Advancing Rural Building Extraction via Diverse Dataset Construction and Model Innovation with Attention and Context Learning

Mingyang Yu, Fangliang Zhou, Haiqing Xu, Shuai Xu · 2023 · Applied Sciences

Summary. Researchers developed AGSC-Net, a deep learning model for automatically extracting rural buildings from satellite imagery. They created a dataset of rural buildings across nine Chinese regions to address the scarcity of training data. The model uses attention mechanisms and context learning to identify buildings despite regional variations in construction styles. AGSC-Net outperformed existing methods and enables better rural planning and disaster assessment.

Read the original

@article{yu-2023-advancing-rural-building-extraction-via,
  title   = {Advancing Rural Building Extraction via Diverse Dataset Construction and Model Innovation with Attention and Context Learning},
  author  = {Mingyang Yu and Fangliang Zhou and Haiqing Xu and Shuai Xu},
  journal = {Applied Sciences},
  year    = {2023},
  doi     = {10.3390/app132413149},
  url     = {https://doi.org/10.3390/app132413149}
}

TY  - JOUR
TI  - Advancing Rural Building Extraction via Diverse Dataset Construction and Model Innovation with Attention and Context Learning
AU  - Mingyang Yu
AU  - Fangliang Zhou
AU  - Haiqing Xu
AU  - Shuai Xu
JO  - Applied Sciences
PY  - 2023
DO  - 10.3390/app132413149
UR  - https://doi.org/10.3390/app132413149
ER  - 

Details

DOI

10.3390/app132413149

Countries

China

Regions

Asia

Categories

agtech, rural-data-and-definitions

Added

2026-04-28