U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas
Arable land is fundamental to agricultural production and a crucial component of ecosystems. However, its complex texture and distribution in remote sensing images make it susceptible to interference from other land cover types, such as water bodies, roads, and buildings, complicating accurate ident...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/5/760 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850030809971425280 |
|---|---|
| author | Yun Chen Yiheng Xie Weiyuan Yao Yu Zhang Xinhong Wang Yanli Yang Lingli Tang |
| author_facet | Yun Chen Yiheng Xie Weiyuan Yao Yu Zhang Xinhong Wang Yanli Yang Lingli Tang |
| author_sort | Yun Chen |
| collection | DOAJ |
| description | Arable land is fundamental to agricultural production and a crucial component of ecosystems. However, its complex texture and distribution in remote sensing images make it susceptible to interference from other land cover types, such as water bodies, roads, and buildings, complicating accurate identification. Building on previous research, this study proposes an efficient and lightweight CNN-based network, U-MGA, to address the challenges of feature similarity between arable and non-arable areas, insufficient fine-grained feature extraction, and the underutilization of multi-scale information. Specifically, a Multi-Scale Adaptive Segmentation (MSAS) is designed during the feature extraction phase to provide multi-scale and multi-feature information, supporting the model’s feature reconstruction stage. In the reconstruction phase, the introduction of the Multi-Scale Contextual Module (MCM) and Group Aggregation Bridge (GAB) significantly enhances the efficiency and accuracy of multi-scale and fine-grained feature utilization. The experiments conducted on an arable land dataset based on GF-2 imagery and a publicly available dataset show that U-MGA outperforms mainstream networks (Unet, A2FPN, Segformer, FTUnetformer, DCSwin, and TransUnet) across six evaluation metrics (Overall Accuracy (OA), Precision, Recall, F1-score, Intersection-over-Union (IoU), and Kappa coefficient). Thus, this study provides an efficient and precise solution for the arable land recognition task, which is of significant importance for agricultural resource monitoring and ecological environmental protection. |
| format | Article |
| id | doaj-art-922b991054364aedb4e40557a04e58a8 |
| institution | DOAJ |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-922b991054364aedb4e40557a04e58a82025-08-20T02:59:07ZengMDPI AGRemote Sensing2072-42922025-02-0117576010.3390/rs17050760U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated AreasYun Chen0Yiheng Xie1Weiyuan Yao2Yu Zhang3Xinhong Wang4Yanli Yang5Lingli Tang6University of Chinese Academy of Sciences, Beijing 100112, ChinaSchool of Earth Sciences and Engineering, Hohai University, Nanjing 211100, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100112, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100112, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100112, ChinaShanghai Aerospace Control Technology Institute, Shanghai 201109, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100112, ChinaArable land is fundamental to agricultural production and a crucial component of ecosystems. However, its complex texture and distribution in remote sensing images make it susceptible to interference from other land cover types, such as water bodies, roads, and buildings, complicating accurate identification. Building on previous research, this study proposes an efficient and lightweight CNN-based network, U-MGA, to address the challenges of feature similarity between arable and non-arable areas, insufficient fine-grained feature extraction, and the underutilization of multi-scale information. Specifically, a Multi-Scale Adaptive Segmentation (MSAS) is designed during the feature extraction phase to provide multi-scale and multi-feature information, supporting the model’s feature reconstruction stage. In the reconstruction phase, the introduction of the Multi-Scale Contextual Module (MCM) and Group Aggregation Bridge (GAB) significantly enhances the efficiency and accuracy of multi-scale and fine-grained feature utilization. The experiments conducted on an arable land dataset based on GF-2 imagery and a publicly available dataset show that U-MGA outperforms mainstream networks (Unet, A2FPN, Segformer, FTUnetformer, DCSwin, and TransUnet) across six evaluation metrics (Overall Accuracy (OA), Precision, Recall, F1-score, Intersection-over-Union (IoU), and Kappa coefficient). Thus, this study provides an efficient and precise solution for the arable land recognition task, which is of significant importance for agricultural resource monitoring and ecological environmental protection.https://www.mdpi.com/2072-4292/17/5/760arable land recognitionhigh-resolution remote sensing imagerylightweight networkattention mechanismmulti-scale information |
| spellingShingle | Yun Chen Yiheng Xie Weiyuan Yao Yu Zhang Xinhong Wang Yanli Yang Lingli Tang U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas Remote Sensing arable land recognition high-resolution remote sensing imagery lightweight network attention mechanism multi-scale information |
| title | U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas |
| title_full | U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas |
| title_fullStr | U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas |
| title_full_unstemmed | U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas |
| title_short | U-MGA: A Multi-Module Unet Optimized with Multi-Scale Global Attention Mechanisms for Fine-Grained Segmentation of Cultivated Areas |
| title_sort | u mga a multi module unet optimized with multi scale global attention mechanisms for fine grained segmentation of cultivated areas |
| topic | arable land recognition high-resolution remote sensing imagery lightweight network attention mechanism multi-scale information |
| url | https://www.mdpi.com/2072-4292/17/5/760 |
| work_keys_str_mv | AT yunchen umgaamultimoduleunetoptimizedwithmultiscaleglobalattentionmechanismsforfinegrainedsegmentationofcultivatedareas AT yihengxie umgaamultimoduleunetoptimizedwithmultiscaleglobalattentionmechanismsforfinegrainedsegmentationofcultivatedareas AT weiyuanyao umgaamultimoduleunetoptimizedwithmultiscaleglobalattentionmechanismsforfinegrainedsegmentationofcultivatedareas AT yuzhang umgaamultimoduleunetoptimizedwithmultiscaleglobalattentionmechanismsforfinegrainedsegmentationofcultivatedareas AT xinhongwang umgaamultimoduleunetoptimizedwithmultiscaleglobalattentionmechanismsforfinegrainedsegmentationofcultivatedareas AT yanliyang umgaamultimoduleunetoptimizedwithmultiscaleglobalattentionmechanismsforfinegrainedsegmentationofcultivatedareas AT linglitang umgaamultimoduleunetoptimizedwithmultiscaleglobalattentionmechanismsforfinegrainedsegmentationofcultivatedareas |