Optical Inspection Vectors for Assembly Unit Defect Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optical inspection methods struggle to accurately predict defects in assembly units in real-time and retrospectively identify defective units without comprehensive knowledge of feature effects on aesthetics and functionality.
Innovation Solution
A method utilizing optical inspection data to generate multi-dimensional vectors representing assembly unit features, grouping them into clusters, and flagging defects or anomalies based on vector proximity and trend analysis, enabling real-time and historical defect prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical inspection methods are used to detect defects in assembly units, then defect detection capability is provided, but the ability to predict defects and identify defective units without comprehensive knowledge of feature effects is insufficient
Solution Approach 1:
The system performs preliminary actions by extracting features and generating vectors before final defect determination. Features are extracted from inspection images and transformed into multi-dimensional vectors that capture assembly unit characteristics, enabling predictions to be made in advance of final inspection results.
Solution Approach 2:
Multi-dimensional vectors serve as an intermediary between raw inspection images and defect predictions. These vectors transform visual data into a format that can be processed by machine learning models, bridging the gap between image processing and defect prediction capabilities.
2Productivity
If real-time defect prediction is implemented using optical inspection data, then productivity is improved by reducing waste, but device complexity increases due to vector generation and clustering processes
Solution Approach 1:
The system segments the defect detection process into distinct stages: feature extraction, vector generation, clustering, and defect prediction. Each stage handles a specific portion of the data processing workload, making the overall complex system more manageable and implementable through modular components.
Solution Approach 2:
The system transforms inspection images into multi-dimensional vectors by changing the parameter representation from raw pixel data to extracted feature parameters. This transformation enables the use of machine learning algorithms that operate on these transformed parameters to make real-time predictions.
3Reliability
If comprehensive knowledge of feature effects on aesthetics and functionality is required for accurate defect prediction, then prediction accuracy improves, but ease of operation decreases due to the need for extensive training data and expertise
Solution Approach 1:
The system performs self-service by automatically extracting features and generating vectors from inspection images without requiring manual intervention or extensive training data input. The machine learning models autonomously learn defect patterns from the extracted features, reducing the need for operator expertise in feature analysis.
Solution Approach 2:
The system replaces manual feature analysis and expert judgment with automated machine learning algorithms. Instead of requiring operators to interpret complex feature effects, the system uses computational models that automatically identify defect patterns from the extracted visual features.
Data Source
AI summary
One variation of a method for predicting manufacturing defects includes: accessing a first set of inspection images of a first set of assembly units recorded by an optical inspection station over a first period of time; generating a first set of vectors representing features extracted from the first set of inspection images; grouping neighboring vectors in a multi-dimensional feature space into a set of vector groups; accessing a second inspection image of a second assembly recorded by the optical inspection station at a second time succeeding the first period of time; detecting a second set of features in the second inspection image; generating a second vector representing the second set of features in the multi-dimensional feature space; and, in response to the second vector deviating from the set of vector groups by more than a threshold difference, flagging the second assembly unit.