In this paper we propose to relate seismic data to facies or petrophysical properties through a colocated window of seismic information instead of the single colocated seismic datum. In this case, the analysis was performed with field data from western venezuela. Multiple available methods deterministic, stochastic and advanced neural network that integrate all the available log and seismic attributes data are available. The classification was done with four attribute signature locations. Seismic interpretation requires the repetitive application of pattern and texture recognition of seismic images, informed by the geologic understanding of a skilled interpreter. Dr smith has a good track record of understanding how the industry works with new technology. Petrel seismic survey design plugin schlumberger software. The method of recognizing a 3d object depends on the properties of an object.
During the crisp seismic survey in 2011 we collected an 11 x 55 km grid of 3d seismic reflection data and highresolvability em122 multibeam data, with coverage extending from the incoming plate to the outershelf. The suite features guided workflows, interactive world maps, 3d imaging, and a 2d color mapping capability that is integrated and interactive with 3d imaging. Pattern recognition by dtw and series data mining in 3d stratum modelling and 3d visualization abstract. Pattern recognition can be defined as the classification of data based on knowledge already gained or on statistical information extracted from patterns andor their representation.
Machine learning deals with the construction and study of systems that can learn from data, rather than follow only explicitly programmed instructions whereas pattern recognition is. Imagine something like drawing number 2 and the program can tell, that it is a number 2 but in 3d and realtime. The seismic data used for the generation of the proposed dataset is a public 3d seismic survey called netherlands offshore f3 block which is available at the open seismic repository. Multiple available methods deterministic, stochastic and advanced neural network that integrate all the available log and seismic attributes. Seismic attribute selection for machinelearningbased facies analysis. In this chapter, the basic concepts of pattern recognition is introduced, focused mainly on.
After his return to the netherlands, he headed a team for the development of 3d interpretation methods using multiattribute statistical and pattern recognition analysis on workstations. From regional exploration to reservoir development and production optimization, geophysics is a critical tool to solve the most complex structural and stratigraphic challengesincluding advanced seismic processing, depth imaging, and 3d, 2d, and prestack seismic interpretation as well as advanced quantitative interpretation. About us acteq 3d seismic survey design software and services. Hi, phd student based in the uk, mostly a geologistpetrophysicist but also interpreting 2d and 3d seismic data over glacial sediments. During the crisp seismic survey in 2011 we collected an 11 x 55 km grid of 3d seismic reflection data and highresolvability em122 multibeam data, with coverage extending from the incoming. Basemap, shapefiles, mapping improves the user interaction in opendtect. Dec 17, 2014 cutlines for a dense 3d survey at surmont field, alberta, canada.
Therefore, the emerging machine learning techniques, particularly the convolutional neural networks, appear most suitable for tackling the problem of annotating various patterns existing in a seismic dataset therefore, implementing deep neural networks into 3d seismic interpretation is the current research focus in the community. Matrox imaging library mil pattern recognition tools. Oct 07, 2010 the som is a powerful cluster analysis and pattern recognition method developed by professor teuvo kohonen of finland during the 1970s and 80s. Tesseract is a powerful integrated survey design package for land, marine, seabed and mixed mode projects.
Seismic geometric decomposition seismic processing. The tremendous amount of samples from numerous seismic attributes exhibit significant organizational structure. Wavelet transform with generalized beta wavelets for seismic timefrequency analysis. It is well suited for indepth hruhr marine seismic processing, realtime marine 2d3d seismic qc, onboard fast track processing, land and marine offline seismic qc, complete processing of nearsurface seismic data reflection, refraction, tomography, masw and vsp processing. Intelligent seismic inversion workflow for highresolution. Welllogging data of strata is taken as time series. For this discussion, seismic data are represented by a 3d seismic survey data volume regularly sampled in location x or y and in time t or in depth z. Developing a seismic texture analysis neural network for. Unsupervised learning endeavors to bring out hidden patterns or structure in the volume of input data through pattern recognition algorithms, without any prior knowledge about the desired output. In forested areas, a designer may choose a pattern that minimizes the number of trees that need to be felled.
Excel 20032007 spreadsheet for quality control of 3d seismic survey geometry for geophysicists and geologists. The teapot dome 3d survey is a land 3d data set from wyoming provided by the u. Rockpredictor uses a 3d geocellular grid, seismic attributes and well data to propagate in 3dimensions, key rock properties such as facies, toc, brittleness, porosity and natural fracture density. Some applications of pattern recognition to oil and gas. The dataset consists of 384km 2 of time migrated 3d seismic data, with 651 inlines and 951 crosslines, located at the north sea, netherlands offshore figure 1. He has performed pioneering work on seismic elastic modeling, seismic pattern recognition and artificial intelligece, reservoir monitoring, and induced seismicity. Building a global seismic texture interpretation network. Mil includes two tools for performing pattern recognition. The application of pattern recognition to oil and gas prospection is very recent, and results are still sparse. Today, paradise distils a variety of information from many attributes simultaneously at full seismic resolution, i.
Seismic interpretation software seisware open inventor 3d sdk. Pattern recognition by dtw and series data mining in 3d. Development of reservoir characterization models using core, well log, and 3d seismic data and intelligent software. Each survey is represented by several attributes, f 1, f 2 f f for example, the attributes might include the amplitude, hilbert transform, envelope, phase, frequency, etc. After obtaining an ftp password, post stack data, well logs, production history, and gis data can be downloaded from the internet. I would like to ask, how hard is it to implement usable pattern recognition system in 3d space. Intelligent seismic inversion workflow for highresolution reservoir characterization. Theres a bit of a gap in the literature on forward modelling of glacial features, and i would quite like to have a go at some forward 2d modelling of what features would look like. Omni 3d is a powerful program designed for seismic survey design and modeling. Therefore, the interpretation of seismic facies has.
Pattern or pattern recognition is the process of taking in raw data and taking an action based on the category of the pattern duda et al. Cutlines for a dense 3d survey at surmont field, alberta, canada. Fred has served the seg as president and has also served on numerous seg committees including the research committee and global affairs committee. They originate from image analysis and are based on the grey level cooccurrence matrix glcm, which describes the relationship between pixels and was developed to capture the. Seismic attribute characteristics are often associated with reservoir lithology and hydrocarbon potential. Pdf seismic facies analysis based on kmeans clustering. As a result, stemmer imaging has developed its cvb polimago software package that has been designed specifically for recognition of objects in various poses. At the conclusion of 3d data processing, the area spanned by a 3d seismic image is divided into a grid of small, abutted subareas called stacking bins. Principal component analysis pca and self organizing maps soms provide multiattribute analyses that have proven to be an excellent pattern recognition approach in the seismic interpretation workflow. Seismic geometric decomposition sgd is an exclusive technique that captures the internal architecture of the seismic reflectors to produce a series of high definition seismic volumes that can be used to improve the delineation of the fault system, the external and internal reservoir architecture and to recognize reflectivity patterns for geomorphological analysis. Apr 22, 2020 radexpro is a seismic processing software on windows. After a period of quality assurance of contractor software for seismic processing, he became responsible for geophysics in the shell learning centre. Pattern recognition is the process of recognizing patterns by using machine learning algorithm. Smith has been a member of the seg since 1967 and is a professional member of seg, gsh, hgs, eage, sipes, aapg, sigma xi, ssa and agu.
Steve carlson senior geophysical advisor ecopetrol. You can quickly and effectively analyze and monitor their seismic acquisition projects. Introduction to 2d3d seismic data acquisition and processing. Threedimension 3d modeling and visualization of stratum plays important role in seismic active fault detection, of course in geoinformation science. Free automated pattern recognition software that recognizes over 170 patterns works on win xp home edition, only, including chart patterns and candlesticks, written by internationally known author and trader thomas bulkowski. The kmeans algorithm jancey, 1966 is one of the first pattern recognition algorithms which was considered to analyze the seismic data, and it is. Som is a powerful nonlinear cluster analysis and pattern recognition approach that helps interpreters identify patterns in their data, some of which can relate to desired geologic characteristics. Theres a bit of a gap in the literature on forward modelling of glacial features, and i would quite like to have a go at some forward 2d modelling of what features. Williamsongeological pattern recognition and modeling with a general. The program can be used for the planning, execution and analysis of land, marine, transition zone, vsp and multicomponent surveys. Seismic facies analysis is considered as a technique for mapping geological changes using seismic data.
Paradise executes and manages workflows based on advanced pattern recognition methods, including selforganizing maps som and principal component analysis pca. The algorithm is applicable to differentiate multiple seismic patterns, which provide an avenue for full pattern recognition by building multiple texture calibration templates models. The kmeans algorithm jancey, 1966 is one of the first pattern recognition algorithms which was considered to analyze the seismic data, and it is used up to now coleou et al. These developments continued into the new millennium, with enhanced visualization and 3d computa. The petrel seismic survey design ssd plugin is a powerful survey design tool fully integrated into the petrel environment. These tools are primarily used to locate complex objects for guiding a gantry, stage, or robot, or for directing subsequent measurement operations. Fred aminzadeh has made significant contributions to the field of exploration geophysics and seismic signal processing through his numerous books, patents, and publications, with a focus on technical advances in discipline boundaries of applied geophysics, petroleum engineering, computer science, and electrical engineering. Acteq 3d seismic survey design software and services 3d. For simplicity, many existing algorithms have focused on recognizing rigid objects consisting of a single part, that is, objects whose spatial transformation is a euclidean motion. We will use omni 3d seismic survey design software to demonstrate acquisition concepts such as geometry layout, azimuth, fold, and offset. This paper presents an introduction to pattern recognition, a summary of previous applications in seismic processing, and several new pattern recognition approaches. Aimachine learning ml promises to alleviate the repetitive nature of this task. I have the opportunity to work on realtime pattern recognition in 3d space, but i have no previous experiences. Som on a 3d seismic survey consisting of a large number.
Literature on 3d seismic cubes and deep learning github. Dynamic warping of seismic images dave hale center for wave phenomena, colorado school of mines, golden co 80401, usa. Seismic attributes a promising aid for geologic prediction. There are a number of ways to lay out sources and receivers for a 3d seismic survey. A 2d seismic line is treated as a 3d survey of one line. We employ a seismic metaattribute workflow to detect and analyze probable faults and fluidpathways in 3d within the sedimentary section offshore southern costa rica. Pdf application of deep learning in first break picking. Geologic pattern recognition from seismic attributes. In the case study shown above, we present results based on som on a 3d seismic survey consisting of a large number of seismic attributes. Seismic pattern recognition has been developing quietly but steadily for twenty years, and the first practical applications are now appearing. Current issue society of exploration geophysicists.
In forested areas, a designer may choose a pattern that minimizes the number of. A seismic attribute is any measurable property of seismic data, such as amplitude, dip, phase, frequency. In recent years, the number of seismic attributes and the size of seismic data have been increased. The som is a powerful cluster analysis and pattern recognition method developed by prof. The variation of seismic data from one window to another. But no matter how new and sophisticated the algorithm, seismic pattern recognition rests on an old and simple foundation. Poststack hydrocarbon prediction methods include pattern recognition, neural networks, and spectral properties. A pattern recognition approach for automatic horizon. Prospect generation and risking using quantitative seismic techniques including multiattribute analysis, pattern recognition, waveform classification, avo, and seismic inversion. One of the important aspects of the pattern recognition is its. Oct 10, 2018 the company launched the paradise multiattribute analysis software in 20, which uses machine learning and pattern recognition to extract greater information from seismic data.
Oct 23, 2018 building a global seismic texture interpretation network october 23, 2018 the primary goal of seismic interpretation is to understand seismic signals, categorize them into various patterns, connect each pattern with a specific depositional event, and finally reconstruct the geologic history. Omni 3d seismic survey design software helps you create optimal 2d and 3d designs for land, marine, oceanbottom cable obc, transition zone, vertical seismic profile vsp, and multicomponent surveys. The som is a powerful cluster analysis and pattern recognition method developed by professor teuvo kohonen of finland during the 1970s and 80s. Neural network technology is used today in financial services software, pattern recognition systems, and many other settings. This is one of the many differences in the application of machine learning and pattern recognition methods available in. Texture attributes are popular in image processing and are used in seismic interpretation to carry out seismic facies analysis and highlight geomorphological features. Comparison and joint evaluation of timelapse pressure and seismic tomography.
Di2018 developing a seismic pattern interpretation network spinet for automated seismic interpretation 1810. Modeling conditional distributions of facies from seismic. Numerous acquisition geometries are supported, including 2d and 3d narrow, wide, and multi and full azimuth. To help work with seismic attributes, dr smith suggests the use of advanced pattern recognition methods such as unsupervised neural networks and principal component analysis. Geo expro seismic interpretation with machine learning. The direct impact on the teaching phase is that a given pattern must be taught in all of its variable positions so that it can be reliably detected in the acquired image. Where land access is easier, designers may opt for a pattern that is efficient for the recording crew to deploy and pick up receivers. In this paper, an interpreter computer interactive software, named seisart, is introduced which is employed for seismic facies analysis. Unsupervised neural networks disruptive technology for. Supervised seismic facies analysis based on image segmentation. The petrel ssd plugin enables users to plan, edit, execute, and analyze marine, land, obc, and vsp surveys. Unsupervised neural networksdisruptive technology for. These technologies are simple, intuitive, and have rapid calculating abilities but a poor forecasting repeatability. The previous screening assumption prevents recognizing and hence utilizing these patterns of seismic data.
1566 1589 555 1207 496 935 1360 278 1539 470 896 1508 1468 310 1543 108 1228 148 1479 1020 1312 893 1361 1317 427 774 591 1459 111 1041 321 1207