Machine Learning for Seismic Data ProcessingMS24

In this talk, we would like to present how machine learning (including SVR and deep learning) can work for seismic data denoising and inversion from training data sets. We introduce deep learning method to seismic noise attenuation without knowing the noise variance (blind denoising). The training set is obtained by partitioning the data from SEG open data and some denoised field data into small patches. Numerical results are provided with comparisons to traditional methods and state-of-the-art methods, showing deep learning method achieve good performance.

This presentation is part of Minisymposium “MS24 - Data-driven approaches in imaging science (3 parts)
organized by: Jae Kyu Choi (Institute of Natural Sciences, Shanghai Jiao Tong University) , Chenglong Bao (Yau Mathematical Sciences Center, Tsinghua University) .

Jianwei Ma (Department of Mathematics, Center of Geophysics, Harbin Institute of Technology)