A deep learning framework based on improved self-supervised learning for ground-penetrating radar tunnel lining inspection

Mathematics

• Ground Penetrating Radar 100%
• Supervised Learning 76%
• Tunnel 70%
• Inspection 59%
• Learning 48%
• Framework 30%
• Mask 30%
• Backbone 22%
• Neural Networks 22%
• Convolution 19%
• Reinforcement 15%
• Voids 14%
• Transfer Learning 11%
• Nondestructive Testing 8%
• Feature Extraction 8%
• Work Flow 8%
• Architecture 6%
• Training 6%
• Experimental Results 5%
• Defects 5%
• Decrease 5%
• Adjacent 4%
• Trace 4%
• Projection 4%

Engineering & Materials Science

• Tunnel linings 70%
• Supervised learning 56%
• Deep learning 47%
• Radar 43%
• Inspection 40%
• Convolution 23%
• Masks 22%
• Neural networks 14%
• Labeled data 13%
• Linings 12%
• Reinforcement 9%
• Transfer learning 7%
• Nondestructive examination 6%
• Feature extraction 4%
• Defects 4%