A method for digitizing image-based data includes receiving an image file including one or more target objects, generating an intermediate image by removing noise from the image file using a denoising machine learning model, identifying the one or more target objects included in the intermediate image using an object segmentation machine learning model, discretizing the one or more target objects that were identified using the trained object segmentation machine learning model, and storing the one or more target objects that were discretized in a data file, visualizing the one or more target objects, or both.

Scattered body waves are isolated to primary, shear, and surface waves as a receiver wavefield from recorded near-surface scattered wave data generated by scatters. The isolated receiver wavefield is backward propagated through an earth model from a final to an initial state. A source wavefield and the receiver wavefields are cross-correlated. A source wavefield and the receiver wavefields are stacked, over all time steps and sources, to generate a subsurface image. A display of the subsurface image is initiated.

A method and a system for implementing the method are disclosed wherein the seismic input data and land acquisition input data may be obtained from a non-flat surface, sometimes mild or foothill topography as well as the shot and receiver lines might not necessarily be straight, and often curve to avoid obstacles on the land surface. In particular, the method and system disclosed, decomposes the cross-spread data into sparse common spread beams, then maps those sparse beams into common-spread depth domain, in order to finally stack them to construct the subsurface depth images. The common spread beam migration and processing have higher signal to noise ratio, as well as faster turn-around processing time, for the cross-spread land acquisition over the common-shot or common offset beam migration/processing. The common spread beam migration method and system disclosed, will eventually help illuminate and interpret the hydro-carbonate targets for the seismic processing.

Techniques are disclosed relating to deblending of sources in multi-source geophysical survey data, including marine or land-based data. Multiple sets of deblended receiver traces are generated by iteratively applying a coherency filter to estimated sets of deblended receiver traces and updating a residual until a termination condition is reached. In some embodiments, applying the coherency filter during a current iteration may include determining coefficients of the coherency filter based on estimated sets of deblended receiver traces from an immediately prior iteration. In further embodiments, applying the coherency filter may include applying a 3D projection filter, such as an fxy projection filter.

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 image processing including filtering a three-dimensional (3D) seismic image for random noise attenuation via multiple processors. The filtering includes receiving a 3D image cube of seismic image data, decomposing the 3D image cube into 3D sub-cubes for parallel computation on the multiple processors, designing and applying a two-dimensional (2D) adaptive filter for image points on 2D image slices of the 3D sub-cubes via the multiple processors to give filtered 3D sub-cubes, and summing the filtered 3D sub-cubes to give a filtered 3D image cube.

A method for enhancing properties of geophysical data with deep learning networks. Geophysical data may be acquired by positioning a source of sound waves at a chosen shot location, and measuring back-scattered energy generated by the source using receivers placed at selected locations. For example, seismic data may be collected using towed streamer acquisition in order to derive subsurface properties or to form images of the subsurface. However, towed streamer data may be deficient in one or more properties (e.g., at low frequencies). To compensate for the deficiencies, another survey (such as an Ocean Bottom Nodes (OBN) survey) may be sparsely acquired in order to train a neural network. The trained neural network may then be used to compensate for the towed streamer deficient properties, such as by using the trained neural network to extend the towed streamer data to the low frequencies.

The present disclosure discloses a method of obtaining a seismic while drilling signal. The method comprises the following steps: arranging geophones by using a first observation method to obtain a first seismic reference signal and a second seismic reference signal; arranging geophones by using a second observation method to obtain first seismic data; arranging geophones by using a third observation method to obtain second seismic data; comparing the first seismic reference signal with the second seismic reference signal to obtain a first output reference signal, and optimizing the first output signal to obtain a second output reference signal. The present disclosure obtains square matrix and near-wellhead seismic while drilling data through the combination of geophone square matrix combined observation, near-wellhead observation, and survey line observation, the data acquisition efficiency is relatively high, the signal-to-noise ratio is high, and thus, the problem of near-surface noise interference is effectively solved.

A method of determining a shear-wave attenuated vertical component vertical seismic profile (VSP) dataset is disclosed. The method includes, obtaining a multi-component VSP dataset, including a vertical and a horizontal component, transforming the vertical component into a vertical spectrum and the horizontal component into a horizontal spectrum, and designing a band-pass filter based, at least in part, on an energetic signal of the horizontal spectrum. The method further includes determining a muted vertical amplitude spectrum by applying the pass-band filter to an amplitude spectrum of the vertical spectrum, determining an estimated noise model based on the muted vertical amplitude spectrum and the vertical spectrum; and determining the shear-wave attenuated vertical component VSP dataset by adaptively subtracting the estimated noise model from the vertical component of the multi-component VSP dataset. A system including a seismic source, a plurality of seismic receivers, and a seismic processor for executing the method is disclosed.

Seismic data is deblended by performing, for each receiver, a first inversion and a second inversion in a transform domain. The first inversion is formulated to minimize a number of non-zero coefficients of the first inversion result. A sub-domain of the transform domain is defined by vectors of a transform domain basis for which the first inversion has yielded the non-zero coefficients. The second inversion is performed in this sub-domain. The solution of the second inversion is used to extract deblended seismic datasets corresponding to each of the distinct signals, from the seismic data.