Streak Defect Severity Classification From Intensity Peak and Variation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting streak-shaped defects in inkjet printing fail to accurately classify the defects based on their severity, leading to inadequate quality control and maintenance measures.
Innovation Solution
A method involving printing a test chart, capturing an image, calculating average intensity values, extracting local maximum values and variations, and using a learned learning model to classify streak-shaped defects based on these features, mimicking human visual perception.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional threshold-based detection methods are used, then streak-shaped defects can be detected, but the defects cannot be classified by severity degree
Solution Approach 1:
The patent segments the defect detection process into multiple stages: initial defect detection using threshold methods, followed by classification into three severity levels (large, medium, small) based on additional analysis of the captured image data. This segmentation allows both detection and classification functions to be performed systematically.
Solution Approach 2:
The patent adds a classification dimension to the existing detection process. Instead of merely detecting defects with binary yes/no results, the system evaluates multiple characteristics (such as defect area, intensity, and pattern) to classify defects into three severity dimensions, thereby preserving and organizing defect severity information.
2Measurement precision
If manual visual inspection is used to classify defect severity, then accurate classification can be achieved, but automation and efficiency are reduced
Solution Approach 1:
The system incorporates feedback mechanisms where the captured image data is analyzed through multiple processing stages. The classification results are generated based on feedback from intermediate analysis steps, allowing the automated system to iteratively refine its classification decisions to match human visual inspection accuracy.
Solution Approach 2:
The patent replaces manual visual inspection (mechanical human process) with an automated image processing system that uses algorithmic analysis of captured image data. The system substitutes human eyes and brain processing with computational algorithms that analyze image characteristics to classify defect severity automatically.
3Productivity
If no defect classification is performed, then processing time is reduced, but quality control and maintenance planning are inadequate
Solution Approach 1:
The patent performs preliminary classification of defects into severity levels during the initial inspection phase. This preliminary action allows the system to quickly categorize defects before any maintenance or corrective actions are planned, enabling efficient quality control without significant time overhead.
Solution Approach 2:
The system changes the analysis parameters dynamically based on the inspection stage. During initial detection, it uses parameters optimized for speed (simple threshold checks). When classification is required, it switches to parameters optimized for accuracy (multi-dimensional image analysis), thereby balancing processing time and quality control effectiveness.
Data Source
AI summary
By causing a learning model to learn a relationship between feature values and classification destinations using at least a peak average intensity value and a variation in average intensity value around a streak-shaped defect as the feature values, a classification model for classifying the streak-shaped defect is generated. Thereafter, when the streak-shaped defect is detected, at least the peak average intensity value and the variation are obtained as the feature amounts representing features of the streak-shaped defect. Then, by inputting the feature amounts to the classification model, the classification destination depending on the features of the detected streak-shaped defect is decided.