Preface......Page 6
Contents......Page 8
1 Introduction......Page 14
2.1 Previous Approaches......Page 15
3.1 Dataset......Page 16
3.2 Railway vs Road Signs and Signals......Page 20
4.1 Faster R-CNN......Page 21
4.2 Evaluation Method......Page 22
5 Results......Page 23
References......Page 27
1 Introduction......Page 29
2 Background......Page 30
3 Network Architecture......Page 31
4 Data......Page 33
5 Experiment and Results......Page 34
Appendix......Page 37
References......Page 39
1 Introduction......Page 40
2 Image Color Representation......Page 42
3 Transfer Learning of DL Models for Image Classification......Page 43
4 Experimental Layout......Page 44
5.2 ResNet......Page 46
6 Findings and Further Work......Page 48
References......Page 50
1 Introduction......Page 52
2 Related Work......Page 53
3 Weak Supervision by NCC......Page 54
5.1 Alignment......Page 57
5.3 Training Stability Analysis......Page 59
A.1 Training Set......Page 63
A.2 Ground Truth Annotation......Page 64
A.3 Match Filtering......Page 65
References......Page 70
Abstract......Page 72
1 Introduction......Page 73
4 The Application of Nature Inspired Algorithms in Deep Learning......Page 74
4.1 Harmony Search Algorithm for Deep Belief Network......Page 75
4.5 Ant Colony Optimization for Deep Belief Network......Page 76
4.9 Genetic Algorithm for Restricted Boltzmann Machine......Page 77
5 The General Overview of the Synergy Between Nature Inspired Algorithms and Deep Learning......Page 78
6 Challenges and Future Research Directions......Page 79
7 Conclusions......Page 80
References......Page 81
1 Introduction......Page 84
3 RBM-SVR Algorithm......Page 87
4.1 Principles of GWO Algorithm......Page 88
4.2 Principles of the DE Algorithm......Page 89
5 Experimental Setup and Results......Page 91
6 Conclusions and Outlook......Page 95
References......Page 96
1 Introduction......Page 99
2 Related Work......Page 100
3 Model......Page 102
3.1 Visual Semantic Embedding......Page 103
3.2 Discriminator......Page 104
4 Experiments and Results......Page 105
4.1 Quantitative Analysis......Page 106
4.2 Qualitative Analysis......Page 107
5 Conclusions......Page 108
References......Page 109
1 Introduction......Page 112
2.1 Inception Architecture......Page 114
3.1 Architecture......Page 115
3.2 Training Methodology......Page 116
4.1 Data Sets......Page 117
4.3 Parameter Sensitivity......Page 118
4.4 Experimental Results......Page 119
5 Conclusion......Page 120
References......Page 121
1 Introduction......Page 122
1.1 Related Work......Page 124
1.2 Single Image Super-Resolution Using GAN......Page 126
2 Proposed Method......Page 128
2.1 Loss Functions......Page 130
3 Experiment Analysis......Page 131
References......Page 137
1 Introduction......Page 141
2.1 What Is Deep Learning......Page 142
2.2 Advantages of Deep Learning......Page 143
2.3 Advantages of Traditional Computer Vision Techniques......Page 145
3.1 Mixing Hand-Crafted Approaches with DL for Better Performance......Page 146
3.2 Overcoming the Challenges of Deep Learning......Page 147
3.4 Problems Not Suited to Deep Learning......Page 148
3.5 3D Vision......Page 149
3.6 Slam......Page 151
3.8 Dataset Annotation and Augmentation......Page 152
4 Conclusion......Page 153
References......Page 154
1 Introduction......Page 158
2 Related Research......Page 159
3.1 Overview of Proposed Methods......Page 160
3.3 Generation of Training/Testing Data......Page 161
3.4 Geometrically Calculate Location......Page 162
3.5 Self-localization by CNN......Page 165
3.6 Self-localization by Convolutional LSTM......Page 166
4 Evaluation of Self-localization......Page 167
5 Conclusion......Page 170
References......Page 171
1 Introduction......Page 172
2.1 Age and Gender Classification......Page 173
2.2 Deep Neural Networks......Page 175
3.2 Age Classification Approach......Page 176
4.1 Model Architecture......Page 177
4.2 Prediction Head and Cross-Modal Learning......Page 179
4.3 Training......Page 180
4.5 Iterative Model Refinement......Page 181
5 System and Evaluation......Page 183
5.2 Results......Page 184
References......Page 186
1 Introduction......Page 191
2 Related Work......Page 192
3.1 Markov Decision Process......Page 193
3.2 Architecture......Page 197
4 Experiments......Page 198
4.2 Results......Page 199
5 Conclusion......Page 201
A Algorithm......Page 202
B Evaluation Overview......Page 203
References......Page 204
1 Introduction......Page 205
1.1 Mobile Sourced Images......Page 206
1.3 Fixed Source Images from Closed-Circuit Television (CCTV)......Page 207
2.1 Model Description......Page 208
2.2 Road Weather Condition Estimation Using CCTV......Page 209
2.3 Road Weather Condition Estimation Using Mobile Cameras......Page 210
3 Experimental Results......Page 211
3.1 Experimental Results on Naturalistic Driving Images......Page 215
References......Page 216
1 Introduction......Page 218
2 Related Work......Page 219
3 Algorithm Description......Page 220
3.1 Design of Feature Extraction Network......Page 221
3.2 Clustering Scheme......Page 223
4.2 Effectiveness Analysis of Feature Extraction......Page 224
4.3 Analysis of the Effectiveness of a Parallel Network......Page 225
4.4 Analysis of Detection Results......Page 226
References......Page 232
1 Introduction......Page 235
2 Regeneration of Subkeys Using a Secret Key......Page 237
4 The Decryption Process......Page 238
5.1 Subkeys Regeneration......Page 239
5.2 Encryption Results......Page 241
5.3 Decryption Results......Page 242
6.2 Scatter Diagrams......Page 243
6.3 Correlation Coefficients......Page 244
6.4 Parameter Mismatch......Page 245
6.5 Differential Attack Analysis......Page 246
7 Conclusions......Page 247
References......Page 248
1 Introduction......Page 250
2 Graphs and Cuts......Page 253
3 Island-Free Reconstructions......Page 254
4 Simple Reconstruction Paths......Page 255
5 Map Reconstructions for 1-Run Boundaries......Page 256
6 Island-Free Reconstructions for 2-Run Boundaries......Page 260
7 MAP Reconstructions for 2-Run Boundaries......Page 263
References......Page 268
1 Introduction......Page 270
2.1 Characteristics of the 3D Neutron Data......Page 272
2.3 DBSCAN-Assisted DVR......Page 274
3 Application on Neutron Data......Page 276
3.1 Feature Extraction in Single Crystal Diffuse Scattering......Page 278
3.2 Interactive Visualization of Neutron Tomography......Page 279
4.1 Involvement of HPC......Page 281
4.2 Expanding DBSCANβs Applications in Neutron Science......Page 282
References......Page 283
1 Introduction......Page 285
2.2 Segmentation......Page 286
3.1 Plate Characterization......Page 287
3.2 Selection of Best Binarization Technique......Page 288
3.3 License Number Segmentation......Page 289
5.1 Character Segmentation......Page 292
5.2 Classification Problems......Page 293
References......Page 295
1 Introduction......Page 297
2 Related Work......Page 298
3.1 Structure of the Human Eye......Page 300
3.2 3D Perception Considerations......Page 303
3.4 Snell Law......Page 304
3.5 Color Spaces......Page 305
3.6 Chroma Key......Page 306
4.2 Capture Module Design......Page 307
4.3 Phase of Coding of Images......Page 310
4.4 Representation Phase......Page 311
5 Experiments and Results......Page 312
References......Page 314
1.1 Background......Page 316
1.2 Related Work......Page 317
2.1 Superpixel Partitioning Using SLIC......Page 318
3.1 Overview......Page 320
3.2 Detection and Manipulation of Superpixels......Page 321
3.4 Crack Cleaning and Simplification......Page 324
4.1 Subjective Assessment......Page 325
4.2 Objective Assessment......Page 326
References......Page 328
1.1 Problem......Page 330
2.2 Previous Research......Page 331
4.1 Research Design......Page 332
5.1 Solution......Page 336
References......Page 338
1 Introduction......Page 339
2 Background......Page 340
3 Method and Approach of Scenario......Page 341
4.2 Image Processing......Page 342
5 Image Recognition Through CNN......Page 343
6.1 Function Evaluate......Page 346
6.2 Confusion Matrix......Page 347
6.3 ROC Curve......Page 348
7 Results Analysis......Page 349
8 Conclusions and Future Works......Page 350
References......Page 351
1 Introduction......Page 352
2 Related Works......Page 353
3 Datasets......Page 354
4.2 License Plate Detection......Page 355
4.4 Character Recognition......Page 356
5.2 License Plate Detection......Page 358
5.3 Recognition......Page 359
References......Page 361
1 Introduction......Page 363
2 The Motivation of Our Approach......Page 364
3.1 Objective Object Weight Map......Page 366
3.3 Segmentation Based on Iterative Graph Cut......Page 369
4 Experiment Result......Page 370
References......Page 372
1 Introduction......Page 374
2 Proposed System......Page 375
2.3 Measurement Stage......Page 376
3.2 System Performance......Page 377
4 Experimental Results......Page 378
References......Page 380
1 Introduction......Page 382
2.1 Problem Statement......Page 383
2.2 Overview......Page 384
3 Optimizations......Page 385
3.1 Element-Wise Addition Fusion......Page 386
3.3 Linear Transformation Fusion......Page 387
3.5 Padding Transformation......Page 388
4.2 Network Architectures......Page 389
4.4 Cost Efficiency......Page 391
4.5 Optimizations......Page 392
5 Related Work......Page 393
References......Page 394
1 Introduction......Page 397
2 Compressive Sensing......Page 398
3.1 The Laplacian Operator......Page 400
3.3 Gaussian Blur......Page 401
5 Results......Page 402
References......Page 406
1 Introduction......Page 407
2 Head of Hades Sample......Page 408
3 3D Reconstruction Model......Page 410
Acknowledgment......Page 413
References......Page 414
1 Introduction......Page 415
2.1 Models......Page 416
2.2 Data......Page 417
3.1 Segmentation Model......Page 419
3.2 Fine-Tuning on Expert Annotations......Page 421
4 Discussion......Page 424
References......Page 426
1 Introduction......Page 429
2.1 Denoising Quality Feature Design......Page 431
2.3 Automatic Parameter Tuning......Page 437
3.1 Image Denoising Quality Benchmark......Page 438
3.2 Regression Validation......Page 439
3.3 Evaluation on Denoising Quality Ranking......Page 440
3.4 Evaluation on Parameter Tuning......Page 441
References......Page 444
1 Introduction......Page 447
2.1 Plant Disease Database Creation......Page 448
2.2 Method......Page 449
3 Results and Discussion......Page 454
References......Page 455
1 Introduction......Page 457
2 Related Works......Page 458
3.1 Face Detection and Alignment......Page 460
3.3 COSFIRE......Page 461
3.4 Fusion Methods......Page 464
4.1 Data Sets......Page 465
4.2 Experiments......Page 466
5 Discussion......Page 467
References......Page 468
1 Introduction......Page 472
3 Methodology......Page 474
3.1 Design of GCP-Marker......Page 475
3.2 Detection of GCP......Page 476
4.1 Data Acquisition......Page 482
5 Results......Page 485
References......Page 487
1 Introduction......Page 490
3 Marker-Based Tracking During Robot-Assisted Radical Prostatectomy......Page 492
3.1 The Target Application......Page 494
3.2 The Simulation of the Surgical Operation......Page 496
4 The Optimization Procedure......Page 497
4.1 Validation......Page 500
5 Evaluation......Page 501
5.1 Considering Different Number of Markings......Page 502
5.2 Adding Deformation......Page 504
5.3 Filtering......Page 505
References......Page 507
1 Introduction......Page 510
2 Laser Scanning System......Page 511
4 3D Shape Deflection Monitoring......Page 512
References......Page 514
1 Introduction......Page 516
2 Related Work......Page 517
3.2 Template Matching......Page 520
3.3 Image Pre-processing......Page 521
3.4 Optical Character Recognition......Page 522
3.5 Text Post Processing......Page 523
4 Experimental Results......Page 524
References......Page 526
1 Introduction......Page 528
2 Literature Review......Page 530
3.2 Flowchart of the Proposed Approach......Page 533
4.1 Image Visibility and Contrast Improvement......Page 541
4.2 3D Quality Improvement......Page 542
5 Conclusion......Page 545
References......Page 546
1 Introduction......Page 548
1.2 Network Structure......Page 549
1.4 Proposed Model......Page 550
2.2 Loss Margin......Page 551
3.2 Dynamic Margin......Page 553
4.2 Model Training......Page 554
5.1 Toy Experiment......Page 556
5.2 Evaluation Metrics......Page 557
5.3 Recall-Time Data Filtering......Page 561
References......Page 562
1 Introduction......Page 564
2 Related Works......Page 566
3.1 Query to Social Media......Page 568
3.2 Image Matching Process and GPS Data Extraction......Page 569
4 Results......Page 571
References......Page 574
1 Introduction......Page 577
2 Methods......Page 579
2.2 Data Collection......Page 580
3 Preliminary Findings (nβ=β14)......Page 584
3.4 Learning/Fatigue Effects......Page 585
4 Discussion......Page 586
5 Limitations and Future Work......Page 588
AppendixβPre-study Questionnaire......Page 589
References......Page 591
1 Introduction......Page 593
2 Problem of Alignment of 2-D Shapes......Page 594
3 Related Work......Page 596
4 Material Used for This Study......Page 598
5.1 Background......Page 599
5.2 Acquisition of Object Contours from Real Images......Page 600
5.3 Approximation......Page 602
6.1 The Alignment Algorithm......Page 603
6.2 Calculation of Point Correspondences......Page 605
6.3 Calculation of the Distance Measure......Page 607
7 Evaluation of Our Alignment Algorithm......Page 608
8 Conclusions......Page 610
References......Page 611
1 Introduction......Page 613
2 Algorithm Framework......Page 614
3 PLK Optical Flow......Page 616
4 EKF......Page 619
5 Experiments......Page 620
References......Page 625
1 Introduction......Page 627
2.2 Method of Sauvola......Page 629
3.1 The Framework of Proposed Method......Page 630
3.2 Character Edge Extraction with High and Low Contrast......Page 631
3.3 Local Threshold Calculation......Page 632
3.4 Image Binarization......Page 634
4.1 Algorithm Parameters......Page 635
4.2 Algorithm Parameters......Page 636
4.4 Time Performance Analysis......Page 638
References......Page 640
1 Introduction......Page 642
2.1 Displacement Accuracy Validation......Page 644
2.2 Laboratory Testing to Obtain Multipoint Displacement of Long-Medium Span Bridge......Page 645
3 Field Testing......Page 648
4 Discussion and Conclusions......Page 649
References......Page 650
1 Introduction......Page 652
2.1 SPM......Page 653
2.4 Image Source......Page 654
3.2 Hausdorf Distance (HD)......Page 655
4 Comparison Results......Page 656
4.1 Images with Different Noise Levels......Page 657
5 Conclusions......Page 658
References......Page 659
1.1 The Challenge of Chronic Respiratory Disease......Page 661
2 Background......Page 662
3.2 Software Development......Page 664
3.3 Development of Buddi-DL, the Open Web App for LibreHealth RIS......Page 666
4 Results......Page 668
6 Conclusion......Page 669
References......Page 670
1 Introduction......Page 671
2.1 Pixel Intensity Based Face Detection Algorithm......Page 673
2.2 Image Fusion......Page 674
2.3 Optimization......Page 676
2.5 MobileNet Architecture......Page 678
3.1 Dataset......Page 679
3.3 Face Recognition Accuracy......Page 680
4 Conclusion......Page 681
References......Page 682
1 Introduction......Page 684
3.2 Model Architecture......Page 685
4.1 Raw Signals......Page 687
4.2 Logarithmic Spectrograms......Page 688
References......Page 689
1 Introduction......Page 691
2.1 Boundary Extraction......Page 692
2.3 Quadratic Function for Curve Fitting......Page 693
3 Problem Representation and Its Solutions with the Help of GA......Page 695
3.1 Initialization......Page 696
3.3 Demonstration......Page 697
References......Page 698
1 Introduction......Page 700
2.1 Differomorphism in Image Registration......Page 701
2.3 Numerical Algorithm of LDDMM......Page 702
3 Fast Sequential Diffeomorphic Atlas Building (FSDAB)......Page 703
3.1 Numerical Algorithm of FSDAB......Page 704
4 Fast Bayesian Principal Geodesic Analysis (FBPGA)......Page 705
5.2 3D Brain Dataset......Page 707
7 Conclusion......Page 710
References......Page 711
1 Introduction......Page 713
2 Literature Review......Page 715
3.1 Linear Cellular Automata Transform......Page 717
3.2 Pre-processing of Data for 2D Vector Maps......Page 719
3.3 Computation of Relative Coordinates......Page 720
3.4 Binary Transform for Cover Data......Page 721
3.5 The Degree of Watermarking......Page 722
4 The Suggested Strategy of Reversible Watermarking......Page 723
4.2 Embedding Algorithm......Page 724
4.3 Extraction Algorithm......Page 725
5.1 Results of Experiments......Page 726
5.2 Robustness Assessment......Page 728
5.3 Assessment of the Capability for Content Authentication......Page 729
References......Page 730
1 Introduction......Page 733
2 Literature Review......Page 734
3 Proposed Methodology......Page 735
4.2 Results......Page 736
4.3 Discussion......Page 737
5 Conclusion......Page 740
References......Page 741
1 Introduction......Page 742
2.1 TPACK and Higher Education......Page 743
3 Study Context......Page 744
3.1 Possible ICT Technologies Used to Carry Out the Proposed Activities......Page 745
4.1 Participants......Page 748
4.3 User Acceptance Questionnaire for TPACK Structure......Page 749
4.4 Focus Group Interview......Page 750
5.1 Results About the Need for Technological Device Usage by University Learners......Page 751
5.2 User Acceptance Results of TPACK Model in Accordance with University Learners Need......Page 753
6.1 User Acceptance and Future of TPACK Approach......Page 754
6.2 Concerns About βTPACK Structure According to University Learnersβ Needβ Approach......Page 756
8 Conclusion......Page 757
References......Page 758
1 Introduction......Page 760
2.1 Online Restaurant Reviews......Page 762
2.2 Dining Experience......Page 763
2.3 Hallyu......Page 764
3.1 Data Collection......Page 765
3.2 Analysis Method......Page 766
4.2 Practical Contributions......Page 768
References......Page 769
Abstract......Page 771
2 Background......Page 772
2.2 Principles of Linked Data......Page 773
2.3 Linked Data Publishing Methodology......Page 774
2.4 Tools......Page 775
2.7 Linked Data in Higher Education......Page 776
3.1 Model of Bodies of Knowledge (BOK)......Page 777
3.3 Academic Offer Domain......Page 778
3.4 Design of HTTP URIs......Page 780
3.5 RDF Generation......Page 781
3.6 Publication and Exploitation of Data......Page 782
4 Conclusions......Page 784
References......Page 785
1.1 Context......Page 787
2 State of the Art......Page 788
3.1 Corpora......Page 791
3.2 Method and Algorithms......Page 792
3.3 Implemented Method System Architecture......Page 794
3.4 Model and Classification Training......Page 796
4 Solution Re-usability and Core Capabilities in a Nutshell......Page 797
5 Conclusions and Future Work......Page 799
References......Page 800
1 Introduction......Page 802
2.3 Modeling Preferences......Page 803
3.2 Selection Process......Page 804
5.1 Linguistic Variables......Page 805
5.2 Choice of the Set of Linguistic Terms......Page 806
6 Methodologies in the Development of New Products......Page 807
6.2 New Products Development Models......Page 808
6.3 Variables and Elements in the Analysis and Design of a New Product Development Model......Page 809
References......Page 810
1 Introduction......Page 812
2 Analysis of Related Work......Page 813
3.1 Scientific Data Sources......Page 817
4 Specification of Ontology Requirements......Page 818
5 Ontology Design......Page 819
5.2 Publication Class......Page 820
5.3 Researcher Profile Ontology......Page 821
6.2 Publication Class Population......Page 822
7 Ontology Enrichment......Page 823
8.1 Researcher Publications......Page 825
8.3 Qualified and Specialized Researchers......Page 826
8.4 Publications by Year......Page 827
8.6 Publications by Gender......Page 828
9 Conclusions......Page 829
References......Page 830
Author Index......Page 831