Seismic interference noise attenuation using DNN
AssigneeCgg Services SAS
InventorsSong Hou, Jing Sun
Life of this patent
Estimating the term…
Abstract
Seismic exploration methods and data processing apparatuses employ a deep neural network to remove seismic interference (SI) noise. Training data is generated by combining an SI model extracted using a conventional model from a subset of the seismic data, with SI free shots and simulated random noise. The trained DNN is used to process the entire seismic data thereby generating an image of subsurface formation for detecting presence and/or location of sought-after natural resources.
Claims20
A seismic exploration method, the method comprising:
- selecting a representative subset of a seismic data acquired during a survey;
- extracting a seismic interference, SI, model from the representative subset using a conventional method;
- combining the SI model with SI free shots and simulated random noise to generate a training dataset;
- training a deep neural network, DNN, using the training data so that the DNN to output the SI model and a sum of the SI free shots and the simulated random noise;
- and
- using the trained DNN to process the seismic data in order to generate an image of subsurface formation for detecting presence and/or location of sought-after natural resources.
A seismic data processing apparatus comprising:
- an interface configured to obtain seismic data acquired during a survey;
- and
- a data processing unit connected to the interface and configured
- to select a representative subset of a seismic data acquired during a survey,
- to extract a seismic interference, SI, model from the representative subset using a conventional method,
- to combine the SI model with SI free shots and simulated random noise to generate a training dataset,
- to train a deep neural network, DNN, using the training data so that the DNN to output the SI model and a sum of the SI free shots and the simulated random noise, and
- to use the trained DNN to process the seismic data in order to generate an image of subsurface formation for detecting presence and/or location of sought-after natural resources.
A computer readable recording medium non-transitorily storing executable codes that when executed by a computer make the computer perform a seismic exploration method, the method comprising:
- selecting a representative subset of a seismic data acquired during a survey;
- extracting a seismic interference, SI, model from the representative subset using a conventional method;
- combining the SI model with SI free shots and simulated random noise to generate a training dataset;
- training a deep neural network, DNN, using the training data, to output the SI model and a sum of the SI free shots and the simulated random noise;
- and
- using the trained DNN to process the seismic data in order to generate an image of subsurface formation for detecting presence and/or location of sought-after natural resources.
Dependencies are inferred from the claim text.