Bad Picture Defect Grading by Size Ratio and Neural Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying defects in display panel production are inefficient, time-consuming, and prone to errors, leading to missed judgments and increased production costs due to manual inspection of bad pictures with substantial impact on product yield.
Innovation Solution
An automated method and apparatus using a neural network model to determine defect grades by measuring defect size and design size ratios, applying specific magnification ranges for different defect types, and labeling defects based on their impact on yield.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual inspection of bad pictures is used, then flexibility in judgment is maintained, but inspection speed is slow and efficiency is low
Solution Approach 1:
The patent replaces manual inspection with an automated system that uses neural network models to classify defect grades. The system automatically processes bad pictures through image recognition algorithms, eliminating the need for human inspectors to manually evaluate each defect, thereby significantly improving inspection speed and efficiency.
Solution Approach 2:
The inspection system performs self-service by automatically evaluating and classifying defect grades without human intervention. The neural network model independently analyzes bad pictures, determines defect severity, and provides inspection results, making the system autonomous and efficient.
2Loss of time
If manual picture judgment is used, then judgment flexibility is maintained, but time consumption is high
Solution Approach 1:
The patent substitutes manual judgment with an automated neural network-based system that rapidly analyzes bad pictures. The system processes images through multiple classification models that can evaluate defect grades in seconds, dramatically reducing time consumption compared to manual inspection.
Solution Approach 2:
The system performs preliminary classification of defect grades automatically before any manual review is needed. By pre-sorting and pre-evaluating bad pictures through automated algorithms, the system identifies the most critical defects quickly, reducing overall time consumption and improving productivity.
3Reliability
If manual inspection is used, then comprehensive review is possible, but error rate increases due to human limitations
Solution Approach 1:
The patent replaces manual inspection with an automated neural network system that provides consistent and reliable defect grading. The system eliminates human errors such as fatigue, bias, and subjective judgment by using objective algorithm-based classification, thereby improving reliability while maintaining fast inspection speeds.
Solution Approach 2:
The system incorporates feedback mechanisms where the neural network models continuously learn from inspection results and can be refined through validation processes. This feedback loop ensures that the system maintains high accuracy in defect grading while operating at rapid speeds, improving both reliability and efficiency.
4Productivity
If automated defect classification is implemented, then inspection efficiency is improved, but system complexity increases
Solution Approach 1:
The patent segments the defect classification task into multiple specialized neural network models, each designed to detect specific types of defects or grade ranges. This segmentation allows the system to handle complex inspection tasks through modular components, improving efficiency while managing complexity through structured organization.
Solution Approach 2:
The system employs a universal neural network framework that can classify multiple types of defects and grade them according to different criteria. This multi-functional approach allows a single system to handle various inspection scenarios, improving efficiency without proportionally increasing complexity through standardized processing architecture.
Data Source
AI summary
Provided are a method and apparatus for identifying the defect grade of a bad picture, and a storage medium. The method includes: determining the defect size of a defect from a bad picture; according to a product model corresponding to the bad picture, determining the design size of a pattern corresponding to a component that is adjacent to the position of the defect; determining the defect grade of the defect according to the defect type of the defect and a magnitude relationship between the defect size and the design size, wherein the defect grade is the degree to which the defect affects product yield.