View in EDS HTML Full Text PDF Full Text

Intertidal Seagrass Mapping Using UAV Visible and Multispectral Imagery: A Comparative Semantic Segmentation Study with Explainability Analysis.

Saved in:
Bibliographic Details
Title: Intertidal Seagrass Mapping Using UAV Visible and Multispectral Imagery: A Comparative Semantic Segmentation Study with Explainability Analysis.
Authors: Lian, Jiali1 (AUTHOR), Mo, Zhanyou1 (AUTHOR), Liu, Zhimin1 (AUTHOR), Peng, Bo1 (AUTHOR), Chang, Ming1 (AUTHOR), Wang, Xuemei1 (AUTHOR), Wang, Weiwen1 (AUTHOR) wwangeci@jnu.edu.cn
Source: Remote Sensing. Jun2026, Vol. 18 Issue 12, p2057. 26p.
Subjects: Multispectral imaging, Image segmentation, Wetlands monitoring, Drone photography, Deep learning, Remote sensing
Abstract: Highlights: A UAV semantic segmentation framework was developed for accurate and explainable monitoring of intertidal seagrass patches by integrating visible bands, multispectral bands, and vegetation indices. The proposed framework improved the accuracy of seagrass pixel extraction in complex intertidal environments, demonstrating better cross-scenario applicability and providing a technical reference for intertidal seagrass mapping. The inclusion of explainability analysis provided a reference for feature input selection, enhancing the transparency of deep learning models in marine ecological monitoring. What are the main findings? UPerNet-ConvNeXtV2-Tiny achieved the best overall performance among the tested models, with 97.45% ACC, 94.63% mIoU, and 97.23% F1 score. Visual comparisons revealed that UPerNet-ConvNeXtV2-Tiny better preserved fine spatial details and structural integrity, confirming its cross-scenario applicability in independent test areas. Explainability analysis showed that model discrimination mainly relied on red and green spectral features, while ESG and SSG exhibited different feature dependence patterns, supporting the value of distinguishing exposed and shallow-submerged seagrass in intertidal mapping. What are the implications of the main findings? The validated high-performance segmentation framework provides a reliable technical solution for fine-grained UAV-based mapping of intertidal seagrass, supporting blue carbon accounting, habitat conservation, and dynamic monitoring of fragmented seagrass beds. The revealed dependence on key spectral features provides a basis for subsequent feature screening, input optimization, and lightweight model design and offers an explainable methodological reference for similar coastal wetland remote sensing tasks. Seagrass meadows are important blue carbon habitats, but their patchy distribution in intertidal zones makes accurate UAV mapping difficult under shallow water cover and complex sediment backgrounds. This study developed a fine-grained semantic segmentation framework with explainability analysis to improve intertidal seagrass extraction from high-resolution UAV visible and multispectral imagery. Exposed seagrass (ESG) and shallow-submerged seagrass (SSG) were mapped separately to represent two observable intertidal states. Visible bands, multispectral bands, and vegetation indices were used as model inputs. U-Net and DeepLabV3+ served as baseline models, while UPerNet-ConvNeXtV2-Tiny was tested under the same settings. Kernel SHAP and permutation importance were used to assess feature contributions. UPerNet-ConvNeXtV2-Tiny achieved the best performance, with an overall accuracy (ACC), mean Intersection over Union (mIoU), and F1 score of 97.45%, 94.63%, and 97.23%, respectively. It outperformed the baseline models in suppressing background interference, preserving patch morphology, and reducing omission errors in weak response and boundary areas, while demonstrating better cross-scenario applicability in independent test areas. Explainability analysis showed that model discrimination was mainly associated with red and green-related features, especially RGB-R, MS-R, MS-G, RGB-G, and NGRDI. ESG and SSG showed different feature dependence patterns, indicating that high-resolution UAV imagery can support accurate seagrass mapping and reveal spectral differences between intertidal seagrass states. These findings provide a practical framework for UAV-based intertidal seagrass mapping and monitoring and offer guidance for feature selection and model explainability analysis. [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: 194915190
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Intertidal Seagrass Mapping Using UAV Visible and Multispectral Imagery: A Comparative Semantic Segmentation Study with Explainability Analysis.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Lian%2C+Jiali%22">Lian, Jiali</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Mo%2C+Zhanyou%22">Mo, Zhanyou</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Liu%2C+Zhimin%22">Liu, Zhimin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Peng%2C+Bo%22">Peng, Bo</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Chang%2C+Ming%22">Chang, Ming</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Xuemei%22">Wang, Xuemei</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Weiwen%22">Wang, Weiwen</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> wwangeci@jnu.edu.cn</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Remote+Sensing%22">Remote Sensing</searchLink>. Jun2026, Vol. 18 Issue 12, p2057. 26p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Multispectral+imaging%22">Multispectral imaging</searchLink><br /><searchLink fieldCode="DE" term="%22Image+segmentation%22">Image segmentation</searchLink><br /><searchLink fieldCode="DE" term="%22Wetlands+monitoring%22">Wetlands monitoring</searchLink><br /><searchLink fieldCode="DE" term="%22Drone+photography%22">Drone photography</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+learning%22">Deep learning</searchLink><br /><searchLink fieldCode="DE" term="%22Remote+sensing%22">Remote sensing</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Highlights: A UAV semantic segmentation framework was developed for accurate and explainable monitoring of intertidal seagrass patches by integrating visible bands, multispectral bands, and vegetation indices. The proposed framework improved the accuracy of seagrass pixel extraction in complex intertidal environments, demonstrating better cross-scenario applicability and providing a technical reference for intertidal seagrass mapping. The inclusion of explainability analysis provided a reference for feature input selection, enhancing the transparency of deep learning models in marine ecological monitoring. What are the main findings? UPerNet-ConvNeXtV2-Tiny achieved the best overall performance among the tested models, with 97.45% ACC, 94.63% mIoU, and 97.23% F1 score. Visual comparisons revealed that UPerNet-ConvNeXtV2-Tiny better preserved fine spatial details and structural integrity, confirming its cross-scenario applicability in independent test areas. Explainability analysis showed that model discrimination mainly relied on red and green spectral features, while ESG and SSG exhibited different feature dependence patterns, supporting the value of distinguishing exposed and shallow-submerged seagrass in intertidal mapping. What are the implications of the main findings? The validated high-performance segmentation framework provides a reliable technical solution for fine-grained UAV-based mapping of intertidal seagrass, supporting blue carbon accounting, habitat conservation, and dynamic monitoring of fragmented seagrass beds. The revealed dependence on key spectral features provides a basis for subsequent feature screening, input optimization, and lightweight model design and offers an explainable methodological reference for similar coastal wetland remote sensing tasks. Seagrass meadows are important blue carbon habitats, but their patchy distribution in intertidal zones makes accurate UAV mapping difficult under shallow water cover and complex sediment backgrounds. This study developed a fine-grained semantic segmentation framework with explainability analysis to improve intertidal seagrass extraction from high-resolution UAV visible and multispectral imagery. Exposed seagrass (ESG) and shallow-submerged seagrass (SSG) were mapped separately to represent two observable intertidal states. Visible bands, multispectral bands, and vegetation indices were used as model inputs. U-Net and DeepLabV3+ served as baseline models, while UPerNet-ConvNeXtV2-Tiny was tested under the same settings. Kernel SHAP and permutation importance were used to assess feature contributions. UPerNet-ConvNeXtV2-Tiny achieved the best performance, with an overall accuracy (ACC), mean Intersection over Union (mIoU), and F1 score of 97.45%, 94.63%, and 97.23%, respectively. It outperformed the baseline models in suppressing background interference, preserving patch morphology, and reducing omission errors in weak response and boundary areas, while demonstrating better cross-scenario applicability in independent test areas. Explainability analysis showed that model discrimination was mainly associated with red and green-related features, especially RGB-R, MS-R, MS-G, RGB-G, and NGRDI. ESG and SSG showed different feature dependence patterns, indicating that high-resolution UAV imagery can support accurate seagrass mapping and reveal spectral differences between intertidal seagrass states. These findings provide a practical framework for UAV-based intertidal seagrass mapping and monitoring and offer guidance for feature selection and model explainability analysis. [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=194915190
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.3390/rs18122057
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 26
        StartPage: 2057
    Subjects:
      – SubjectFull: Multispectral imaging
        Type: general
      – SubjectFull: Image segmentation
        Type: general
      – SubjectFull: Wetlands monitoring
        Type: general
      – SubjectFull: Drone photography
        Type: general
      – SubjectFull: Deep learning
        Type: general
      – SubjectFull: Remote sensing
        Type: general
    Titles:
      – TitleFull: Intertidal Seagrass Mapping Using UAV Visible and Multispectral Imagery: A Comparative Semantic Segmentation Study with Explainability Analysis.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Lian, Jiali
      – PersonEntity:
          Name:
            NameFull: Mo, Zhanyou
      – PersonEntity:
          Name:
            NameFull: Liu, Zhimin
      – PersonEntity:
          Name:
            NameFull: Peng, Bo
      – PersonEntity:
          Name:
            NameFull: Chang, Ming
      – PersonEntity:
          Name:
            NameFull: Wang, Xuemei
      – PersonEntity:
          Name:
            NameFull: Wang, Weiwen
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 15
              M: 06
              Text: Jun2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 20724292
          Numbering:
            – Type: volume
              Value: 18
            – Type: issue
              Value: 12
          Titles:
            – TitleFull: Remote Sensing
              Type: main
ResultId 1