Bu et al. (2026) A Dual-Branch Collaborative Network Based on CNN and Transformer With Multiscale Dual Coordinate Attention for Farmland Segmentation

Identification

• Journal: IEEE Transactions on Geoscience and Remote Sensing
• Year: 2026
• Date: 2026-01-01
• Authors: Qiang Bu, Xuyu Xiang, Jiaohua Qin, Yuanjing Luo, Wenyan Pan, Yifan Tan
• DOI: 10.1109/tgrs.2026.3661067

Research Groups

[Information not available in the provided text.]

Short Summary

A novel dual-branch collaborative network, integrating CNN and Transformer architectures with multiscale dual coordinate attention, is proposed for improved farmland segmentation from imagery.

Objective

• To design and evaluate a deep learning model, specifically a dual-branch collaborative network combining CNN and Transformer with multiscale dual coordinate attention, for accurate farmland segmentation from imagery.

Study Configuration

• Spatial Scale: Farmland areas (e.g., from individual fields to agricultural regions, typically ranging from hectares to tens of square kilometers).
• Temporal Scale: Not explicitly stated, likely static image analysis, but could involve multi-temporal data if "multiscale" refers to temporal aspects.

Methodology and Data

• Models used: Dual-Branch Collaborative Network, Convolutional Neural Network (CNN), Transformer, Multiscale Dual Coordinate Attention.
• Data sources: Remote sensing imagery (e.g., satellite or aerial photographs) containing farmland, accompanied by ground truth segmentation masks for training and evaluation.

Main Results

[Information not available in the provided text.]

Contributions

[Information not available in the provided text.]

Funding

[Information not available in the provided text.]

Citation

@article{Bu2026DualBranch,
  author = {Bu, Qiang and Xiang, Xuyu and Qin, Jiaohua and Luo, Yuanjing and Pan, Wenyan and Tan, Yifan},
  title = {A Dual-Branch Collaborative Network Based on CNN and Transformer With Multiscale Dual Coordinate Attention for Farmland Segmentation},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2026},
  doi = {10.1109/tgrs.2026.3661067},
  url = {https://doi.org/10.1109/tgrs.2026.3661067}
}