Proc. 4 inter. conf. Computer Vision and Image Processing, CVIP 2019, Part 2

Preface......Page 6
Organization......Page 7
Contents – Part II......Page 9
Contents – Part I......Page 14
Neural Network......Page 18
 1 Introduction......Page 19
 3 Proposed Model......Page 21
 4 Experimental Setup and Results......Page 22
 5 Conclusion......Page 27
 References......Page 29
 1 Introduction......Page 31
  2.1 CNN Architectures......Page 33
 3 Multi Channel CNN Based Image Aesthetic Assessment......Page 34
  3.1 Image Pre-processing and Feature Selection......Page 35
  3.2 Proposed Multi Channel CNN for IAA......Page 36
 4 Experiments......Page 37
  4.1 Comparison with Various Approaches......Page 38
 5 Conclusions......Page 39
 References......Page 40
 1 Introduction......Page 41
 2 Proposed Work......Page 43
  2.2 Proposed Model......Page 44
  3.1 Training Phase......Page 46
  3.3 Classification Report and Confusion Matrix......Page 48
 References......Page 49
  1.1 Breast Cancer......Page 52
 2 Breast Histopatholgy......Page 54
 3 Deep Neural Network Architecture......Page 55
  4.1 Nuclei Analysis......Page 56
  4.4 Mitotic Activity Analysis......Page 57
  4.5 Other Tasks......Page 58
  5.2 Conclusion......Page 60
 References......Page 61
 1 Introduction......Page 65
 2 Proposed Approach......Page 68
 3 Experiments and Results......Page 69
 References......Page 75
 1 Introduction......Page 78
  1.1 Malware Analysis......Page 79
  1.2 Related Work......Page 80
  2.1 Convolutional Neural Networks......Page 82
 4 Experimental Results......Page 84
 References......Page 88
 1 Introduction......Page 90
  1.1 Evidence Theory......Page 92
 2 Belief Theory Based Hyper-Credal N.N. Classifier......Page 93
  2.2 Combining the Basic Belief Assignments......Page 94
 3 Belief Theory Based Hyper-Credal CNN Classifier......Page 95
 4 Results......Page 96
 References......Page 100
 1 Introduction......Page 102
 2 The Prototype Network: DDDN......Page 104
  3.1 Deep Visualization and Pruning of DDDN-B4......Page 105
  3.2 Rank Analysis Using VBMF......Page 106
  3.3 The Proposed Light DDDN Model (DDDN-L)......Page 107
  4.1 Dataset......Page 108
  4.3 Heterogeneous Computing Based Implementation......Page 109
  4.4 Results and Discussion......Page 110
 References......Page 112
 1 Introduction......Page 114
  2.2 Decision Tree (DT)......Page 115
  2.3 Artificial Neural Network (ANN)......Page 116
  3.2 Pre Processing of Dataset......Page 117
  3.3 Feature Selection Using DT......Page 118
  3.5 Classification Models......Page 119
 4 Result Analysis......Page 120
 References......Page 122
 1 Introduction......Page 123
  2.2 Training of FEN......Page 125
  2.3 Training of Classifier......Page 126
  3.2 Performance of CFMNN......Page 128
 4 Conclusion......Page 130
 References......Page 131
 1 Introduction......Page 133
 2 Anomalous Event Detection and Localization Using Stacked Autoencoder......Page 135
  2.2 Feature Extraction......Page 136
  2.3 Training of Stacked Autoencoder......Page 137
  2.4 Event Detection Model......Page 138
  3.2 Performance Evaluation Protocol......Page 139
  3.3 Results and Discussion......Page 140
 4 Conclusion......Page 143
 References......Page 144
 1 Introduction......Page 146
 2 Extended Locality Preserving Projection......Page 147
 3 Kernel-Methods for Linear Dimensionality Reduction Methods......Page 148
  4.1 Proposed Euler-ELPP......Page 149
  4.2 Proposed IGO-ELPP......Page 151
 5 Experiments......Page 153
  5.2 Face Reconstruction Results......Page 155
 References......Page 157
 1 Introduction......Page 159
 3 Design......Page 161
  3.1 Synthetic Trace Generation......Page 162
  3.3 Reinforcement Learning......Page 163
  3.4 Smart Strategy Module......Page 164
  3.5 On-Device Training......Page 165
  5.1 QoE Comparision vs Google’s Dash Player......Page 166
  5.2 QoE Comparision vs. Pensieve......Page 167
  5.3 On-Device Training......Page 168
 6 Conclusion......Page 169
 References......Page 170
 1 Introduction......Page 171
  2.1 Comparison with Style Transfer......Page 172
 3 Proposed Method......Page 173
 4 Results......Page 176
  4.1 Error Metrics......Page 177
 References......Page 178
 1 Introduction......Page 179
 2 Proposed Method......Page 181
 3 Experiments and Results......Page 182
 4 Conclusion......Page 184
 References......Page 185
 1 Introduction......Page 186
  1.1 Related Work......Page 187
 2 Proposed Deep CNN Architecture......Page 188
 3 Experiments and Results......Page 190
 4 Conclusion and Future Work......Page 193
 References......Page 194
 1 Introduction......Page 196
 2 Dataset......Page 197
  3.1 Pre-processing......Page 198
  3.3 Network Architecture......Page 199
  3.4 Experimental Setup......Page 200
 4 Experimental Results......Page 201
 5 Discussion......Page 202
 References......Page 204
 1 Introduction......Page 206
 2 Previous Work......Page 208
  3.1 Keyframes Extraction......Page 209
  3.2 Training a Convolutional Neural Network......Page 210
 4 Experiments and Results......Page 211
 References......Page 214
 1 Introduction......Page 215
  2.1 Traditional Compression Methods......Page 216
  2.2 DL Based Compression Methods......Page 217
 3 Generative Adversarial Network......Page 218
  4.1 Encoder Framework: Conventional Method......Page 219
  4.3 Reconstruction Framework: DL Based......Page 221
  5.2 Comparison with Conventional Methods......Page 223
 6 Conclusion......Page 225
 References......Page 226
 1 Introduction......Page 227
 2 Proposed Methodology......Page 228
  2.1 GAN Model......Page 229
  2.2 Training Phase......Page 230
  3.1 Environment, Dataset, and Evaluation Metrics......Page 232
  3.2 Baselines and Simulated Results......Page 234
 4 Conclusion......Page 236
 References......Page 237
 1 Introduction......Page 239
  2.2 Word Alteration Detection Using CNN......Page 242
 3 Creation of the Data Set......Page 243
 4 Experimental Results and Discussion......Page 245
 References......Page 247
 1 Introduction......Page 249
  2.1 Dataset Creation......Page 251
  2.2 Ensemble Methods......Page 253
 3 Results......Page 254
 References......Page 257
 1 Introduction......Page 259
 2 Related Work......Page 260
 3 Proposed Method......Page 262
  3.1 DeepRNNetSeg Model......Page 263
  4.1 Data Set and Experimental Setup......Page 265
  4.2 Results......Page 266
 5 Conclusion......Page 267
 References......Page 268
 1 Introduction......Page 270
 2 Related Work......Page 271
  3.1 Preprocessing......Page 272
  3.3 Feature Extraction......Page 273
 4 Experimental Results......Page 276
  4.1 Results of Preprocessing......Page 277
  4.3 Results of Density Classification......Page 278
  4.4 Comparison with the Existing Techniques......Page 279
 References......Page 280
 1 Introduction......Page 282
 2 Literature Survey......Page 283
  3.1 Convolutional LSTM (ConvLSTM)......Page 284
  3.2 U-Net......Page 285
  4.1 Framework......Page 286
  4.2 Dataset Details......Page 288
  4.3 Implementation Details......Page 290
  5.2 Results......Page 291
  5.3 Validation......Page 293
  5.4 2017 Mumbai Floods......Page 294
 References......Page 297
 1 Introduction......Page 299
 2 Related Work......Page 300
 4 Proposed Work......Page 301
  4.1 Preprocessing......Page 302
  4.2 Convolutional Neural Networks......Page 303
  4.4 Combined CNN-RNN Model......Page 304
 5 Results and Evaluation......Page 306
 References......Page 307
 1 Introduction......Page 310
  2.1 Model-Based Approach......Page 312
 3 Model Architecture......Page 313
  3.1 Encoder EZ......Page 314
  3.4 Decoder D......Page 315
  4.3 Pixel-Wise Reconstruction Loss......Page 316
  6.1 Experiment on the CLF and UTKFace Data Set......Page 317
  6.2 Qualitative Results......Page 318
  6.3 Quantitative Analysis......Page 319
 References......Page 320
 1 Introduction......Page 323
  1.1 Related Work......Page 324
 2 Description About Dataset......Page 326
 3 Convolutional Neural Network......Page 327
  3.1 Convolution Layer......Page 328
 4 Proposed Method......Page 329
  4.1 Experimental Results and Analysis......Page 331
 5 Conclusion and Future Work......Page 332
 References......Page 333
 1 Introduction......Page 335
 2 Related Work......Page 336
  3.1 Traditional Image Processing Approach......Page 337
  3.2 Deep Learning Approach......Page 339
  3.3 Face Detection Algorithm......Page 340
  3.4 Dataset Description......Page 341
  4.1 Results of Traditional Image Processing Approach......Page 342
  4.2 Results of Fine-Tuned Xception Deep Learning Model......Page 343
  4.4 Real Time Implementation......Page 344
 References......Page 345
 1 Introduction......Page 347
  2.1 Denoising and Deblurring......Page 349
  3.1 Dataset......Page 350
  3.2 Architecture......Page 351
  3.3 Objective Function......Page 352
 4 Experiments and Results......Page 353
  4.5 Naturalness Image Quality Evaluator......Page 354
 References......Page 357
 1 Introduction......Page 359
 2 Related Work......Page 360
 3 Proposed Work......Page 361
  3.1 Database Processing......Page 362
 4 Experiments......Page 363
  4.1 Dataset......Page 364
  4.2 Results and Discussion......Page 365
 References......Page 368
 1 Introduction......Page 370
 2 Related Work......Page 371
 3 Proposed Method......Page 372
 4 Experiments and Discussions......Page 374
  4.3 Results and Discussions......Page 375
 References......Page 378
 1 Introduction......Page 380
  1.1 Literature Survey......Page 381
 2 Proposed Method......Page 382
  2.1 Feature Extraction......Page 383
 4 Database Information......Page 384
 6 Conclusion......Page 385
 References......Page 386
 1 Introduction......Page 389
 2 Object Detection Approach for Localized Affected Area in MRI Images......Page 391
  2.1 An Architecture Description with Feature Mapping Approach......Page 392
  2.2 A Workflow of the Model for Detecting OA Affected Region......Page 393
 3 Result and Discussion......Page 397
 References......Page 398
Object Detection......Page 400
 1 Introduction......Page 401
 2 Problems in Current 360 Face Detectors......Page 402
  3.1 High Recall Model......Page 403
  3.3 Refinement Network: High Precision Model......Page 404
  3.4 Data Augmentation......Page 405
  4.1 Setup......Page 406
  4.2 Dataset......Page 407
  4.3 Model Analysis......Page 408
 5 Conclusion......Page 410
 References......Page 411
 1 Introduction......Page 412
 2 Literature Survey......Page 413
 3 The Proposed Method......Page 415
  3.2 Ear Localization......Page 416
  3.3 Ear Landmark Detection......Page 417
 4 Results......Page 418
 References......Page 423
Object Recognition......Page 424
 1 Introduction......Page 425
 2 Related Works......Page 426
 3 Ambulance Detection......Page 427
  4.1 Difficulties of Video Processing......Page 428
  4.2 Proposed Method......Page 429
  4.4 ResNet......Page 430
  5.1 Results......Page 431
  5.2 Analysis......Page 432
 References......Page 433
 1 Introduction......Page 435
 2 Literature Review......Page 437
  4.1 Pre-processing of Video......Page 438
  4.4 Speed Calculation......Page 439
 6 Conclusion......Page 442
 References......Page 443
 1 Introduction......Page 444
 2 Related Work......Page 445
  3.2 Image Descriptors......Page 446
  3.3 Proposed Methodology......Page 449
  3.4 Visual Concept Detection Based on Bagging Approach......Page 450
 4 Experimental Results and Analysis......Page 451
 References......Page 453
 1 Introduction......Page 455
  2.1 ID Document Photo Matching......Page 457
 3 Datasets......Page 458
  3.4 Private Profile and Frontal Face......Page 459
 4 Methodology......Page 460
  4.2 Feature Extraction......Page 461
  4.3 Verification......Page 462
  5.1 Experimental Results......Page 463
 6 Conclusions and Future Work......Page 464
 References......Page 465
Online Handwriting Recognition......Page 467
 1 Introduction......Page 468
 2 Gurmukhi Script......Page 470
 3 Data Collection and Annotation......Page 471
 4 OHWR Data Format......Page 472
 5 OHWR-Gurmukhi_1.0 Format Description......Page 473
 7 Conclusion......Page 475
 References......Page 476
Optical Character Recognition......Page 478
 1 Introduction......Page 479
 2 Literature Survey......Page 480
  3.1 Binary Classification of Segmented Blocks Using CNN ch42cnnspsmain......Page 481
  3.2 Classification of Characters......Page 482
  3.3 Capsule Network......Page 483
 4 Experimental Results......Page 486
  4.2 Dataset......Page 487
  4.3 Comparison with Other Approaches......Page 488
 References......Page 490
Security and Privacy......Page 492
 1 Introduction......Page 493
  1.1 Motivation......Page 495
  2.1 Embedding Algorithm......Page 496
 3 Proposed Method......Page 497
  3.1 Embedding Algorithm......Page 498
 4 Experiment and Results......Page 499
 5 Conclusion......Page 500
 References......Page 501
 1 Introduction......Page 503
 2 Related Work......Page 505
 3 Proposed Framework......Page 506
  3.1 Frame Anomaly Detection......Page 507
  3.2 Threat-Level Determination......Page 508
 4 Experimental Results......Page 509
  4.2 Performance Measures......Page 510
  4.3 Threat Level Analysis......Page 511
 References......Page 513
Unsupervised Clustering......Page 515
 1 Introduction......Page 516
 3 Methodology......Page 518
  3.1 Dataset Description......Page 519
  3.2 Feature Representation......Page 520
 4 Experimentation and Results......Page 521
  4.1 Calculating the Maximum Permutation Accuracy......Page 523
 5 Conclusion and Future Work......Page 524
 References......Page 525
Author Index......Page 528