Data Extrapolation

It is used to synthesize low-frequency components of the seismic signals based on the recorded high-frequency components or vise versa.

Used ML Approaches:

  • Artificial Neural Networks

Used Neural Networks:

  • CNN

  • FC

  • Autoencoder

  • U-Net

  • DenseNet

Used Learning Procedures:

  • Supervised Learning

References:

  1. Jia, Z., & Lu, W. (2019). CNN-based ringing effect attenuation of vibroseis data for first-break picking. IEEE Geoscience and Remote Sensing Letters, 16(8), 1319-1323.

  2. Ovcharenko, O., Kazei, V., Kalita, M., Peter, D., & Alkhalifah, T. (2019). Deep learning for low-frequency extrapolation from multioffset seismic data. Geophysics, 84(6), R989-R1001.

  3. Sun, H., & Demanet, L. (2020). Extrapolated full-waveform inversion with deep learning. Geophysics, 85(3), R275-R288.

  4. Fang, J., Zhou, H., Elita Li, Y., Zhang, Q., Wang, L., Sun, P., & Zhang, J. (2020). Data-driven low-frequency signal recovery using deep-learning predictions in full-waveform inversion. Geophysics, 85(6), A37-A43.

  5. Li, Y., Song, J., Lu, W., Monkam, P., & Ao, Y. (2020). Multitask learning for super-resolution of seismic velocity model. IEEE Transactions on Geoscience and Remote Sensing.

Last updated