Bucket Tool Wear Detection Using Pixel Count Trends

Solution Overview

Problem

Existing wear detection systems for ground engaging tools (GET) rely on computationally expensive neural networks and machine learning, which are prone to false positives, making them ineffective for real-time monitoring and resource-intensive.

Innovation Solution

A system and method using a camera and processor to analyze video streams of a work machine's bucket, identifying tool images and determining wear levels based on pixel counts over time, reducing false positives and processing resources by focusing on pixel count changes and trends.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If neural networks and machine learning are used for wear detection, then detection capability is improved, but computational cost and false positives increase

Engineering Contradiction:

Improvewear detection accuracyVSAvoidcomputational cost

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts and isolates the specific feature of pixel count from images, separating it from complex neural network processing. By focusing only on the relevant pixel count metric rather than analyzing entire images with machine learning, the system achieves wear detection without the high computational cost and false positives associated with neural networks.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the complex machine learning system with a simpler computational approach based on pixel counting. This substitution uses basic image processing techniques instead of resource-intensive neural networks, maintaining detection capability while dramatically reducing computational requirements.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If neural networks are used for wear detection, then detection capability is improved, but false positives increase

Engineering Contradiction:

Improvewear detection accuracyVSAvoidfalse positive rate

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent extracts and isolates the specific feature of pixel count from images, separating it from complex neural network processing. By focusing only on the relevant pixel count metric rather than analyzing entire images with machine learning, the system achieves wear detection without the high computational cost and false positives associated with neural networks.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses template images as references to compare against captured images. By creating idealized template representations of GET in various wear states and comparing pixel counts against these templates, the system achieves reliable detection without the false positives that plague machine learning approaches.

Inventive Principle:

Principle #26Copying

3Measurement precision

If video streams are processed for wear detection, then monitoring capability is improved, but processing time increases

Engineering Contradiction:

Improvewear monitoring capabilityVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts and isolates the specific feature of pixel count from images, separating it from complex neural network processing. By focusing only on the relevant pixel count metric rather than analyzing entire images with machine learning, the system achieves wear detection without the high computational cost and false positives associated with neural networks.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by processing only the necessary pixel count information from video frames rather than performing complete image analysis. This selective approach processes only the critical data needed for wear detection, reducing overall processing time while maintaining monitoring capability.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12252870B2Ground engaging tool wear and loss detection system and method

Publication Date: 2025.03.18 CATERPILLAR INC

AI summary

A wear detection system can be configured to receive a video stream including a plurality of images of a bucket of the work machine from a camera associated with the work machine. The bucket has one or more ground engaging tools (GET). The wear detection system can also be configured to identify a plurality of tool images from the video stream over a period of time. The plurality of tool images depict the GET at a plurality of instances over a period of time. The wear detection system can also be configured to determine a plurality of tool pixel counts from the plurality of tool image and determine a wear level for the GET based on the plurality of tool pixel counts.