A new approach for detecting topographic and geologic information in seismic data

Geophysics ◽  
2012 ◽  
Vol 77 (2) ◽  
pp. V81-V90 ◽  
Author(s):  
Adam Gersztenkorn
Keyword(s):  
Seismic Data ◽  
Computational Cost ◽  
Wavelet Domain ◽  
3D Seismic ◽  
Limited Extent ◽  
Complex Matrix ◽  
New Approach ◽  
Additional Information ◽  
Seismic Volumes ◽  
Topographic Aspect

Interpreting 3D seismic volumes can be an intensive and time-consuming endeavor. Algorithms that provide additional information and expedite this process can therefore be useful tools for the interpreter. To further this goal, an algorithm that gives a topographic perspective of seismic data is described. After applying the continuous wavelet transform to the data, templates having a directional orientation are constructed locally in the complex wavelet domain for a number of scales. For each scale, a complex matrix is formed having real and imaginary parts, which are independently designed for a specific purpose and then combined to produce the final result. Whereas the composition of the real matrix is not well suited for dealing with the topographic aspect of the data, the imaginary matrix is. Using basic concepts from graph theory, the imaginary matrix is constructed to reveal the topographic nature of the underlying data. To a limited extent, dip scans provide similar results. Nonetheless, comparisons with dip scans reveal significant differences in the final results and computational efficiency. Although the general features seem to be similar, detailed features appear to be missing from the dip scan results. For the dip scans, semblance is measured over a number of dips and the highest value is used to determine the dip. The computational cost can vary, depending on factors such as the number of dips tested and implementation, but a comparison indicates that dip scans can be computationally more costly. In contrast, the algorithm to be described uses a single suite of wavelets convolved with the data to produce a number of scale-dependent complex matrices that are summed in a specific way. Furthermore, convolutions may be performed in the frequency domain. This reduces the computational cost, making this algorithm an effective and relatively fast interpretation tool.

Relative time seislet transform

Geophysics ◽  
2020 ◽  
Vol 85 (2) ◽  
pp. V223-V232 ◽  
Author(s):  
Zhicheng Geng ◽  
Xinming Wu ◽  
Sergey Fomel ◽  
Yangkang Chen
Keyword(s):  
Seismic Data ◽  
Computational Cost ◽  
Lifting Scheme ◽  
Real Data ◽  
Data Sets ◽  
Relative Time ◽  
New Approach ◽  
New Formulation ◽  
Wavelet Lifting ◽  
Seismic Images

The seislet transform uses the wavelet-lifting scheme and local slopes to analyze the seismic data. In its definition, the designing of prediction operators specifically for seismic images and data is an important issue. We have developed a new formulation of the seislet transform based on the relative time (RT) attribute. This method uses the RT volume to construct multiscale prediction operators. With the new prediction operators, the seislet transform gets accelerated because distant traces get predicted directly. We apply our method to synthetic and real data to demonstrate that the new approach reduces computational cost and obtains excellent sparse representation on test data sets.

Deep learning for multitrace sparse-spike deconvolution

Geophysics ◽  
2021 ◽  
pp. 1-64
Author(s):  
Xintao Chai ◽  
Genyang Tang ◽  
Kai Lin ◽  
Zhe Yan ◽  
Hanming Gu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Seismic Data ◽  
Resolution Enhancement ◽  
Parameter Tuning ◽  
Seismic Exploration ◽  
3D Seismic ◽  
Spatial Continuity ◽  
Additional Information ◽  
3D Data ◽  
3D Seismic Data

Sparse-spike deconvolution (SSD) is an important method for seismic resolution enhancement. With the wavelet given, many trace-by-trace SSD methods have been proposed for extracting an estimate of the reflection-coefficient series from stacked traces. The main drawbacks of the trace-by-trace methods are that they neither use the information from the adjacent seismograms and nor take full advantage of the inherent spatial continuity of the seismic data. Although several multitrace methods have been consequently proposed, these methods generally rely on different assumptions and theories and require different parameter settings for different data applications. Therefore, the traditional methods demand intensive human-computer interaction. This requirement undoubtedly does not fit the current dominant trend of intelligent seismic exploration. Therefore, we have developed a deep learning (DL)-based multitrace SSD approach. The approach transforms the input 2D/3D seismic data into the corresponding SSD result by training end-to-end encoder-decoder-style 2D/3D convolutional neural networks (CNNs). Our key motivations are that DL is effective for mining complicated relations from data, the 2D/3D CNNs can take multitrace information into account naturally, the additional information contributes to the SSD result with better spatial continuity, and parameter tuning is not necessary for CNN predictions. We report the significance of the learning rate for the training process's convergence. Benchmarking tests on the field 2D/3D seismic data confirm that the approach yields accurate high-resolution results that are mostly in agreement with the well logs; the DL-based multitrace SSD results generated by the 2D/3D CNNs are better than the trace-by-trace SSD results; and the 3D CNN outperforms the 2D CNN for 3D data application.

Impact of 3D Seismic Technologies on Sedimentology Study

2012 ◽  
Vol 463-464 ◽  
pp. 1041-1046
Author(s):  
Ru Tai Duan ◽  
Zhen Kui Jin ◽  
Chong Hui Suo
Keyword(s):  
High Resolution ◽  
Seismic Data ◽  
Vital Role ◽  
Rock Properties ◽  
Data Mapping ◽  
3D Seismic ◽  
Analytical Methodology ◽  
New Approach ◽  
Basin Scale ◽  
3D Seismic Data

Progress of 3D seismic technologies has played a vital role in the developments of sedimentology in terms of analytical methodology and concepts. High-density and high-resolution 3D seismic data can be used to reconstruct 3D views of sedimentary paleo-evironment by direct imaging of depositional elements and can also be used to analyze sedimentary paleo-evironment evolution in 3D detail by mapping facies variability at a specific geologic time by slicing though it. And such data connected with well logging data can be used for predictions of rock properties distribution to delineate sedimentologic heterogeneity. High resolution of 3D seismic data mapping can also be used to image the geometry of diagenesis front to a resolution of a few meters over thousands of square kilometers, which is a new approach to the study of diagenesis process in basin scale. The potential for future developments in this field is considerable. Relative methods and examples of such Studies on the aspects mentioned above are presented.

Sobel filter for edge detection of hexagonally sampled 3D seismic data

Geophysics ◽  
2016 ◽  
Vol 81 (6) ◽  
pp. N41-N51 ◽  
Author(s):  
Haroon Ashraf ◽  
Wail A. Mousa ◽  
Saleh Al Dossary
Keyword(s):  
Edge Detection ◽  
Seismic Data ◽  
Computational Cost ◽  
3D Seismic ◽  
Quality Of Data ◽  
Hexagonal Sampling ◽  
Efficient Data ◽  
2D Data ◽  
2D And 3D ◽  
3D Seismic Data

In today’s industry, automatic detection of geologic features such as faults and channels is a challenging problem when the quality of data is not good. Edge detection filters are generally applied for the purpose of locating such features. Until now, edge detection has been carried out on rectangularly sampled 3D seismic data. The computational cost of edge detection can be reduced by exploring other sampling approaches instead of the regular rectangular sampling commonly used. Hexagonal sampling is an alternative to rectangular sampling that requires 13.4% less samples for the same level of accuracy. The hexagonal approach is an efficient method of sampling with greater symmetry compared with the rectangular approach. Spiral architecture can be used to handle the hexagonally sampled seismic data. Spiral architecture is an attractive scheme for handling 2D images that enables processing 2D data as 1D data in addition to the inherent hexagonal sampling advantages. Thus, the savings in number of samples, greater symmetry, and efficient data handling capability makes hexagonal sampling an ideal choice for computationally exhaustive operations. For the first time to our knowledge, we have made an attempt to detect edges in hexagonally sampled seismic data using spiral architecture. We compared edge detection on rectangular and hexagonally sampled seismic data using 2D and 3D filters in rectangular and hexagonal domains. We determined that hexagonal processing results in exceptional computational savings, when compared with its rectangular processing counterpart.

ChannelSeg3D: channel simulation and deep learning for channel interpretation in 3D seismic images

Geophysics ◽  
2021 ◽  
pp. 1-50
Author(s):  
Hang Gao ◽  
Xinming Wu ◽  
Guofeng Liu
Keyword(s):  
Seismic Data ◽  
Structural Deformation ◽  
3D Seismic ◽  
Channel Models ◽  
Meandering Channel ◽  
Seismic Discontinuities ◽  
Seismic Volumes ◽  
Seismic Images ◽  
Channel Structures

Seismic channel interpretation involves detecting channel structures which often appear as meandering shapes in 3D seismic images. Many conventional methods are proposed for delineating channel structures using different seismic attributes. However, these methods are often sensitive to seismic discontinuities (e.g., noise and faults) that are not related to channels. We propose a convolutional neural network (CNN) method to improve the automatic channel interpretation. The key problem in applying the CNNs method into channel interpretation is the absence of the labeled field seismic images for training the CNNs. To solve this problem, we propose a workflow to automatically generate numerous synthetic training datasets with realistic channel structures. In this workflow, we first randomly simulate various meandering channel models based on geological numerical simulation. We further simulate structural deformation in the form of stratigraphic folding referred to as “folding structures” and combine them with the previously generated channel models to create reflectivity models and the corresponding channel labels. Convolved with a wavelet, the reflectivity models can be transformed into learnable synthetic seismic volumes. By training the designed CNN with synthetic seismic data, we obtain a CNN which learns the characterization of channel structures. Although trained on only synthetic seismic volumes, this CNN shows an outstanding performance on field seismic volumes. This indicates that the synthetic seismic images created in this workflow are realistic enough to train the CNN for channel interpretation in field seismic images.

scholarly journals ADAPTIVE SINGULAR VALUE DECOMPOSITION FILTERING TO ENHANCE REFLECTORS AND GEOLOGICAL STRUCTURES IN 3D SEISMIC DATA

2016 ◽  
Vol 34 (2) ◽  
Author(s):  
Washington Oliveira Martins ◽  
Milton José Porsani ◽  
Michelângelo G. da Silva
Keyword(s):  
Singular Value Decomposition ◽  
Adaptive Filtering ◽  
Seismic Data ◽  
Singular Value ◽  
3D Seismic ◽  
Data Filtering ◽  
Geological Structures ◽  
Value Decomposition ◽  
Seismic Volumes ◽  
3D Seismic Data

ABSTRACT. We applied an adaptive seismic data filtering method, based on the singular value decomposition (SVD) to improve the identification of reflectors and geological structures in 3D stacked seismic volumes...Keywords: seismic data processing, SVD filtering, 3D pos-stacked filtering, adaptive filtering. RESUMO. Nós aplicamos um método de filtragem adaptativa de dados sísmicos, baseado na decomposição em valores singulares (SVD), para melhorar a identificação de refletores e estruturas geológicas em volumes sísmicos empilhados 3D...Palavras-chave: processamento sísmico, filtragem SVD, filtragem pós-stack 3D, filtragem adaptativa.

scholarly journals Detection and extraction of fault surfaces in 3D seismic data

Geophysics ◽  
2006 ◽  
Vol 71 (4) ◽  
pp. P21-P27 ◽  
Author(s):  
Israel Cohen ◽  
Nicholas Coult ◽  
Anthony A. Vassiliou
Keyword(s):  
Seismic Data ◽  
Semantic Representation ◽  
Region Of Interest ◽  
3D Seismic ◽  
Fault Surface ◽  
Directional Filtering ◽  
Seismic Discontinuities ◽  
Ultimate Result ◽  
Seismic Volumes ◽  
3D Seismic Data

We propose an efficient method for detecting and extracting fault surfaces in 3D-seismic volumes. The seismic data are transformed into a volume of local-fault-extraction (LFE) estimates that represents the likelihood that a given point lies on a fault surface. We partition the fault surfaces into relatively small linear portions, which are identified by analyzing tilted and rotated subvolumes throughout the region of interest. Directional filtering and thresholding further enhance the seismic discontinuities that are attributable to fault surfaces. Subsequently, the volume of LFE estimates is skeletonized, and individual fault surfaces are extracted and labeled in the order of decreasing size. The ultimate result obtained by the proposed procedure provides a visual and semantic representation of a set of well-defined, cleanly separated, one-pixel-thick, labeled fault surfaces that is readily usable for seismic interpretation.

Automatic channel detection using deep learning

2019 ◽  
Vol 7 (3) ◽  
pp. SE43-SE50 ◽  
Author(s):  
Nam Pham ◽  
Sergey Fomel ◽  
Dallas Dunlap
Keyword(s):  
Neural Network ◽  
Seismic Data ◽  
Field Data ◽  
3D Seismic ◽  
Data Set ◽  
Channel Detection ◽  
Detection Model ◽  
Uncertainty Measurement ◽  
Human Interpretation ◽  
Seismic Volumes

We have developed a method based on an encoder-decoder convolutional neural network for automatic channel detection in 3D seismic volumes. We use two architectures borrowed from computer vision: SegNet for image segmentation together with Bayesian SegNet for uncertainty measurement. We train the network on 3D synthetic volumes and then apply it to field data. We test the proposed approach on a 3D field data set from the Browse Basin, offshore Australia, and a 3D Parihaka seismic data in New Zealand. Applying the weights estimated from training on 3D synthetic volumes to a 3D field data set accurately identifies channel geobodies without the need for any human interpretation on seismic attributes. Our proposed method also produces uncertainty volumes to quantify the trustworthiness of the detection model.

scholarly journals COMPRESSION OF SEGMENTED 3D SEISMIC DATA

2004 ◽  
Vol 02 (03) ◽  
pp. 269-281 ◽  
Author(s):  
VALERY A. ZHELUDEV ◽  
DAN D. KOSLOFF ◽  
EUGENE Y. RAGOZA
Keyword(s):  
Wavelet Transform ◽  
Seismic Data ◽  
Preliminary Investigation ◽  
Vertical Direction ◽  
3D Seismic ◽  
Discrete Cosine Transforms ◽  
Coding Scheme ◽  
Seismic Volumes ◽  
Spiht Coding ◽  
3D Seismic Data

We present a preliminary investigation of compression of segmented 3D seismic volumes for the rendering purposes. Promising results are obtained on the base of 3D discrete cosine transforms followed by the SPIHT coding scheme. An accelerated version of the algorithm combines 1D discrete cosine transform in vertical direction with the 2D wavelet transform of horizontal slices. In this case the SPIHT scheme is used for coding the mixed sets of cosine-wavelet coefficients.

Predictive painting of 3D seismic volumes

Geophysics ◽  
2010 ◽  
Vol 75 (4) ◽  
pp. A25-A30 ◽  
Author(s):  
Sergey Fomel
Keyword(s):  
Data Analysis ◽  
Plane Wave ◽  
Numerical Algorithm ◽  
Seismic Data ◽  
Local Structure ◽  
Field Data ◽  
Seismic Events ◽  
3D Seismic ◽  
Data Space ◽  
Seismic Volumes

Predictive painting is a numerical algorithm that spreads information in 3D volumes according to the local structure of seismic events. The algorithm consists of two steps. First, local spatially variable inline and crossline slopes of seismic events are estimated by the plane-wave-destruction method. Next, a seed trace is inserted in the volume, and the information contained in that trace is spread inside the volume, thus automatically “painting” the data space. Immediate applications of this technique include automatic horizon picking and flattening in applications to both prestack and poststack seismic data analysis. Synthetic and field data tests demonstrate the effectiveness of predictive painting.

Sign in / Sign up

Export Citation Format

Share Document