Ground Engaging Tool Wear Assessment with Perspective Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately predict the optimal time to replace worn machine parts, such as ground engaging tools, due to the complexity of wear estimation and the need to consider various operating conditions, leading to inefficiencies in fuel consumption and component wear.
Innovation Solution
A method using image recognition and neural networks to determine wear estimation by measuring the size of machine parts through segmentation polygons, correcting for perspective distortion, and identifying occluded parts, with a system that includes training models for precise wear estimation and reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical measurements are used to estimate wear, then measurement precision is improved, but loss of time increases due to time-consuming measurements
Solution Approach 1:
The patent replaces physical measurement systems with an optical/image-based measurement system. Cameras capture images of the machine part, and image processing algorithms automatically measure dimensions and calculate wear, eliminating the need for manual physical measurements while maintaining accuracy.
Solution Approach 2:
The patent creates a digital copy (image) of the machine part and performs measurements on this copy rather than the physical object itself. This allows wear estimation to be done from images, significantly reducing time while maintaining measurement precision through automated image analysis.
2Loss of time
If image recognition is used to estimate wear, then loss of time is reduced, but measurement precision deteriorates due to perspective distortion and occlusion
Solution Approach 1:
The patent changes the parameter space by introducing perspective distortion correction through scaling factors. The system captures images at multiple positions, calculates perspective distortion parameters, and corrects measurements by applying scaling factors derived from these parameters, thereby maintaining precision despite perspective effects.
Solution Approach 2:
The patent performs preliminary calibration by capturing images at two or more predefined positions to establish perspective distortion characteristics before actual wear measurement. This preliminary action creates reference data that compensates for perspective distortion during subsequent measurements, improving accuracy.
Solution Approach 3:
The patent implements a feedback mechanism where the system evaluates detection confidence levels and uses this feedback to determine whether additional images are needed. If confidence is below a threshold, the system captures more images from different positions, iteratively improving measurement precision through feedback-driven data collection.
3Measurement precision
If multiple images are captured to improve detection accuracy, then measurement precision is improved, but use of energy increases
Solution Approach 1:
The patent applies partial action by capturing multiple images only when necessary - specifically when detection confidence is below a threshold or when perspective distortion correction is needed. The system balances the need for accuracy with energy conservation by using multiple images selectively rather than continuously.
Solution Approach 2:
The system uses feedback from detection confidence levels to control image capture. When confidence is sufficient, no additional images are taken; when confidence is low, the system triggers additional captures. This feedback mechanism optimizes energy usage by capturing images only when needed for accurate measurement.
Data Source
AI summary
A technique is directed to methods and systems for assessing wear for a ground engaging tool (GET), such as a tooth on a bucket on a machine. A GET monitoring system can utilize convolutional neural networks for detecting individual instances of GETs in an image. The GET monitoring system can records the initial state of a GET on a bucket during a calibration process. After calibration, the GET monitoring system can estimate how much the GET has worn down, as length or surface area, since the GET was installed. In some embodiments, as a machine operates the motion of the machine can induce perspective distortion on the captured images of the GETs. The GET monitoring system can correct for perspective distortion, induced by bucket motion, by building a motion history map of the GET.