Automatic Industry PCB Board DIP Process Defect Detection System Based on Deep Ensemble Self-Adaption Method

Keyphrases

• Systems-based 100%
• Printed Circuit Board 100%
• Self-adaption 100%
• Process Defects 100%
• Deep Ensemble 100%
• Defect Detection System 100%
• Detection Performance 66%
• Detection Rate 66%
• Inspection Interval 66%
• False Alarm Rate 66%
• YOLOv2 66%
• ResNet101 66%
• Environmental Variability 33%
• Repair Time 33%
• Real Tests 33%
• Self-service 33%
• Semi-automatic 33%
• Artificial Intelligence Models 33%
• Convolutional Neural Network Model 33%
• Automated Optical Inspection 33%
• Efficiency Boost 33%
• Manual Labeling 33%
• Data Collection Methods 33%
• Feature Pyramid Network 33%
• Foreground Detector 33%
• Soldering Defects 33%
• Faster R-CNN 33%
• Ensemble Convolutional Neural Networks 33%
• Professional Guidance 33%

Engineering

• Defect Detection 100%
• Production Line 100%
• Printed Circuit Board 100%
• Detection Performance 66%
• False Alarm Rate 66%
• Network Model 33%
• Repair Time 33%
• Automated Optical Inspection 33%
• Convolutional Neural Network 33%