Automated Petrographic Image Analysis for Rock Sample Characterization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional thin section analysis methods, such as point counting, are expensive, subjective, and prone to human error, limiting their accuracy and efficiency in determining mineral composition and texture, especially in hydrocarbon exploration.
Innovation Solution
An automated system using computer programs to analyze petrographic thin section images by determining feature vectors for each pixel, clustering regions, and segmenting grains, which enables accurate and consistent analysis of rock samples, including porosity and grain properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated image processing methods are used, then analysis speed and consistency are improved, but implementation complexity increases
Solution Approach 1:
The automated analysis system segments the thin section image into multiple processing stages: preprocessing (color normalization, noise reduction), feature extraction (color, texture, shape features), classification (mineral identification using machine learning), and measurement (grain size, porosity calculation). This segmentation allows each stage to be optimized independently while maintaining overall system productivity.
Solution Approach 2:
The patent introduces intermediate data structures and processing layers between the raw image input and final analysis output. Feature vectors serve as intermediaries that transform pixel data into meaningful characteristics, while machine learning models act as intermediary classifiers that bridge image processing and mineral identification. This reduces implementation complexity by creating modular, reusable components.
2Measurement precision
If automated machine learning methods are used, then analysis accuracy and objectivity are improved, but computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary processing steps before applying computationally intensive machine learning algorithms. Image preprocessing (color normalization, contrast enhancement) and feature pre-extraction are performed first to reduce the complexity of the data fed into the neural network. This preliminary action reduces computational requirements while maintaining accuracy.
Solution Approach 2:
The patent implements a hierarchical classification approach where not all image regions are processed with the full machine learning pipeline. Simple regions (clear mineral grains) use straightforward classification, while complex regions (mixed minerals, fine textures) receive more intensive processing. This partial application of complex algorithms reduces overall computational requirements while maintaining accuracy where needed.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for automated analysis of petrographic thin section images. In one aspect, a method includes determining a first image of a petrographic thin section of a rock sample, and determining a feature vector for each pixel of the first image. Multiple different regions of the petrographic thin section are determined by clustering the pixels of the first image based on the feature vectors, wherein one of the regions corresponds to grains in the petrographic thin section. The method further includes determining a second image of the petrographic thin section, including combining images of the petrographic thin section acquired with plane-polarized light and cross-polarized light. Multiple grains are segmented from the second image of the petrographic thin section based on the multiple different regions from the first image, and characteristics of the segmented grains are determined.