AI Display Panel Age Prediction via Virtual Aging Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting the age of display panels, particularly those with light-emitting elements, face challenges in accurately determining aging characteristics over time, leading to inefficiencies in luminescence retention and overall panel performance.
Innovation Solution
An inspecting method utilizing an artificial intelligence model trained on age data from test display panels, which generates virtual age data by analyzing aging characteristics such as luminance retention, color, grayscale, temperature, and current, to predict the age of display panels with improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional aging prediction methods are used, then the inspection process is simple, but the prediction accuracy of display panel age is insufficient
Solution Approach 1:
The patent creates virtual aging data by training an AI model on real aging characteristics from test display panels. Instead of requiring extensive physical aging tests on production panels, the system copies the aging patterns learned from test panels and applies them to predict ages of production panels, significantly improving accuracy while reducing inspection complexity
Solution Approach 2:
The patent transforms multiple aging parameters (luminance retention, color shift, grayscale degradation, temperature, current) into a comprehensive age prediction model. By changing from single-parameter to multi-parameter analysis, the system achieves higher prediction accuracy without proportionally increasing system complexity
2Measurement precision
If multiple test display panels are used for training, then the prediction accuracy improves, but the inspection time and resources increase
Solution Approach 1:
The patent performs preliminary aging tests on a set of test display panels to build the AI model before actual production inspection. By pre-training the model with comprehensive aging data, the system reduces the time required for inspecting each production panel, as the heavy computational work is done in advance
Solution Approach 2:
The system copies aging characteristics from test panels to virtual models, allowing rapid prediction on production panels without requiring equivalent physical testing time. This virtual copying approach maintains high accuracy while dramatically reducing inspection time
Data Source
AI summary
An inspecting method of a display panel includes training an artificial intelligence model based on age data of a test display panel for an aging characteristic; generating virtual age data of a virtual display panel from the aging characteristic using the artificial intelligence model; and predicting an age of the display panel based on the virtual age data.