Optical Inspection Anomaly Detection for Assembly Unit Defects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current optical inspection methods lack efficiency in predicting defects in assembly units in real-time and identifying manufacturing drift, leading to potential yield reduction and waste, as they rely on manual inspection and lack advanced anomaly detection capabilities.
Innovation Solution
A method that utilizes multi-dimensional feature vectors from inspection images to group similar assembly units, detect defects, and predict anomalies by comparing new images to a database of past outcomes, allowing for real-time identification of defective units and trends in manufacturing processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual inspection methods are used, then device complexity is reduced, but productivity and measurement precision deteriorate
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated optical inspection system that captures images and uses machine learning algorithms to detect defects. This substitution of mechanical/manual processes with automated optical and computational systems directly resolves the contradiction by dramatically increasing inspection speed while accepting the necessary increase in system complexity.
Solution Approach 2:
The inspection system performs self-service through automated defect detection and classification without requiring manual intervention. The machine learning model automatically analyzes inspection images, identifies defects, and flags problematic assembly units, enabling the system to serve itself and achieve high productivity.
2Manufacturing precision
If real-time defect detection is implemented, then manufacturing precision is improved, but loss of time increases due to processing requirements
Solution Approach 1:
The system performs preliminary defect detection during the assembly process itself, capturing images at intermediate stages rather than waiting for final assembly. This allows defects to be identified early when they can be more easily detected and when corrective action can be taken immediately, improving manufacturing precision without significant time loss.
Solution Approach 2:
The inspection process operates continuously alongside production, with images being captured and analyzed in real-time as assembly units move through the manufacturing line. This continuous operation ensures that defect detection does not interrupt production flow, maintaining both high manufacturing precision and minimal time loss.
3Reliability
If comprehensive inspection of all assembly units is performed, then reliability is improved, but productivity deteriorates due to inspection time
Solution Approach 1:
The machine learning model focuses inspection efforts on specific regions and features most likely to contain defects, rather than uniformly analyzing entire assembly units. By concentrating computational resources on critical areas identified through training data, the system achieves high reliability in defect detection while maintaining high throughput by avoiding unnecessary analysis of defect-free regions.
Solution Approach 2:
The system performs inspection on all assembly units (excessive action) but uses the machine learning model to quickly identify and flag only those units with potential defects. This approach ensures comprehensive coverage for reliability while the intelligent filtering maintains productivity by minimizing follow-up inspection time for clearly defective units.
4Measurement precision
If advanced anomaly detection capabilities are added, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces a machine learning model as an intermediary between the optical inspection system and the defect identification process. This intermediary layer processes inspection images and provides structured defect classifications, enabling advanced anomaly detection capabilities while managing system complexity through modular architecture where the ML model can be trained and updated independently.
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.