Integrated seismic texture segmentation and clustering analysis to improved delineation of reservoir geometry

Abstract

In recent years, 3D volumetric attributes have gained wide acceptance by geosciences interpreters. The early introduction of single-trace complex trace attributes was quickly followed by seismic sequence attribute mapping workflows. 3D geometric attributes such as coherence and curvature are also widely used. Most of these attributes correspond to a very simple easy-to-understand measures of a waveform or surface morphology. However, not all geologic features can be so easily quantified. For this reason, simple statistical measures of the seismic waveform such as RMS amplitude prove to be quite valuable in delineating more chaotic stratigraphy. In this paper, we show how modern texture analysis based on the gray-level co-occurrence matrix, when coupled with recent developments in self-organizing maps clustering technology, extends such statistical measures to delineate features that geoscientists can see, but not easily describe.

Scale-free Texture Segmentation: Expert Feature-based versus Deep Learning strategies

“…In the last decade, the tremendous increase in computational/storage capabilities has triggered a massive use of deep learning for segmentation, because deep convolutional neural networks have the potential to discover and aggregate relevant information from large scale shapes to fine scale structures. Recently, numerous real-world applications, related to biological tissues [3], [4], geological samples [5], satellite images [6],. .…”

Integrated seismic texture segmentation and cluster analysis applied to channel delineation and chert reservoir characterization

“…Such texture attributes hold significant promise in quantifying geological features such as mass transport complexes, amalgamated channels, and dewatering features that exhibit a distinct lateral pattern beyond simple edges. Like seismic waveform classification (Coléou et al, 2003), and spectral components, GLCM attributes are amenable to subsequent clustering analysis using selforganizing and generative-topographic maps (Angelo et al, 2009;West et al, 2002;Gao, 2007;Wallet et al, 2009).…”

Seismic characterization of a Mississippi Lime resource play in Osage County, Oklahoma, USA

“…Works by Nissen et al (2006), , White et al (2012), and White (2013) show the relationship between seismic curvature attributes and fracture trend detection. Angelo (2010), Yenugu et al (2010, Matos et al (2011), and Roy et al (2012) use gray level co-occurrence matrix (GLCM) seismic texture attributes, selforganizing maps (SOM), and unsupervised 3D seismic facies analysis to identify petrofacies within the Mississippi Lime. Finally, work by Dowdell et al (2012) and Dowdell (2013) show the use of poststack and simulta-neous prestack seismic inversion to predict zones of tripolitic chert using petrophysical properties measured on well logs.…”