Raster Log Digitization Using Segmentation Masks for Header Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately digitize and identify headers, plots, and tracks in raster images of logs due to their underrepresentation, which hinders effective analysis and processing of geologic data.
Innovation Solution
A method and system utilizing a trained machine learning model to generate segmentation masks for raster images of logs, enabling the identification and digitization of headers, plots, and tracks, thereby enhancing data processing and analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional digitization methods are used to process raster images of logs, then processing speed may be maintained, but identification accuracy of headers, plots, and tracks deteriorates due to their underrepresentation
Solution Approach 1:
The system performs preliminary segmentation of the raster image into distinct regions (headers, plots, tracks) before detailed analysis. This pre-processing step creates structured data that improves subsequent identification accuracy without requiring complex processing during the main analysis phase
Solution Approach 2:
The patent introduces an intermediary processing layer that uses machine learning models to bridge the gap between raw raster images and structured log data. This intermediary system handles the complexity of identifying underrepresented elements while providing accurate results to the final output
2Measurement precision
If manual digitization methods are used to ensure accurate identification of underrepresented elements, then identification accuracy improves, but processing time and productivity deteriorate
Solution Approach 1:
The system employs self-service mechanisms where machine learning models automatically identify and segment log elements without human intervention. The models learn from training data to autonomously distinguish headers, plots, and tracks, maintaining high accuracy while enabling rapid automated processing
Solution Approach 2:
The patent transforms the processing approach by changing parameters from manual inspection to automated machine learning-based detection. This parameter change enables the system to process images rapidly while maintaining identification accuracy through trained algorithms that recognize patterns in the raster data
3Productivity
If automated processing is used to maintain productivity, then processing speed is maintained, but identification accuracy of underrepresented elements deteriorates
Solution Approach 1:
The system performs preliminary segmentation of the raster image into distinct regions (headers, plots, tracks) before detailed analysis. This pre-processing step creates structured data that improves subsequent identification accuracy without requiring complex processing during the main analysis phase
Solution Approach 2:
The patent incorporates feedback mechanisms where the machine learning model receives input about the segmented regions and adjusts its processing to improve identification accuracy. The system uses the segmented data as feedback to refine its detection of underrepresented elements while maintaining automated processing speed
Data Source
AI summary
A method can include receiving raster images of logs, where the logs include plots, tracks and headers and where, compared to the plots and the tracks, the headers are underrepresented; processing the raster images using a trained machine learning model to generate segmentation masks for the logs for identification of the plots, the tracks and the headers; and generating digitized versions of the logs using the segmentation masks, where each of the digitized versions of the logs includes a digitized plot with an associated digitized header and an associated digitized track.