Abstract: A geometric method is described for 3D structural restoration of a subsurface model including receiving data representative of a subsurface volume of interest including one or more chronohorizons and the geometry and topology of any faults of relevance; developing a fault framework model of the subsurface volume of interest; selecting a horizon, the deposition of which represents the geologic time to which the structural model should be restored; developing coordinate transformation constrained by a single datum horizon and, optionally, additional geologic constraints; applying the 3D transformation to all geologic features below and, optionally, above the datum surface; and scaling the vertical coordinates to accurately relate vertical and horizontal dimensions. The method may be executed by a computer system.

Abstract: A geometric method is described for 3D structural restoration of a subsurface model including receiving data representative of a subsurface volume of interest including one or more chronohorizons and the geometry and topology of any faults of relevance; developing a fault framework model of the subsurface volume of interest; selecting a horizon, the deposition of which represents the geologic time to which the structural model should be restored; developing coordinate transformation constrained by a single datum horizon and, optionally, additional geologic constraints; applying the 3D transformation to all geologic features below and, optionally, above the datum surface; and scaling the vertical coordinates to accurately relate vertical and horizontal dimensions. The method may be executed by a computer system.

Abstract: An automated method of processing a fault-enhanced 3-D seismic volume (V) to locate and interpret faults is disclosed. The method includes processing of individual lateral slices (LS.sub.j) of the volume (V). For each slice (LS.sub.j), stripe artifacts are eliminated (36) by adjusting pixel values to account for lines that are unduly bright or dim (and thus artifacts of processing). Linear features are enhanced by applying a modified Gumey-Vanderburg algorithm (62) to pixel windows, such that intensity values of pixels are enhanced according to the extent to which the pixels reside in a line. Detection of lines in the enhanced lateral slice is then performed by summing pixel intensities over a window at varying directions (72), and associating, with a center pixel, an amplitude value corresponding to the maximum sum and a direction value associated with this sum.