A Noise Removal Method for Seismic Data Based on Perceptron Residual Autoencoder Network

Abstract

The invention discloses a method for denoising seismic data of perceptron residual self-encoding network. The feature data reconstructed by self-encoding network encoding is output by the decoding part after the multi-layer perceptron residual convolution self-encoding block composed of multi-layer perceptron convolution layer, multi-scale layer, BN layer and auto-encoder , and combined with the optimized convolution kernel, a deep learning-based perceptron residual autoencoder network seismic data denoising method is proposed. The seismic data denoising method disclosed by the invention can simultaneously remove multiple waves and random noises under the condition that the local details of the seismic data are completely preserved and artifacts are not generated.

Description

Technical field [0001] The invention relates to the technical field of deep learning, in particular to a denoising method for seismic data of a perceptron residual self-encoding network in seismic data denoising. Background technique [0002] Seismic data noise can be divided into coherent noise and random noise. Traditional denoising needs to select different methods for denoising according to the difference of signal and noise characteristics. Traditional methods for removing multiples can be divided into two categories. One is the filtering method based on the difference between the significant wave and the multiple. The more commonly used methods are Radon transform and predictive deconvolution, and the other is based on Wave theory methods, among which the wave equation extrapolation method and feedback method are more commonly used. Radon transform denoising will also produce more artifacts; the predictive deconvolution method is more dependent on parameter settings; the wa...

AI Technical Summary

Problems solved by technology

Among them, based on Auto-Encoder (AE) network, Convolutional Neural Network (CNN), Generative Adversarial Networks (GAN) are widely used in the field of denoising, but the earthquake denoising method based on deep learning Less, the random noise removal algorithm based on the residual convolutional neural network has stronger denoising performance; based on the convolutional neural network denoising model, it can remove seismic random noise with unknown variance; the residual convolutional neural network image The denoising method applied to seismic data denoising can effectively remove random noise, but the above deep learning method can only remove random noise and cannot suppress multiple waves

Images