One or more first sensors may be configured to sense seismic signals and one or more second sensors may be configured to sense electromagnetic crosstalk signals. The second sensors are not responsive to the seismic signals. The data from the first and second sensors may be recorded as first data and second data, respectively. The first data may be modified based on the second data to remove the electromagnetic crosstalk noise form the seismic data.

Systems and methods for filtering interface waves from single component seismic data are disclosed. In one embodiment, a method of filtering seismic data includes comparing amplitude coefficients of a matrix storing the seismic data in a time-frequency domain against an amplitude threshold, and comparing frequencies of the matrix against a maximum expected frequency of noise. The method further includes, for each amplitude coefficient having less than the amplitude threshold and an associated frequency less than the maximum expected frequency of noise, scaling the amplitude coefficient to reduce its value. The method also includes performing an inverse time-frequency transformation on the matrix to generate a noise model in a time domain, and subtracting the noise model from the seismic data in the time domain to generate filtered seismic data.

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.

One or more first sensors may be configured to sense seismic signals and one or more second sensors may be configured to sense electromagnetic crosstalk signals. The second sensors are not responsive to the seismic signals. The data from the first and second sensors may be recorded as first data and second data, respectively. The first data may be modified based on the second data to remove the electromagnetic crosstalk noise form the seismic data.

A method of detecting an event by: obtaining a first sample data set; determining a frequency domain feature(s) of the first sample data set over a first time period; determining a first threshold for the a frequency domain feature(s) using the first sample data set; determining that the frequency domain feature(s) matches the first threshold; determining the presence of an event during the first time period based on determining that the frequency domain feature(s) matches the first threshold; obtaining a second sample data set; determining a frequency domain feature(s) of the second sample data set over a second time period; determining a second threshold for the frequency domain feature(s) using the second sample data set; determining that the frequency domain feature(s) matches the second threshold; and determining the presence of the event during the second time period based on determining that the frequency domain feature(s) matches the second threshold.

Methods, systems, and computer-readable medium to perform operations including: generating a first time-frequency spectrum of a first seismic trace from an original seismic dataset; generating a second time-frequency spectrum of a second seismic trace from an enhanced seismic dataset, where the second seismic trace; calculating a difference between the first time-frequency spectrum and the second time-frequency spectrum to generate a noise estimate in the first seismic trace; constructing, based on (i) the noise estimate, (ii) the first time-frequency spectrum, and (iii) the second time-frequency spectrum, a time-frequency mask (TFM); and using the constructed TFM to generate a third time-frequency spectrum of an output trace that corresponds to the first and second seismic traces.

This disclosure is directed to methods and systems to evaluate noise contend of seismic data received during a marine survey. The seismic data includes pressure and particle motion data generated by collocated pressure and particle motion sensors of a seismic data acquisition system. The pressure and particle motion data are cross ghosted and temporal and spatial wavelet transforms are applied to the cross-ghosted pressure and particle motion data in order to compute pressure energies and particle motion energies in temporal and spatial scales of a temporal and spatial scale domain. The pressure and particle motion energies may be compared to evaluate noise content in the pressure and particle motion data, evaluate changes in the noise content during the marine survey, and adjust marine survey parameters to reduce the noise content.

Systems and methods for determining object location may include a memory and a processor. The processor may be configured to collect seismic data and geophysical data to determine object location. The processor may be configured to determine one or more seismic attributes associated with a plurality types of noises based on the seismic data and the geophysical data using one or more machine learning algorithms. The processor may be configured to eliminate unwanted noises from noise classifications based on the one or more seismic attributes. The processor may be configured to predict the object location by comparing time and velocity data of the object with recorded timing and velocity data. The processor may be configured to validate the object location by comparing the determined noise with image data. The systems and methods may be used in, for example, detecting missing planes such as Malaysian Airlines Flight 370.

The invention notably relates to a computer-implemented method for processing a 4D seismic signal relative to a subsoil, the subsoil including a zone subject to extraction and/or injection, the method comprising: providing the 4D seismic signal; identifying a part of the 4D seismic signal corresponding to a zone of the subsoil distinct from the zone subject to extraction and/or injection; determining a noise model of the 4D seismic signal based on the identified part of the 4D seismic signal; and processing the 4D seismic signal based on the noise model. This improves the field of 4D seismic data processing.

Techniques are disclosed relating to machine learning in the context of noise filters for sensor data, e.g., as produced by geophysical surveys. In some embodiments, one or more filters are applied to sensor data, such a harsh filter determined to cause a threshold level of distortion in measured reflections, a mild filter determined to leave a threshold level of remaining noise signals, or an acceptable filter. In some embodiments, the system trains a machine learning classifier based on outputs of the filtering procedures and uses the classifier to determine whether other filtered sensor data from the same survey exhibits acceptable filtering. This may improve accuracy or performance in detecting unacceptable filtering, in some embodiments.