View in EDS HTML Full Text PDF Full Text

Rid-HRNet: A Lightweight Multi-Scale Network for Sand Ridge Line Extraction from Landsat Imagery.

Saved in:
Bibliographic Details
Title: Rid-HRNet: A Lightweight Multi-Scale Network for Sand Ridge Line Extraction from Landsat Imagery.
Authors: Huang, Xuanjing1 (AUTHOR), Liu, Xinchao1,2 (AUTHOR), Mu, Jiayue1,3 (AUTHOR), Zhu, Ye1,2 (AUTHOR), Wang, Zhaobin1,2 (AUTHOR) wangzhb@lzu.edu.cn, Zhang, Yaonan3 (AUTHOR)
Source: Remote Sensing. May2026, Vol. 18 Issue 9, p1439. 22p.
Subjects: Landsat satellites, Geomorphological mapping, Remote sensing, Feature extraction, Artificial neural networks
Abstract: Highlights: This study proposes Rid-HRNet, a lightweight high-resolution network for sand ridge line extraction from Landsat imagery. The model maintains high-resolution representations throughout the network and incorporates a Multi-Scale Information Aggregation (MSIA) module together with an Improved Contextual Fusion Module (ICFM) to enhance multiscale feature interaction and contextual fusion, enabling accurate detection of thin ridge structures under complex desert textures. What are the main findings? Maintaining high-resolution representations throughout the network significantly improves the extraction of thin ridge structures under complex desert textures and low-contrast conditions. The proposed MSIA module and the ICFM module enhance multi-scale feature aggregation and contextual fusion, leading to higher F1-score and reduced structural fragmentation compared with baseline models. What are the implications of the main findings? Lightweight high-resolution architectures provide an effective solution for extracting fine linear geomorphological structures from medium-resolution satellite imagery. The proposed framework offers a practical approach for large-scale desert geomorphology mapping and can potentially be extended to other remote sensing tasks involving elongated structures. Sand ridge lines serve as key geomorphological indicators for interpreting aeolian dynamics and assessing desertification intensity. However, automated extraction of continuous ridge structures from remote sensing imagery remains challenging due to the multi-scale morphology of dunes, complex surface textures, and strong shadow interference. Conventional edge detection models often rely on computationally heavy backbones or suffer from structural discontinuities in subtle ridge branches, limiting their applicability in large-scale desert monitoring. To address these challenges, we propose Rid-HRNet, a lightweight high-resolution network specifically designed for efficient and structurally coherent sand ridge extraction. Unlike traditional encoder–decoder architectures, Rid-HRNet maintains parallel high-resolution representations throughout the network to preserve fine spatial details. A Multi-Scale Information Aggregation (MSIA) module enhances cross-scale feature interaction by integrating shallow structural cues with deeper semantic representations. In addition, an Improved Contextual Fusion Module (ICFM) employs pixel-wise attention to adaptively fuse multi-level predictions, reinforcing ridge continuity while suppressing background interference. Experiments on Landsat-8 desert imagery demonstrate that Rid-HRNet achieves an Optimal Dataset Scale (ODS) of 0.790, an Optimal Image Scale (OIS) of 0.806, an Average Precision (AP) of 0.710, and an AC(R50) score of 0.744. The proposed model outperforms classical VGG-based detectors, including HED and RCF, as well as recent lightweight baselines such as PiDiNet and LDC, in terms of overall accuracy and structural consistency. Notably, Rid-HRNet contains only 0.20M parameters and requires 0.55 GFLOPs, operating at 279.23 FPS with a GPU memory footprint of 0.02 GB. These results indicate that Rid-HRNet achieves a favorable balance between detection performance and computational efficiency, supporting large-scale geomorphological mapping and operational desert monitoring based on high-resolution satellite imagery. [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.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 193715470
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Rid-HRNet: A Lightweight Multi-Scale Network for Sand Ridge Line Extraction from Landsat Imagery.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Huang%2C+Xuanjing%22">Huang, Xuanjing</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Liu%2C+Xinchao%22">Liu, Xinchao</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Mu%2C+Jiayue%22">Mu, Jiayue</searchLink><relatesTo>1,3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhu%2C+Ye%22">Zhu, Ye</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Zhaobin%22">Wang, Zhaobin</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> wangzhb@lzu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Yaonan%22">Zhang, Yaonan</searchLink><relatesTo>3</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Remote+Sensing%22">Remote Sensing</searchLink>. May2026, Vol. 18 Issue 9, p1439. 22p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Landsat+satellites%22">Landsat satellites</searchLink><br /><searchLink fieldCode="DE" term="%22Geomorphological+mapping%22">Geomorphological mapping</searchLink><br /><searchLink fieldCode="DE" term="%22Remote+sensing%22">Remote sensing</searchLink><br /><searchLink fieldCode="DE" term="%22Feature+extraction%22">Feature extraction</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+neural+networks%22">Artificial neural networks</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Highlights: This study proposes Rid-HRNet, a lightweight high-resolution network for sand ridge line extraction from Landsat imagery. The model maintains high-resolution representations throughout the network and incorporates a Multi-Scale Information Aggregation (MSIA) module together with an Improved Contextual Fusion Module (ICFM) to enhance multiscale feature interaction and contextual fusion, enabling accurate detection of thin ridge structures under complex desert textures. What are the main findings? Maintaining high-resolution representations throughout the network significantly improves the extraction of thin ridge structures under complex desert textures and low-contrast conditions. The proposed MSIA module and the ICFM module enhance multi-scale feature aggregation and contextual fusion, leading to higher F1-score and reduced structural fragmentation compared with baseline models. What are the implications of the main findings? Lightweight high-resolution architectures provide an effective solution for extracting fine linear geomorphological structures from medium-resolution satellite imagery. The proposed framework offers a practical approach for large-scale desert geomorphology mapping and can potentially be extended to other remote sensing tasks involving elongated structures. Sand ridge lines serve as key geomorphological indicators for interpreting aeolian dynamics and assessing desertification intensity. However, automated extraction of continuous ridge structures from remote sensing imagery remains challenging due to the multi-scale morphology of dunes, complex surface textures, and strong shadow interference. Conventional edge detection models often rely on computationally heavy backbones or suffer from structural discontinuities in subtle ridge branches, limiting their applicability in large-scale desert monitoring. To address these challenges, we propose Rid-HRNet, a lightweight high-resolution network specifically designed for efficient and structurally coherent sand ridge extraction. Unlike traditional encoder–decoder architectures, Rid-HRNet maintains parallel high-resolution representations throughout the network to preserve fine spatial details. A Multi-Scale Information Aggregation (MSIA) module enhances cross-scale feature interaction by integrating shallow structural cues with deeper semantic representations. In addition, an Improved Contextual Fusion Module (ICFM) employs pixel-wise attention to adaptively fuse multi-level predictions, reinforcing ridge continuity while suppressing background interference. Experiments on Landsat-8 desert imagery demonstrate that Rid-HRNet achieves an Optimal Dataset Scale (ODS) of 0.790, an Optimal Image Scale (OIS) of 0.806, an Average Precision (AP) of 0.710, and an AC(R50) score of 0.744. The proposed model outperforms classical VGG-based detectors, including HED and RCF, as well as recent lightweight baselines such as PiDiNet and LDC, in terms of overall accuracy and structural consistency. Notably, Rid-HRNet contains only 0.20M parameters and requires 0.55 GFLOPs, operating at 279.23 FPS with a GPU memory footprint of 0.02 GB. These results indicate that Rid-HRNet achieves a favorable balance between detection performance and computational efficiency, supporting large-scale geomorphological mapping and operational desert monitoring based on high-resolution satellite imagery. [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=193715470
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.3390/rs18091439
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 22
        StartPage: 1439
    Subjects:
      – SubjectFull: Landsat satellites
        Type: general
      – SubjectFull: Geomorphological mapping
        Type: general
      – SubjectFull: Remote sensing
        Type: general
      – SubjectFull: Feature extraction
        Type: general
      – SubjectFull: Artificial neural networks
        Type: general
    Titles:
      – TitleFull: Rid-HRNet: A Lightweight Multi-Scale Network for Sand Ridge Line Extraction from Landsat Imagery.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Huang, Xuanjing
      – PersonEntity:
          Name:
            NameFull: Liu, Xinchao
      – PersonEntity:
          Name:
            NameFull: Mu, Jiayue
      – PersonEntity:
          Name:
            NameFull: Zhu, Ye
      – PersonEntity:
          Name:
            NameFull: Wang, Zhaobin
      – PersonEntity:
          Name:
            NameFull: Zhang, Yaonan
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Text: May2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 20724292
          Numbering:
            – Type: volume
              Value: 18
            – Type: issue
              Value: 9
          Titles:
            – TitleFull: Remote Sensing
              Type: main
ResultId 1