MATLAB code for fault individualization from 2017. The code is based on simple image processing tools to automatically filter, separate and extract individual 3D faults from a fault volume, such as a fault likelihood cube (Hale, 2013), or a fault volume resulting from supervised fault identification.
The method is described in Bugge et al., 2018:
Bugge, A. J., S. R. Clark, J. E. Lie, and J. I. Faleide, 2018, A case study on semiautomatic seismic interpretation of unconformities and faults in the southwestern Barents Sea: Interpretation, 6, SD29-SD40
The method for semi-automatic fault extraction is based on the assumption that each fault surface can be targeted as a 3D object in a binary representation of a fault volume. Prior to this, interference within the cube is addressed by separating intersecting faults through morphological filter operations and by assigning objects to different dip cubes. Objects that meet a set of user-defined filter criterions, related to size, are extracted from each of the dip cubes and assumed to represent individual fault surfaces.
Figures from Bugge et al., 2018. Automatically filtered and individualized faults from a 3D fault likelihood volumes (Hale, 2013). The figure to the right shows three individual fault surfaces after being exported to petrel.