Automated Rock Type Classification via Image Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated tools, such as neural networks, require significant human intervention for feature selection and are limited in their ability to accurately and rapidly classify rock types from images without pre-selecting features, hindering efficient rock classification processes.
Innovation Solution
An automated image classification algorithm and system that partitions images into sub-regions, uses convolutional neural networks for classification, and provides confidence scores, allowing for the identification of rock types without pre-selecting features, applicable to various photo types including core and CT scan images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated tools such as neural networks are used for rock classification, then classification speed is improved, but human intervention for feature selection is still required which reduces automation extent
Solution Approach 1:
The system performs self-service by automatically selecting features and classifying rock types without requiring human intervention for feature selection. The automated classification tool independently identifies relevant features from input images and proceeds with classification, eliminating the need for scientists to manually pre-select features while maintaining high classification speed.
Solution Approach 2:
The classification process is segmented into distinct automated stages: feature selection, feature extraction, and classification. Each stage is handled automatically by the system through separate modular components, allowing the tool to process rock images through multiple processing steps without continuous human intervention, thereby improving both speed and automation extent.
2Measurement precision
If feature pre-selection is performed to improve classification accuracy, then measurement precision is improved, but loss of time due to manual intervention increases
Solution Approach 1:
The system performs preliminary automated feature selection before classification, identifying and selecting relevant features automatically without human intervention. This preliminary action ensures that the most informative features are chosen in advance, maintaining high classification accuracy while eliminating the time loss associated with manual feature selection by scientists.
Solution Approach 2:
The manual mechanical process of feature selection by scientists is replaced with an automated computational system. The automated classification tool uses algorithms to select features based on their relevance to rock type classification, substituting human manual work with automated processing that maintains accuracy while significantly reducing the time required for feature selection.
3Adaptability or versatility
If all information in the image is used for classification without pre-selection, then adaptability is improved, but device complexity increases
Solution Approach 1:
The automated classification tool is designed with universal functionality to handle various types of rock images and photos without requiring customization or pre-selection for each case. The system can process different image types (core photos, CT scans, thin sections, well-log converted images) using the same automated feature selection and classification algorithms, improving adaptability while managing complexity through standardized multi-functional processing.
Solution Approach 2:
The system manages complexity by dynamically adjusting processing parameters based on the input image characteristics. The automated feature selection process identifies relevant parameters and features specific to each image, allowing the classification system to adapt to different rock types and image conditions without requiring complex manual configuration, thereby maintaining versatility while controlling system complexity through adaptive parameter changes.
Data Source
AI summary
A system and method of automated classification of rock types includes: partitioning, by a processing device, an image into partitions; extracting, by the processing device, sub-images from each of the partitions; first-level classifying, by an automated classifier, the sub-images into corresponding first classes; and second-level classifying, by the processing device, the partitions into corresponding second classes by, for each partition of the partitions, selecting a most numerous one of the corresponding first classes of the sub-images extracted from the partition. A method of displaying automated classification results on a display device is provided. The method includes: receiving, by a processing device, an image partitioned into partitions and classified into corresponding classes; and manipulating, by the processing device, the display device to display the image together with visual identification of the partitions and their corresponding classes. The method may include generating a depth-aligned log file of the classifications.