Seismic Noise Suppression Method Based on Unequilibrium Depth Desired Block Log-Likelihood Network

Abstract

The seismic noise suppression method based on the unbalanced depth expected block log-likelihood network belongs to the field of machine learning and seismic image processing technology. The regularization term parameter in the expected block log-likelihood algorithm only changes with the variance of the overall noise, so that the strong noise cannot be completely suppressed. problem, the present invention proposes a seismic noise suppression method based on an unbalanced depth expected block log likelihood network, in which the end-to-end denoising network consists of an expected block log likelihood denoising main network and an unbalanced multi-layer The perceptron parameter estimation network is composed of noise-containing seismic images as the input and clean seismic images as the output to learn network parameters, and for the first time adopts unbalanced block signal-to-noise ratios to adjust multi-layer perceptrons to estimate network parameters. The invention can realize the Each block in the seismic image estimates accurate regularization parameters, better controls the denoising intensity of each block, and then improves the block denoising effect, and is superior to traditional seismic denoising in suppressing strong noise in the desert and maintaining signal details algorithm.

Description

technical field [0001] The invention belongs to the technical field of machine learning and seismic image processing, and in particular relates to a seismic noise suppression method based on an unbalanced depth expected block log-likelihood network. Background technique [0002] The exploration of oil, natural gas and other resources has been an enduring hotspot. Seismic exploration is currently the main means of exploration and exploitation of underground energy such as oil and gas. However, affected by the environment of seismic exploration areas, the collected seismic images are often mixed with a large amount of random noise. These noises seriously damage the effective signals and increase the difficulty of extracting reflected seismic signals. Therefore, the suppression of seismic random noise is the fundamental problem in seismic data processing to improve seismic quality and extract underground structure information from noise-disturbed signals. In recent years, my c...

AI Technical Summary

Problems solved by technology

Solving the Desert Noise Suppression Problem with Unequalized Deep Desired Block Log-Likelihood Networks

Images