EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentation
Segmenting farmland images is challenging due to their high color similarity to the background and irregular shapes, resulting in over/undersegmentation. To tackle these challenges, we propose the Edge Guided Hybrid CNN-Mamba UNet (EGCM-UNet) and design the oriented residual convolutional edge branc...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Geocarto International |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/10106049.2024.2440407 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846114842252935168 |
|---|---|
| author | Jianhua Zheng Yusha Fu Xiaohan Chen Ruolin Zhao Junde Lu Huanghui Zhao Qian Chen |
| author_facet | Jianhua Zheng Yusha Fu Xiaohan Chen Ruolin Zhao Junde Lu Huanghui Zhao Qian Chen |
| author_sort | Jianhua Zheng |
| collection | DOAJ |
| description | Segmenting farmland images is challenging due to their high color similarity to the background and irregular shapes, resulting in over/undersegmentation. To tackle these challenges, we propose the Edge Guided Hybrid CNN-Mamba UNet (EGCM-UNet) and design the oriented residual convolutional edge branch (ORCEB) to mine prior edge information. Additionally, the model designs a MaUNet module, which introduces the Visual State Space (VSS) block fused with Mamba to manage long-distance dependencies of image features, and uses the Edge-Guided Semantic Aggregation Module (EGSAM) for precise segmentation by fusing edge features with the VSS block’s output. Lastly, comparative experiments were conducted using selected baseline models on the AgriculturalField-Seg dataset. The results show that EGCM-UNet outperformed U-Net with a Mean Intersection over Union (mIoU) of 0.394 vs. 0.379 on the test set. This indicates the proposed model delivers good performance in the semantic segmentation task of farmland remote sensing images. |
| format | Article |
| id | doaj-art-e009023acf134e95aad27f2b51c8a31d |
| institution | Kabale University |
| issn | 1010-6049 1752-0762 |
| language | English |
| publishDate | 2025-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Geocarto International |
| spelling | doaj-art-e009023acf134e95aad27f2b51c8a31d2024-12-20T07:33:47ZengTaylor & Francis GroupGeocarto International1010-60491752-07622025-12-0140110.1080/10106049.2024.2440407EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentationJianhua Zheng0Yusha Fu1Xiaohan Chen2Ruolin Zhao3Junde Lu4Huanghui Zhao5Qian Chen6College of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou, ChinaCollege of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou, ChinaCollege of Computer and Information Security, Guilin University of Electronic Technology, Guilin, ChinaCollege of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou, ChinaCollege of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou, ChinaCollege of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou, ChinaCollege of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou, ChinaSegmenting farmland images is challenging due to their high color similarity to the background and irregular shapes, resulting in over/undersegmentation. To tackle these challenges, we propose the Edge Guided Hybrid CNN-Mamba UNet (EGCM-UNet) and design the oriented residual convolutional edge branch (ORCEB) to mine prior edge information. Additionally, the model designs a MaUNet module, which introduces the Visual State Space (VSS) block fused with Mamba to manage long-distance dependencies of image features, and uses the Edge-Guided Semantic Aggregation Module (EGSAM) for precise segmentation by fusing edge features with the VSS block’s output. Lastly, comparative experiments were conducted using selected baseline models on the AgriculturalField-Seg dataset. The results show that EGCM-UNet outperformed U-Net with a Mean Intersection over Union (mIoU) of 0.394 vs. 0.379 on the test set. This indicates the proposed model delivers good performance in the semantic segmentation task of farmland remote sensing images.https://www.tandfonline.com/doi/10.1080/10106049.2024.2440407Remote sensing images of farmlandsemantic segmentationfeature fusionMamba |
| spellingShingle | Jianhua Zheng Yusha Fu Xiaohan Chen Ruolin Zhao Junde Lu Huanghui Zhao Qian Chen EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentation Geocarto International Remote sensing images of farmland semantic segmentation feature fusion Mamba |
| title | EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentation |
| title_full | EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentation |
| title_fullStr | EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentation |
| title_full_unstemmed | EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentation |
| title_short | EGCM-UNet: Edge Guided Hybrid CNN-Mamba UNet for farmland remote sensing image semantic segmentation |
| title_sort | egcm unet edge guided hybrid cnn mamba unet for farmland remote sensing image semantic segmentation |
| topic | Remote sensing images of farmland semantic segmentation feature fusion Mamba |
| url | https://www.tandfonline.com/doi/10.1080/10106049.2024.2440407 |
| work_keys_str_mv | AT jianhuazheng egcmunetedgeguidedhybridcnnmambaunetforfarmlandremotesensingimagesemanticsegmentation AT yushafu egcmunetedgeguidedhybridcnnmambaunetforfarmlandremotesensingimagesemanticsegmentation AT xiaohanchen egcmunetedgeguidedhybridcnnmambaunetforfarmlandremotesensingimagesemanticsegmentation AT ruolinzhao egcmunetedgeguidedhybridcnnmambaunetforfarmlandremotesensingimagesemanticsegmentation AT jundelu egcmunetedgeguidedhybridcnnmambaunetforfarmlandremotesensingimagesemanticsegmentation AT huanghuizhao egcmunetedgeguidedhybridcnnmambaunetforfarmlandremotesensingimagesemanticsegmentation AT qianchen egcmunetedgeguidedhybridcnnmambaunetforfarmlandremotesensingimagesemanticsegmentation |