README.md

SAM-Based Efficient Feature Integration Network for Remote Sensing Change Detection: A Case Study on Macao Sea Reclamation

published in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing publication information (JSTARS),

paper link: 10.1109/JSTARS.2025.3584145

The paddle implementation for EFI-SAM the weights on AI Studio the entire project can achieve AI Studio

Requirements

• Python 3.10
• paddle 3.0.0

@ARTICLE{11058393,
  author={Huang, Junqing and Bao, Junqi and Xia, Min and Yuan, Xiaochen},
  journal={IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing}, 
  title={SAM-Based Efficient Feature Integration Network for Remote Sensing Change Detection: A Case Study on Macao Sea Reclamation}, 
  year={2025},
  volume={},
  number={},
  pages={1-13},
  keywords={Feature extraction;Remote sensing;Land surface;Transformers;Semantics;Data mining;Accuracy;Visualization;Decoding;Computational modeling;Random fourier features;remote sensing change detection;sea reclamation;segment anything model},
  doi={10.1109/JSTARS.2025.3584145}}

Macao Land Change Detection (MLCD) Dataset

The MLCD utilizes imagery sourced from Google Earth Engine, covering a period of 15 years from 2008 to 2023. By precisely annotating these images, we have created a comprehensive dataset for sea-land change, focusing on the geographical evolution of reclaimed areas and the dynamics of surrounding vegetation coverage. This dataset includes 10,000 pairs of $256 \times 256$ images with a spatial resolution ranging from 0.5 to 2 m, providing valuable resources for studying coastal and land transformations. you can get it at https://www.modelscope.cn/datasets/chuntsing/MLCD