The Extraction of Fractional-order Curvature Attributes and Applications in the Structure Interpretation

Abstract

Curvature attribute interpretation is a new interpretation technology rising in the recent two years. Curvature attributes provide images of structure and stratigraphy that complement those seen by the well accepted coherence algorithms. The article expounds the definition of the curvature, the physical meaning and its shortages in structural interpretation from the most basic mathematical definition. On this basis, we describe 2D curvature of the surface, list and compare the definition, physical meaning and its main usages of surface-derived attributes. After interpolation and 2D median filtering of t0 contour data, we extract 8 attributes with fractional-order method including mean curvature, Gaussian curvature, maximum curvature, minimum curvature, most positive curvature, most negative curvature, dip curvature, and strike curvature. Finally, we apply most positive curvature to detect fault and fracture and achieve better results.