AI Defect Inspection System for Automated Visual Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing precision requirements in electronic component manufacturing lead to a need for more efficient defect inspection processes, which are currently labor-intensive and require manual re-inspection, making them costly and inefficient.
Innovation Solution
A defect inspection system utilizing artificial intelligence that includes a re-inspection server, training terminal, classification terminal, and re-inspection terminal to automatically classify defect images, reducing the need for manual verification and minimizing manpower costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual re-inspection is used to ensure high precision defect detection, then inspection accuracy is improved, but labor costs and inspection time increase significantly
Solution Approach 1:
The inspection process is segmented into two stages: optical inspection for initial defect detection and AI-based re-inspection for automated verification. This segmentation allows each stage to specialize in specific tasks, improving overall efficiency while maintaining accuracy through the AI's focus on verifying optical inspection results
Solution Approach 2:
The patent replaces manual mechanical inspection with an AI-based automated inspection system. The AI model learns from labeled defect data and automatically performs defect classification and verification, substituting human inspectors with an intelligent system that operates faster and consistently without fatigue
2Reliability
If more manpower is deployed for manual defect verification, then inspection accuracy is improved, but labor costs increase
Solution Approach 1:
The AI-based re-inspection system performs self-learning and self-improvement by continuously processing labeled defect data. The system automatically updates its defect recognition capabilities without requiring additional human inspectors, achieving improved reliability through autonomous learning rather than increased manpower
Solution Approach 2:
The system implements a feedback mechanism where AI classification results are fed back into the training process. Labeled defect images from the re-inspection stage are used to retrain and improve the AI model, creating a continuous improvement loop that enhances reliability without requiring proportional increases in human resources
3Ease of manufacture
If traditional optical inspection is used, then initial defect detection is achieved, but follow-up manual re-inspection is required increasing process complexity
Solution Approach 1:
The patent merges optical inspection and AI-based re-inspection into a unified automated inspection system. The two inspection methods are combined in sequence, with the AI system automatically receiving and processing outputs from the optical inspection stage, eliminating the need for separate manual re-inspection processes and reducing overall process complexity
Data Source
AI summary
A defect inspection system, connected to an automatic visual inspection device, is provided, including the followings. A re-inspection server (VRS) receives a defect image and a defect location. A training terminal stores trained modules. A classification terminal receives the defect image and the defect location, reads a target trained module corresponding to the defect image, classifies the defect image according to the target trained module to obtain a labeled defect image, and sends the labeled defect image to the VRS. A re-inspection terminal receives the labeled defect image from the VRS, and sends a verified operation corresponding to the labeled defect image to the VRS. A labeling re-inspection terminal receives the verified operation and the labeled defect image, and a labeling result corresponding to the labeled defect image. The VRS sends the labeling result and the labeled defect image to the training terminal to train a corresponding training module.