066, outperforming the representative remote-sensing baseline DAF-Net by 0.0223 in S-measure and 0.0358 in max F-measure. What are the implications of the main finding? The results indicate that combining efficient global dependency modeling with explicit boundary enhancement and region-differentiated decoding is more effective than relying on convolution-only or uniform decoding strategies for complex remote sensing scenes with cluttered backgrounds, weak boundaries, low contrast, and densely distributed objects. The proposed hybrid Mamba design provides a favorable balance between detection accuracy and computational efficiency, showing practical potential for salient object detection in high-resolution optical remote sensing imagery. Salient object detection in optical remote sensing images remains challenging due to complex backgrounds, blurred boundaries, small objects, unstable foreground–background contrast, and dense object distributions. Existing convolution-based methods are effective at modeling local structures, but they are limited in capturing long-range dependencies, whereas Transformer-based approaches usually incur substantial computational cost when handling high-resolution remote sensing imagery. To address these issues, this paper proposes EGMamba-Net, an edge-guided global–local collaborative network for salient object detection in optical remote sensing images. Specifically, a hybrid global–local backbone is first constructed to preserve shallow texture, edge, and geometric details while introducing Mamba-based global modeling in deeper stages for efficient long-range dependency representation. An Edge Prior Enhancement Module (EPEM) is then designed to explicitly extract boundary priors from shallow features and refine feature representations through edge-guided modulation. To alleviate the representation conflict between global semantics and local details, a Global–Local Interaction Module (GLIM) is further developed, where convolutional local modeling and Mamba-based global modeling interact through cross-gating for complementary feature learning. Moreover, a Region-Adaptive Routing Decoder (RARD) is introduced to dynamically assign different refinement paths according to regional saliency response, boundary intensity, and contextual complexity, thereby improving the recovery of small, low-contrast, and densely distributed objects. In addition, a Difficulty-Aware Joint Loss (DAJL) is designed to enhance optimization on boundary regions and hard samples, improving robustness under challenging conditions. Extensiveexperiments on ORSSD, EORSSD, and ORSI-4199 datasets demonstrate the superiority of the proposed method. In particular, on the more challenging EORSSD dataset, EGMamba-Net achieves 0.9389 S-measure, 0.8972 max F-measure, and 0.0066 MAE. Compared with the representative remote-sensing method DAF-Net, it improves S-measure and max F-measure by 0.0223 and 0.0358, respectively, indicating stronger capability in background suppression, structural preservation, and boundary recovery. [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: 194141093 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: EGMamba-Net: Edge-Guided Global–Local Mamba Network with Region-Adaptive Routing for Salient Object Detection in Optical Remote Sensing Images. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Zhang%2C+Fubin%22">Zhang, Fubin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Zichi%22">Zhang, Zichi</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Feihu%22">Zhang, Feihu</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> feihu.zhang@nwpu.edu.cn</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Remote+Sensing%22">Remote Sensing</searchLink>. May2026, Vol. 18 Issue 10, p1568. 37p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Optical+remote+sensing%22">Optical remote sensing</searchLink><br /><searchLink fieldCode="DE" term="%22Object+recognition+%28Computer+vision%29%22">Object recognition (Computer vision)</searchLink><br /><searchLink fieldCode="DE" term="%22Image+enhancement+%28Imaging+systems%29%22">Image enhancement (Imaging systems)</searchLink><br /><searchLink fieldCode="DE" term="%22Optimization+algorithms%22">Optimization algorithms</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Highlights: What are the main findings? We propose EGMamba-Net, an edge-guided global–local collaborative network for salient object detection in optical remote sensing images, which integrates a hybrid CNN–Mamba backbone, edge prior enhancement, global–local interaction, region-adaptive routing, and difficulty-aware joint optimization. EGMamba-Net achieves strong performance on ORSSD, EORSSD, and ORSI-4199. On the more challenging EORSSD dataset, it reaches an S-measure of 0.9389, a max F-measure of 0.8972, and an MAE of 0.0066, outperforming the representative remote-sensing baseline DAF-Net by 0.0223 in S-measure and 0.0358 in max F-measure. What are the implications of the main finding? The results indicate that combining efficient global dependency modeling with explicit boundary enhancement and region-differentiated decoding is more effective than relying on convolution-only or uniform decoding strategies for complex remote sensing scenes with cluttered backgrounds, weak boundaries, low contrast, and densely distributed objects. The proposed hybrid Mamba design provides a favorable balance between detection accuracy and computational efficiency, showing practical potential for salient object detection in high-resolution optical remote sensing imagery. Salient object detection in optical remote sensing images remains challenging due to complex backgrounds, blurred boundaries, small objects, unstable foreground–background contrast, and dense object distributions. Existing convolution-based methods are effective at modeling local structures, but they are limited in capturing long-range dependencies, whereas Transformer-based approaches usually incur substantial computational cost when handling high-resolution remote sensing imagery. To address these issues, this paper proposes EGMamba-Net, an edge-guided global–local collaborative network for salient object detection in optical remote sensing images. Specifically, a hybrid global–local backbone is first constructed to preserve shallow texture, edge, and geometric details while introducing Mamba-based global modeling in deeper stages for efficient long-range dependency representation. An Edge Prior Enhancement Module (EPEM) is then designed to explicitly extract boundary priors from shallow features and refine feature representations through edge-guided modulation. To alleviate the representation conflict between global semantics and local details, a Global–Local Interaction Module (GLIM) is further developed, where convolutional local modeling and Mamba-based global modeling interact through cross-gating for complementary feature learning. Moreover, a Region-Adaptive Routing Decoder (RARD) is introduced to dynamically assign different refinement paths according to regional saliency response, boundary intensity, and contextual complexity, thereby improving the recovery of small, low-contrast, and densely distributed objects. In addition, a Difficulty-Aware Joint Loss (DAJL) is designed to enhance optimization on boundary regions and hard samples, improving robustness under challenging conditions. Extensiveexperiments on ORSSD, EORSSD, and ORSI-4199 datasets demonstrate the superiority of the proposed method. In particular, on the more challenging EORSSD dataset, EGMamba-Net achieves 0.9389 S-measure, 0.8972 max F-measure, and 0.0066 MAE. Compared with the representative remote-sensing method DAF-Net, it improves S-measure and max F-measure by 0.0223 and 0.0358, respectively, indicating stronger capability in background suppression, structural preservation, and boundary recovery. [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=194141093 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.3390/rs18101568 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 37 StartPage: 1568 Subjects: – SubjectFull: Optical remote sensing Type: general – SubjectFull: Object recognition (Computer vision) Type: general – SubjectFull: Image enhancement (Imaging systems) Type: general – SubjectFull: Optimization algorithms Type: general Titles: – TitleFull: EGMamba-Net: Edge-Guided Global–Local Mamba Network with Region-Adaptive Routing for Salient Object Detection in Optical Remote Sensing Images. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Zhang, Fubin – PersonEntity: Name: NameFull: Zhang, Zichi – PersonEntity: Name: NameFull: Zhang, Feihu IsPartOfRelationships: – BibEntity: Dates: – D: 15 M: 05 Text: May2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 20724292 Numbering: – Type: volume Value: 18 – Type: issue Value: 10 Titles: – TitleFull: Remote Sensing Type: main |
| ResultId | 1 |
