Deep Learning Techniques for Automation and Industrial Applications

Index

• Absolute layer, 111, 114–115
• Accuracy chart, 39
• Activation, 111, 115–116, 120
• Adaboost classifier, 102
• Adulterated images, 110
• ALASKA challenge, 111–112
• ALASKA-II challenge, 112
• Analytical hierarchy process (AHP), 129, 136–139, 147
• Applications in automation and industry, 171
• Artificial intelligence, 39
• Artificial intelligence machine learning (AIML models), 137, 140–141, 144, 146
• Artificial neural network (ANN), 132, 137, 140, 145
• Augmentation, 39–40
• Automobile systems emission, 181
• Average pooling, 115, 118

• Batch normalization (BN), 111, 118

• Challenges of building a classification model from scratch, 154–159
• Ciede2000, 92–93
• Classification, 101, 109–110, 112, 116
• CNN (convolutional neural network), 39–41, 91
• CO₂ emissions in sectors, 180
• Color attenuation prior, 89–90
• Compound scaling, 118
• Computer vision, 38, 109–110, 112, 115–116
• Computer vision systems, 224
• Contrast enhancement, 87–88
• Convolutions, 115, 118
• Cover, 109–111, 113–116, 119–120, 123
• COVID-19, 236

• Dark channel prior, 88–89
• Data augmentation, 39
• Data augmentation in fruit leaf classification, 163–164
• Dataset, 24
• Decision tree (DT), 129, 137–138, 140–141
• Deep learning, 37–42, 91
• Deep learning techniques, 245–246
• Depth to groundwater (DTW), 140
• Different techniques of single image dehazing, 87–91
• Digital elevation model (DEM), 139

• Electrical resistivity test (ERT), 139
• Ensemble classifier (EC), 110, 114
• Evidential belief function (EBF), 137
• Experiments and results,
  • DenseNet family, 57 ...