Borehole Image Labeling With Optimum Polygons for Deep Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Interpreting borehole images for geological features requires intensive manual work, and there is a lack of properly labeled data for training machine learning models to automate this process.
Innovation Solution
A method and system for automatically converting interpreted borehole images to digital labeled data, using a borehole data acquisition system, a geological locator, and computer processors to determine descriptors, including optimum polygons enclosing geological features, which are then used to create a labeled dataset for training deep learning models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual interpretation of borehole images is performed, then geological features can be identified with high accuracy, but intensive manual work and time consumption are required
Solution Approach 1:
The system performs preliminary automated labeling of borehole images by detecting geological features and generating digital labels with bounding polygons. This preliminary action creates a labeled dataset that can be used for training deep learning models, thereby reducing the need for time-consuming manual interpretation while maintaining identification accuracy.
Solution Approach 2:
The system creates digital copies of manually interpreted borehole images by automatically generating labeled datasets that replicate the geological feature identification results. These digital labels serve as substitutes for manual interpretation, enabling automated training of deep learning models without requiring continuous manual intervention.
2Reliability
If deep learning models are trained with manually labeled data, then model accuracy improves, but the labeling process requires extensive manual effort
Solution Approach 1:
The system enables self-service labeling by automatically detecting geological features in borehole images and generating digital labels without human intervention. The automated labeling system serves itself by using computer vision algorithms to identify and label features such as fractures, faults, and bedding planes, thereby creating high-quality training data with high productivity.
Solution Approach 2:
The system replaces the mechanical manual labeling process with an automated computer-based system. Instead of manually drawing bounding polygons and annotating geological features, the system uses image processing algorithms and deep learning models to automatically detect and label features, substituting human manual work with computational processes.
3Extent of automation
If automated feature detection is implemented, then manual work is reduced, but properly labeled training data is lacking
Solution Approach 1:
The system performs preliminary automated labeling to create a dataset of properly labeled borehole images. This preliminary labeled data serves as training material for deep learning models, enabling subsequent automated feature detection without requiring manual intervention. The system thus prepares the necessary training data in advance to support high-level automation.
Solution Approach 2:
The system establishes a continuous process where automated labeling generates training data, which then trains deep learning models that perform automated feature detection. This continuous cycle of labeling, training, and detection ensures that properly labeled training data is continuously available, supporting sustained automation of the interpretation process.
Data Source
AI summary
A method for determining descriptors associated with borehole images. The method includes obtaining N≥1 borehole images, where N is an integer, and locating, in each borehole image within the N borehole images, one or more geological features associated with the borehole image. The method further includes determining, for each borehole image within the N borehole images, one or more descriptors associated with the borehole image, where each descriptor of the one or more descriptors includes an optimum polygon enclosing a geological feature of the one or more geological features associated with the borehole image.