27th January 2022
Speaker: Eshan Naeini, Earth Analytics
Topic: Denoising with Machine Learning with Application to Seismic Data
Conditioning of seismic data is a key step in interpretative and quantitative exploration workflows. A key step in seismic data conditioning is the removal of various noise signatures. While various denoising operations are applied throughout the seismic processing workflow, typically we still observe a remnant of noise in post-migration seismic images.
We present a comparison of supervised and unsupervised ML based seismic denoising methods against traditional denoising operators. Our findings show that supervised deep learning-based denoising operators can provide good results but are challenged due to the need for good synthetic datasets and valid assumptions about the underlying noise characteristics that might be present in the seismic dataset they are being applied on. Unsupervised techniques however promise a good potential in seismic data conditioning workflows given their adaptive nature.
Furthermore, the application on a downstream task such as coloured inversion shows that the direct evaluation of these tasks on resulting denoised volumes not only provides a good benchmark for various denoising operators but can also directly highlight added value for structural and quantitative interpretation.