AutofocusSAR

Codes and dataset for machine learning-based Synthetic Aperture Radar (SAR) autofocus algorithms. Please see AutofocusSAR Github or AutofocusSAR Webpage. Any commercial use is prohibited!

Folder Description

• Dataset: contains all information about dataset.
• ECELMs : contains code of Ensemble Convolutional Extreme Learning Machine based Autofocus Algorithms, models: Bagging-ECELMs.
• PAFnet : contains code of AFnet and PAFnet: Fast and Accurate SAR Autofocus Based on Deep Learning, models: AFnet, PAFnet.
• VDNNAF : contains code of A Finely Focusing Method of SAR Using Very Deep Neural Network, (implenmented by me, not the official).
• CNNAF : contains code of SAR Autofocus based on Convolutional Neural Networks.

Algorithms

1. Dataset or Bagging-ECELMs: Fast SAR Autofocus based on Ensemble Convolutional Extreme Learning Machine, 2021, pdf, doi
2. CNN-AF: SAR Autofocus based on Convolutional Neural Networks
3. AFnet and PAFnet: Fast and Accurate SAR Autofocus Based on Deep Learning, 2022, pdf, doi

Usage

Dependencies

You need first to install our SAR library ( torchsar ) by excuting the following command:

Please see torchsar for details (e.g. install on Windows). The package can be installed by

pip install torchsar

Now, all platforms are supported and part of the source code is open!

Citation

If you find the datasets or codes are useful, please kindly cite our papers and star our pakcage AutofocusSAR, torchsar, torchbox on GitHub:

@article{Liu2021Fast,
  title={Fast SAR Autofocus Based on Ensemble Convolutional Extreme Learning Machine},
  author={Liu, Zhi and Yang, Shuyuan and Feng, Zhixi and Gao, Quanwei and Wang, Min},
  journal={Remote Sensing},
  volume={13},
  number={14},
  pages={2683},
  year={2021},
  publisher={Multidisciplinary Digital Publishing Institute},
  doi={https://doi.org/10.3390/rs13142683}
}

@article{Liu2022PAFnet,
  title={AFnet and PAFnet: Fast and Accurate SAR Autofocus Based on Deep Learning},
  author={Liu, Zhi and Yang, Shuyuan and Gao, Quanwei and Feng, Zhixi and Wang, Min and Jiao, Licheng},
  journal={IEEE Transactions on Geoscience and Remote Sensing},
  year={2022},
  volume={60},
  number={},
  pages={1-13},
  doi={10.1109/TGRS.2022.3217063}
}