Defect Classification Condition Feedback for Visual Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for setting classification conditions for defect types in visual inspection are inefficient and prone to errors, especially when the number of defects increases, making it difficult to confirm the appropriateness of the classification conditions.
Innovation Solution
A classification condition setting support apparatus that includes a basic information storage unit, classification condition setting unit, basic defect type classification unit, and display unit to generate a classification result confirmation screen, providing visual feedback on the appropriateness of classification conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of defect occurrences increases, then the classification work becomes more comprehensive, but the time required for associating defect types increases and errors easily occur
Solution Approach 1:
The system automatically classifies defects by comparing imaging information against stored classification conditions and defect type databases, eliminating the need for manual association work. The classification unit processes defects autonomously using predefined rules, thereby reducing both time consumption and human error while maintaining comprehensive classification coverage.
Solution Approach 2:
Manual mechanical classification work is replaced by an automated information processing system. The classification unit uses computational algorithms to compare imaging data with defect databases, substituting human operators with an automated system that processes classifications faster and more accurately.
2Adaptability or versatility
If the number of defect types to be classified increases, then the classification coverage is improved, but the number of feature amounts constituting classification conditions increases making analysis work difficult
Solution Approach 1:
The classification system divides the complex classification task into manageable segments by organizing defect types hierarchically and assigning specific feature amounts to each defect category. This segmentation allows the system to handle multiple defect types without requiring analysis of all feature amounts for every defect, thereby reducing complexity while maintaining comprehensive coverage.
Solution Approach 2:
The system dynamically selects and adjusts which feature amounts are applied based on the specific defect type being classified. Rather than using all possible feature amounts for every classification, the system changes parameters adaptively, applying only the relevant feature sets needed for each defect category, thus reducing overall complexity.
3Ease of operation
If manual classification work is performed, then flexibility in adjusting classification conditions is maintained, but the work becomes difficult and error-prone when defect numbers increase
Solution Approach 1:
The system incorporates feedback mechanisms where classification results are continuously evaluated and used to refine classification conditions. The display unit presents classification outcomes that can be reviewed, and the system allows iterative adjustment of classification conditions based on observed performance, maintaining flexibility while improving reliability through data-driven refinements.
Data Source
AI summary
Provided is a classification condition setting support apparatus including: a basic information storage unit configured to store basic information including basic imaging information and a basic defect type; a classification condition setting unit; a basic defect type classification unit configured to classify the basic imaging information according to the classification condition; a classification result confirmation screen generator configured to generate a classification result confirmation screen including the number of pieces of classification basic imaging information, the basic defect type associated with the classification basic imaging information, and the number of pieces of correct answer basic imaging information, by classifying the target basic imaging information according to the classification condition; and a display unit.