CC-SGAE

Cross-Cycle Structured Graph Autoencoder for Unsupervised Cross-Sensor Image Change Detection.

📋 Environment Requirements

The code has been tested in the following environment. We recommend using these specific versions for reproducibility:

• python==3.12.2
• numpy==1.26.4
• torch==2.3.0
• torch_geometric==2.5.0
• scikit-learn==1.4.1
• scikit-image==0.22.0
• opencv-python==4.9.0
• imageio==2.34.0
• scipy==1.12.0

You can install the dependencies using pip:

pip install numpy scikit-learn scikit-image opencv-python imageio scipy torch_geometric
# Note: Please ensure PyTorch is installed according to your CUDA version.

📂 Project Structure

This repository currently contains the core implementation of the proposed method:

• Networks.py: Implementation of the Cross-Cycle Structured Graph Autoencoder.
• utils.py: Utility functions.
• data_loader.py: Data loading and preprocessing logic for cross-sensor datasets.

🔗 Related Datasets and Supporting Algorithms

The datasets used in this work are publicly available from the following sources:

• Dataset #2 & Dataset #4: Download from Professor Max Mignotte's webpage: http://www-labs.iro.umontreal.ca/~mignotte/
• Dataset #3: Download from Dr. Han's GitHub repository: https://github.com/rshante0426/MCD-datasets

The following repositories are related to the graph-based structural consistency and change alignment mechanisms. Great thanks to the authors for their excellent works:

• SCASC: https://github.com/yulisun/SCASC
• SRF: https://github.com/yulisun/SRF

📧 Contact

If you have any queries, please do not hesitate to contact us at: dearhyk@126.com