A separate three-dimensional (3D) integration filter mask if precomputed for each of x, y, and z dimensions with a given operator length. A portion of a 3D post-stack seismic data set is received for processing and loaded into a generated 3D-sub-cube. The separate 3D integration filter masks are applied to the loaded 3D-sub-cube to generate filtered 3D-sub-cube data. The square mean of the 3D-sub-cube is calculated to generate smoothed 3D-sub-cube 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.

Seismic survey data is received, indexed into index sets, and each index set partitioned into data blocks. For each particular data block of a particular index set, the particular data block is sliced into frequency slices. For each particular frequency slice of the particular data block, the particular frequency slice is processed to remove random and erratic noise by: forming a Hankel matrix from the particular frequency slice: determining an optimal rank for the Hankel matrix, determining a clean signal and erratic noise from the ranked Hankel matrix, and returning the clean signal and erratic noise for the particular frequency slice. A clean signal is assembled from the index sets.

Systems and methods include a computer implemented method for evaluating rock physical properties. Drilling acoustic signals are received in real time during a drilling operation through rock at a drilling location. Transformed data is generated in a frequency domain from the drilling acoustic signals. The transformed data includes frequency and amplitude information for the drilling acoustic signals. De-noised transformed data is generated from the transformed data by filtering noise including background noise generated in a recording system and top drive rotation generated traces. A lithological significant frequency range that includes de-noised significant data points is determined from the de-noised transformed data. Physical properties of the rock are determined in real time using drill bit rotation rates and the amplitudes of the de-noised significant data points.

Systems and methods of performing a seismic survey are described. The system can receive seismic data in a first domain, and transform the seismic data into a tau-p domain. The system can identify a value on an envelope in the tau-p domain, select several values on the tau-p envelope using a threshold, and then generate a masking function. The system can combine the masking function with the tau-p transformed seismic data, and then perform an inverse tau-p transform on the combined seismic data. The system can adjust amplitudes in the inverse tau-p transformed combined seismic data, and identify one or more coherent events corresponding to subsea lithologic formations or hydrocarbon deposits.

Computing device, computer instructions and method for removing cross-talk noise from seismic data and generating an image of a surveyed subsurface. The method includes receiving input seismic data D generated by firing one or more seismic sources so that source energy is overlapping, and the input seismic data D is recorded with seismic sensors over the subsurface; generating a cross-talk noise model N by replacing at least one original shot gather with a reconstructed shot gather; subtracting the cross-talk noise model N from the input seismic data D to attenuate coherent cross-talk noise to obtain processed seismic data D.sub.p; deblending the processed seismic data D.sub.p with a deblending algorithm to attenuate a residual noise to obtain deblended seismic data D.sub.d; and generating the image of the subsurface based on the deblended seismic data D.sub.d.

A data set representing features of a geologic formation is formed from two or more signal acquisition data set representing independent aspects of the same wavefield. A wavelet transform is performed on the two or more signal acquisition data sets, and the data sets are further transformed to equalize signal portions of the data sets. Remaining differences in the data sets are interpreted as excess noise and are removed by different methods to improve the signal-to-noise ratio of any resulting data set.

The present disclosure is related to methods, systems, and machine-readable media for interference attenuation of a residual portion of seismic data, such as may be recorded in a marine seismic survey. Recorded seismic data can be separated into a portion attributed to a source and a residual portion. Seismic interference attenuation can be performed on the residual portion.

Embodiments of non-uniformly spaced seismic receiver arrays and associated noise removal techniques are disclosed. In one embodiment of a method of seismic data acquisition, a plurality of seismic receivers may be positioned in an array having a plurality of regions, each region in the array having a respective average spacing between seismic receivers, with the average spacing in a second region of the plurality of regions being greater than the average spacing in a first region of the plurality of regions that is adjacent to the second region. Seismic data may be acquired utilizing the plurality of seismic receivers, and noise may be removed therefrom.

Systems and methods may use machine learning to automate the parameterization process for multi-stage iterative source separation. Seismic signals that are generated by a plurality of sources are received by a plurality of sensors within a field as a blended signal. An automated machine learning model that has been trained on blended and unblended signals determines if the incoming blended signal has a relatively high or low signal to noise ratio and then selects a threshold value based on the detected signal to noise ratio. The blended signal is then separated according to the source of the seismic data. A seismic image based on the separated seismic data is then generated which can then be used to adjust one or more control parameters in a machine or tool within the field.