Well Log Vectorization via Anisotropic Image-Guided Shortest Path
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for digitizing well log curves from raster images are inefficient and prone to errors due to geometric skew, image noise, and the need for manual selection of numerous points, which complicates the vectorization process.
Innovation Solution
An automated image-guided method using the anisotropic image-guided shortest path algorithm for interpolation between manually selected points, allowing for a combination of manual and automatic digitization, and improving image rectification to reduce distortions and noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual point selection is used for curve digitization, then the operator can control the digitization process, but the time required and number of points needed increase significantly
Solution Approach 1:
The system enables self-service automation where the computer automatically selects points along the curve and performs digitization without continuous human intervention. The automated point selection algorithm identifies curve points based on image features, reducing manual labor while maintaining accuracy through algorithmic curve following capabilities
Solution Approach 2:
The manual mechanical process of moving a mouse to select points is replaced with an automated image processing system that uses algorithms to detect and trace curves. The system substitutes human operator actions with computer-based image analysis and automated point selection mechanisms
2Manufacturing precision
If more manual points are selected along the curve, then the curve representation accuracy improves, but the complexity and time of the digitization process increase
Solution Approach 1:
The complex manual process of selecting numerous points is replaced with automated image processing algorithms that intelligently identify and select optimal points along the curve. The system uses computer vision techniques to automatically trace curves and generate digitized representations with appropriate point density
3Measurement precision
If image rectification is performed to remove geometric skew, then the quality of extracted curves improves, but additional processing steps and time are required
Solution Approach 1:
Image rectification is performed as a preliminary step before curve extraction and digitization. The system corrects geometric skew and aligns the image properly beforehand, ensuring that subsequent curve detection and point selection operations work on corrected imagery, which improves overall extraction quality
Solution Approach 2:
The patent combines multiple processing functions including image rectification, curve detection, point selection, and digitization into an integrated automated workflow. These previously separate manual or semi-automated steps are merged into a unified computer-based system that processes the entire digitization pipeline
Data Source
AI summary
A system and method for well log vectorization are described that assist geologists or other technicians to rapidly and accurately vectorize well log curves that are visually evident in a raster well log image. The resulting digital well log curves are instrumental for qualitative interpretation and quantitative analysis of geologic information measured along well bores. The system and method simplifies and makes more efficient and accurate the process of vectorizing well log curves that are displayed in a digital raster log image.