Extraction of 3D structural data from Virtual Outcrop Models: problems and best practices.

• DOI: 10.5194/egusphere-egu23-7446
• Publication Date: 2023-02-25T20:12:41Z
• Authors: Stefano Tavani; Amerigo Corradetti; Marco Mercuri
• Keywords: Outcrop

The rapid improvements of computer vision–based photogrammetric image processing pipelines (i.e., Structure from Motion–Multi View Stereophotogrammetry: SfM-MVS), coupled with the availability various low-cost and portable acquisition tools, such as Digital Single-Lens Reflex (DSLR), mirrorless cameras, Unmanned Aerial Vehicle (UAV) even smartphones, have revolutionized outcrop studies in structural geology brought traditional field into digital age. This has had a transformative impact on Virtual Outcrop Models (VOMs), which been promoted mostly visualization media to fully interrogable quantitative objects. Among several applications VOMs geology, extraction near planar features (e.g., fracture bedding surfaces) is one most important. Various procedures aimed at this purpose exist, spanning automated segmentation best fitting point clouds manual picking 3D polylines both textured meshes.Here we illustrate pros cons, practices, drawbacks main for geological data VOMs. While or supervised recognition subsequent best-fitting coplanar patches received remarkable attention, its application generally limits rare case studies. Indeed, commonly, outcrops do not expose surfaces are large enough carry out robust fitting, interpretation only permits procedures. In latter case, use meshes must be preferred clouds, during digitization accuracy mesh considered, well intrinsic roughness any surfaces. analysis coplanarity collinearity picked pointsets may help identifying traces that diverge idealized (low) collinear (high) configurations. However, typically suggested threshold values often produces small datasets. Nonetheless, goodness based merely visual inspection best-fit plane, handling real-time through live computation planes pointsets, acceptable.