FCDNet is a supervised-learning-based deep learning framework for denoising the FORGE DAS dataset with different types of noise. It has an U-shaped structure, containing several downsampling convolutional blocks (DCB) and upsampling convolutional blocks (UCB). FCDNet has been demonstrated on invisible, weakly visible, and visible signals denoising immediately and efficiently.
If you find this package useful, please do not forget to cite the following paper.
Yang, L., Fomel, S., Wang, S., Chen, X., Saad, O.M., and Chen, Y., (2023). Denoising of distributed acoustic sensing data using supervised deep learning, Geophysics, 88(1), doi: 10.1190/geo2022-0138.1.
BibTeX:
@article{DASD2023,
title={Denoising of distributed acoustic sensing data using supervised deep learning},
author={Liuqing Yang and Sergey Fomel and Shoudong Wang and Xiaohong Chen and Wei Chen and Omar M. Saad and Yangkang Chen},
journal={Geophysics},
year={2023},
volume={88},
issue={1},
pages={in press},
doi={10.1190/geo2022-0138.1},
}
GNU General Public License, Version 3
(http://www.gnu.org/copyleft/gpl.html)
- Tensforflow-gpu 1.9.0
- Keras 2.2.5
The best trained model can be downloaded directly. Test_data only display three real DAS data, which are invisible, weakly visible, and visible signals. All FORGE DAS data can be downloaded here.
If you have any suggestions and questions, please contact me:
Liuqing Yang
yangliuqingqin@163.com