Automated Image Processing for Display Panel Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual image processing in the manufacturing of display panels is time-consuming, inconsistent, and prone to errors due to operator reliance, leading to low efficiency and accuracy in identifying defects.
Innovation Solution
An automated image processing method that screens target images based on non-periodic pixel information, determines feature values, and calculates similarities between images to categorize them into sets, improving efficiency and accuracy by filtering out interference fringes and using specific similarity formulas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual processing is used to process image data, then the processing can be performed with simple equipment, but the processing efficiency is low and the processing accuracy is inconsistent
Solution Approach 1:
The system performs self-service by automatically processing images through neural network algorithms without requiring manual operator intervention. The automated processing includes receiving images, determining target areas with non-periodic pixel information, calculating feature values, and identifying defects, thereby improving efficiency while maintaining consistent accuracy through standardized computational procedures
Solution Approach 2:
The patent replaces manual mechanical processing with automated computational processing. Instead of operators manually examining and analyzing images, the system uses computer-based neural network algorithms to automatically detect target areas, calculate feature values, and identify defects, thereby eliminating human factors that cause inconsistency while improving processing speed
2Measurement precision
If manual processing is used to identify defects in images, then the processing cost is low, but the processing accuracy is low and consistency is poor
Solution Approach 1:
The system performs preliminary action by pre-processing images to determine target areas with non-periodic pixel information before conducting detailed defect analysis. This preliminary step of identifying regions of interest based on pixel periodicity patterns prepares the data for more accurate and efficient defect detection, reducing both time and improving accuracy
Solution Approach 2:
The patent applies parameter changes by transforming image data into feature values through neural network algorithms. The system changes the representation of image data from raw pixel information to extracted features such as target area coordinates, pixel information characteristics, and defect probability scores, thereby improving measurement precision while maintaining efficient processing
3Productivity
If automated processing with neural network algorithms is used, then processing efficiency and accuracy are improved, but the requirement for high-quality training images increases
Solution Approach 1:
The system applies segmentation by dividing the image processing task into distinct stages: receiving images, determining target areas with non-periodic pixel information, calculating feature values, and identifying defects. This segmentation allows the neural network to focus on specific aspects of image analysis separately, improving overall efficiency while maintaining high accuracy through specialized processing at each stage
Data Source
AI summary
An image processing method includes: receiving a processing instruction for an image; screening out a plurality of target images from a plurality of acquired images; determining a first feature value and a second feature value of the target area of each of the target images; determining a similarity between the first feature values of the target areas in every two target images according to all the determined first feature values, and determining a similarity between the second feature values of the target areas in every two target images according to all the determined second feature values; and, if a first target similarity is within a first preset range, and a second target similarity is within a second preset range, determining that a first image and a second image belong to the same image set.