Display Panel Defect Inspection with Multi-Filter Gray-Level Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic optical inspection systems for display panels struggle with misjudgment of defects due to similar optical characteristics between dust and panel defects, leading to false detections or missed inspections, especially for color defects with similar gray levels, affecting production efficiency and requiring manual re-judgment.
Innovation Solution
A defect inspection system that captures multiple monochrome images of display panels through filters of different colors, using an analysis unit to determine defect types based on brightness ratios and change rates between gray levels of these images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If inspection parameters are set more strictly to reduce false detections, then false detection rate decreases, but missed inspection rate increases
Solution Approach 1:
The patent segments the inspection process by dividing defects into different categories (in-panel defects vs. dust particles) with distinct characteristic analysis methods. In-panel defects are identified by their location within pixel boundaries and specific optical characteristics, while dust particles are identified by their location on pixel boundaries and different optical properties. This segmentation allows the system to apply appropriate detection criteria for each type, reducing both false detections and missed inspections.
Solution Approach 2:
The patent changes multiple inspection parameters simultaneously: inspection criteria (strict vs. loose based on defect type), gray level thresholds (different thresholds for different defect types), and detection sensitivity. By dynamically adjusting these parameters based on the specific defect category being inspected, the system achieves high detection accuracy without increasing missed inspection rates.
2Productivity
If single gray level inspection is used to maintain fast inspection speed, then productivity is high, but color defects with similar gray levels cannot be detected
Solution Approach 1:
The patent implements a dynamic inspection approach where the system first performs a rapid single gray level inspection to identify candidate defect areas, then dynamically switches to multi-gray level inspection only for those specific areas where color defects are suspected. This dynamic adjustment of inspection depth maintains high overall productivity while ensuring color defects are not missed.
Solution Approach 2:
The patent performs preliminary filtering using single gray level inspection to identify potential defect locations before conducting more thorough multi-gray level analysis. This preliminary action eliminates the need for full multi-gray level inspection across the entire panel, maintaining speed while enabling color defect detection when necessary.
3Measurement precision
If multi-filter monochrome images are captured to improve defect recognition accuracy, then defect detection precision improves, but inspection complexity increases
Solution Approach 1:
The patent segments the image processing workflow into distinct stages: capture of multiple monochrome images through different filters, separate analysis of each filter's image data, and integrated defect classification. By segmenting the processing of different filter images and analyzing them independently according to their specific characteristics, the system manages complexity while maintaining high recognition accuracy.
Solution Approach 2:
The patent makes the inspection system multi-functional by enabling it to detect and classify multiple defect types (in-panel defects, dust particles, color defects) using a unified multi-filter imaging platform. The same hardware system performs diverse inspection functions by analyzing different aspects of the captured images, avoiding the need for separate specialized systems for each defect type.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves defect recognition accuracy by reducing false positives and negatives, enhancing the system's ability to detect various types of defects, including color defects, thereby improving production efficiency.
Implementation Method 1
an imaging unit adapted to capture a plurality of monochrome images of at least one gray level image of a display panel through a plurality of filters, wherein filter colors of the filters differ from each other
Data Source
AI summary
A defect inspection system including an imaging unit and an analysis unit is provided. The imaging unit is adapted to capture a plurality of monochrome images of at least one gray level image of a display panel through a plurality of filters. Filter colors of the filters differ from each other. The analysis unit is adapted to determine a type of defect of the display panel in an abnormal area according to a brightness ratio or a brightness change rate between two gray levels of two of the monochrome images corresponding to the abnormal area of the display panel.