Attribute Expression Using Gray Level Co-Occurrence

Sipuikinene Angelo Marcilio MatosKurt J Marfurt ConocoPhillips School of Geology amp Geophysics University of Oklahoma Seismic resolution remains a major limitation in the world of seismic interpretation The goal of reflection seismology is to analyze seismic amplitude and character to predict lithologic facies and rock properties such as porosity and thickness Seismic attribute analysis is a technique that is commonly used by oil industry to delineate stratigraphic and structural features of interest Seismic attributes are particularly important in allowing the interpreter to extract subtlest at the limits below seismic resolution For example some attributes such as coherence and curvature are particularly good at identifying edges and fractures Attributes such as spectral components tend to be more sensitive to stratigraphic thickness Many commercial seismic interpretation packages contain RMS amplitude and relative impedance which is sensitive to acoustic impedance My proposed research focuses upon seismic textural analysis borrowing upon techniques commonly used in remote sensing to enhance and detect terrain vegetation and land use information Textures are frequently characterized as different patterns in the underlying data Seismic texture analysis was first introduced by Love and Simaan 1984 to extract patterns of common seismic signal character Recently several workers West et al 2002 Gao2003 Chopra and Vladimir 2005 have extended this technique to seismic through the uses of graylevel cooccurrence matricesGLCMThe gray level allows the recognition of patterns significantly more complex than simple edges This set of texture attributes is able to delineate complicated geological features such as mass complex transport and amalgamated channels that exhibit a distinct lateral pattern

Embed this Presentation