Digital Outcrop Model Reconstruction With Geological Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for reconstructing Digital Outcrop Models (DOMs) result in low visual quality due to the absence of segmentation and decimation processes, leading to excessive triangles in the triangular mesh, high computational costs, and reliance on powerful GPU systems.
Innovation Solution
A method involving segmentation and decimation of dense clouds and triangular meshes based on geological relevance levels to optimize DOMs, allowing high visual quality without requiring robust GPUs, using image acquisition devices, alignment, densification, and texturing processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional SfM-MVS algorithms are applied without segmentation and decimation, then complete geometric representation is achieved, but file size and computational cost increase excessively
Solution Approach 1:
The patent applies segmentation by dividing the outcrop into multiple areas of geological interest, each processed independently with appropriate detail levels. This allows the model to maintain geometric completeness in critical regions while reducing detail in less important areas, thereby reducing overall file size without sacrificing essential geometric representation.
Solution Approach 2:
The patent implements local quality control through differential decimation, where different areas of the outcrop are subjected to different levels of geometric simplification based on their geological importance. High-priority areas maintain full geometric detail while low-priority areas undergo aggressive decimation, optimizing the balance between completeness and file size.
2Manufacturing precision
If high-density triangular mesh is generated without decimation, then visual quality is maximized, but processing and storage costs increase
Solution Approach 1:
The patent segments the triangular mesh into multiple portions corresponding to different areas of geological interest. Each portion is then independently decimated according to its priority level, allowing high visual quality to be maintained in critical areas while reducing polygon count in less important areas, thereby reducing processing and storage costs.
Solution Approach 2:
The patent applies differential decimation strategies to different mesh portions based on their geological relevance. High-priority areas undergo minimal decimation to preserve visual quality, while low-priority areas are aggressively simplified. This local quality approach optimizes the trade-off between visual fidelity and processing efficiency.
3Measurement precision
If complete geometric detail is preserved throughout the model, then accuracy in all areas is maintained, but machine resource requirements increase
Solution Approach 1:
The patent divides the outcrop into multiple areas and processes each with appropriate detail levels. This segmentation allows the system to maintain high geometric accuracy in areas of geological interest while using coarser representation elsewhere, reducing overall machine resource requirements without sacrificing critical measurement precision.
Solution Approach 2:
The patent implements local quality control by applying different levels of geometric fidelity to different areas based on their geological importance. High-priority areas maintain full geometric accuracy while low-priority areas use simplified representations, optimizing the balance between measurement precision and computational resource requirements.
4Stability of the object's composition
If uniform processing is applied to all outcrop areas, then consistency is maintained, but areas of geological interest lack optimized detail
Solution Approach 1:
The patent segments the outcrop into multiple areas and assigns different processing priorities to each segment. This allows the system to maintain consistent processing workflows while applying area-specific optimization strategies, ensuring that geological areas of interest receive appropriate detail levels without compromising overall model consistency.
Solution Approach 2:
The patent applies local quality control by differentiating processing parameters for different areas based on their geological relevance. Areas of interest receive higher detail and less decimation, while other areas use optimized lower-detail settings. This maintains processing consistency through standardized workflows while achieving superior geological detail accuracy where needed.
Data Source
AI summary
The present invention relates to a method for reconstructing a Digital Outcrop Model (DOM), comprising placing an image acquisition device; capturing, processing and aligning a plurality of images; generating a sparse alignment cloud and performing its densification to generate a dense cloud; performing a first segmentation of the dense cloud into two portions based on two categories of geological relevance levels; carrying out decimation of the dense cloud portion of the low geological relevance category; joining the two portions of dense cloud; generating a three-dimensional triangular mesh; performing a second segmentation of the three-dimensional triangular mesh into two portions based on two categories of geological relevance levels; carrying out decimation of the three-dimensional triangular mesh portions of the low and high geological relevance categories; carrying out texturing of the three-dimensional triangular mesh portions of the high and low geological relevance categories; and joining the two three-dimensional triangular mesh portions to form an object defined as DOM.