Multilayer representation for geological information systems

Abstract

In this thesis we propose the use of the Stack-Based Representation of Terrains (SBRT) for volumetric geological data. This data structure encodes geological structures represented as stacks using a compact data representation. The SBRT is further formalized with a framework based on the geo-atom theory to provide a precise definition and determine its properties. Also, we introduce QuadStacks, a novel data structure that improves the compression results provided by the SBRT, by exploiting in its data arrangement the redundancy often found in layered dataset. This thesis also provides direct visualization methods for the SBR and QuadStacks based on the well-known raycasting algorithm. By keeping the whole dataset in the GPU in a compact way, the methods are fast enough to provide real-time frame rates.