Proc. 3 inter. conf. Computer Vision and Image Processing, CVIP 2018, Vol. 2

Preface......Page 6
Contents......Page 8
Editors and Contributors......Page 12
 1 Introduction......Page 20
  2.1 Feature Detection and Matching......Page 22
  2.3 RANSAC......Page 23
 3 Proposed Method......Page 24
 4 Results and Discussion......Page 27
 References......Page 29
 1 Introduction......Page 31
 2 Visual Cryptography for Biometric Privacy......Page 32
 3 The Proposed Architecture......Page 33
  3.1 Enrollment Phase......Page 34
  3.2 Authentication Phase......Page 37
 4 Experimental Results and Discussion......Page 38
 5 Security and Privacy Issues......Page 41
 References......Page 42
 1 Introduction......Page 44
  2.2 Image Captioning......Page 46
  3.1 Overview......Page 47
  3.2 Modules......Page 48
 4 Experimental Results......Page 52
 References......Page 54
Facial Expression Recognition Using Improved Adaptive Local Ternary Pattern......Page 56
 1 Introduction......Page 57
  2.2 Facial Feature Extraction and Dimensionality Reduction......Page 58
  2.3 Kernel Extreme Learning Machine (K-ELM) Classifier......Page 61
  3.2 Parameter Selection......Page 62
  3.3 Results on CK+ Dataset......Page 65
  3.4 Results on Jaffe Dataset......Page 66
  3.5 Results on RFD Dataset......Page 67
 References......Page 68
 1 Introduction......Page 70
  2.1 Cell Extraction......Page 71
  2.2 Text Normalization......Page 72
  3.1 Cell Extraction......Page 73
  3.2 Horizontal-Scale Correction......Page 75
  4.2 Implementation......Page 77
  4.3 Results and Discussions......Page 78
 5 Conclusion......Page 79
 References......Page 80
 1 Introduction......Page 82
 2 Model......Page 84
  2.3 Refine Units......Page 85
  3.2 Training......Page 86
  3.3 Quantitative Comparison......Page 87
 4 Conclusion......Page 88
 References......Page 89
Copy–Move Image Forgery Detection Using Gray-Tones with Texture Description......Page 91
 1 Introduction......Page 92
  2.1 Preprocessing......Page 93
  2.2 Feature Extraction......Page 94
  2.4 Similarity Detection and False Match Removal......Page 95
  3.2 Experimental Setup......Page 96
  3.3 Performance Evaluation......Page 97
  3.4 Comparative Analysis......Page 98
 4 Conclusion......Page 101
 References......Page 102
Object Labeling in 3D from Multi-view Scenes Using Gaussian–Hermite Moment-Based Depth Map......Page 103
 1 Introduction......Page 104
  2.1 Estimation of Depth Map......Page 105
  2.2 Feature Extraction......Page 107
  2.3 Object Labeling and Probability Maps......Page 108
  3.2 Setup......Page 109
  3.4 Evaluation Metrics......Page 110
  3.5 Results......Page 111
 References......Page 113
Zernike Moment and Mutual Information Based Methods for Multimodal Image Registration......Page 116
 2 Previous Methods......Page 117
  3.1 Common Features......Page 119
  3.2 Proposed Principal Component Analysis and Zernike Moment-Based Method (PCA–ZM)......Page 120
  3.3 Proposed Canny Edge Detector and Zernike Moment-Based Method (CED–ZM)......Page 122
 4 Result......Page 125
 5 Conclusion and Future Work......Page 127
 References......Page 128
 1 Introduction......Page 129
  2.1 Model Architecture......Page 130
  2.2 Loss Function......Page 132
  3.2 Noise Model......Page 133
  3.3 Experiments on Duke Dataset......Page 134
  3.4 Experiments on OPTIMA Dataset......Page 135
  3.5 Experimental Setup......Page 136
 4 Conclusion......Page 137
 References......Page 138
D2ehazing: Real-Time Dehazing in Traffic Video Analytics by Fast Dynamic Bilateral Filtering......Page 141
 1 Introduction......Page 142
 2 Detection of Vehicles in Adaptive RoI......Page 143
 3 Detailed Algorithm for Dehazing Traffic Video in Real Time......Page 144
 4 Experimental Results......Page 147
 5 Conclusion......Page 149
 References......Page 150
 1 Introduction......Page 152
 2 View Synthesis Distortion......Page 153
 4 SSIM-Based RDO with Nonlinear Depth Distortion......Page 155
 5 Bit Allocation Algorithm......Page 156
 7 Conclusions......Page 158
 References......Page 160
 1 Introduction......Page 161
  2.1 Fuzzy C-Means Clustering......Page 162
  2.2 Extensions of FCM......Page 163
 3 Proposed Methodology......Page 164
 4 Experimental Results and Analysis......Page 165
 References......Page 170
 1 Introduction......Page 172
  2.1 Word Spotting......Page 173
  2.2 Dynamic Time Warping......Page 174
 3 Features......Page 175
 4 Feature Matching Using DTW......Page 176
 5 Methodology......Page 178
 6 Dataset......Page 179
 7 Experiments......Page 180
 8 Results......Page 181
 References......Page 182
Physical Intrusion Detection System Using Stereo Video Analytics......Page 184
 1 Introduction......Page 185
 3 Our Method......Page 186
  3.1 Depth Estimation......Page 187
  3.3 Alarm Filtering and Visualization......Page 188
  4.2 Virtual Volume......Page 189
  4.3 Narrow Baseline PIDS for Near Range (20 M)......Page 190
  4.4 Wide Baseline Far Range Checker Board Experiments......Page 191
 References......Page 192
 1 Introduction......Page 194
 2 Proposed Methodology......Page 197
  2.2 Feature Extraction......Page 198
 3 Experimental Result and Discussion......Page 200
  3.1 Experimental Data Set Accumulation......Page 201
  3.2 Performance Accuracy and Comparison......Page 202
 References......Page 205
 1 Introduction......Page 207
 2 Related Works......Page 208
  3.1 Bird Parts Region Extraction Using Faster RCNN......Page 209
  3.2 Refining FRCNN Regions Using CNN Part Classifier......Page 211
  4.1 Datasets and Implementation Details......Page 212
  5.1 On CUB Dataset......Page 213
  5.2 On IIT Mandi Dataset......Page 217
 6 Conclusion......Page 220
 References......Page 222
 1 Introduction......Page 223
 2 Related Work......Page 224
 3 Context-Free Grammar......Page 227
 4 Methodology......Page 228
 5 Results......Page 232
 6 Conclusions......Page 233
 References......Page 234
L1-Regulated Featureaut]Shekar, B. H. Selection  and Classificationaut]Dagnew, Guesh of Microarray  Cancer Data Using Deep Learning......Page 236
 2 Related Works......Page 237
  3.1 L1-Regularized Feature Selection......Page 238
  3.2 Fully Connected Neural Network Method for Multiclass Microarray Cancer Data Classification......Page 240
 5 Evaluation Metrics......Page 242
 6 Experimental Results and Discussion......Page 243
 7 Conclusion......Page 248
 References......Page 250
 1 Introduction......Page 252
 2 Related Work......Page 253
  3.2 Skip-Gram in NLP......Page 254
  4.1 Image Embedding......Page 255
  4.2 Abnormality Map......Page 260
 5 Discussion and Conclusion......Page 263
 References......Page 264
Optimal Number of Seed Point Selection Algorithm of Unknown Dataset......Page 265
 1 Introduction......Page 266
  3.1 Preliminaries......Page 267
  3.2 Dunn Index......Page 268
 4 Proposed Seed Point Selection Algorithm......Page 269
  4.2 Multidimensional Entropy Estimation of Image Data......Page 270
 5 Experimental Results......Page 271
 References......Page 275
TexFusionNet: An Ensemble of Deep CNN Feature for Texture Classification......Page 278
 1 Introduction......Page 279
 2 Proposed TexFusionNet......Page 281
  3.1 Databases Used......Page 283
  3.3 Texture Classification Results......Page 285
 References......Page 288
Person Identificationaut]Kushwaha, Riti Using Footprintaut]Nain, Neeta Minutiae......Page 291
 1 Introduction......Page 292
 2 Related Work/Background......Page 293
 3 Capturing Footprint Images......Page 294
 4 Methodology to Compute Minutiae Features......Page 296
 5 Performance Evaluation......Page 302
 References......Page 304
Target Tracking Based Upon Dominant Orientation Template and Kalman Filter......Page 306
 2 Related Work......Page 307
  3.1 DOT-Based Template Matching......Page 308
  3.2 Kalman Filtering......Page 310
  3.3 Proposed Approach......Page 312
 4 Results......Page 315
 5 Conclusion......Page 316
 References......Page 317
 1 Introduction......Page 318
  1.1 Related Works......Page 319
  1.2 Our Contribution......Page 320
 2 Proposed Method......Page 321
 3 Experimental Results......Page 323
 4 Conclusion......Page 325
 References......Page 329
 1 Introduction......Page 330
 2 Related Work......Page 332
  3.2 Network Architecture......Page 333
  3.3 Implementation Detail......Page 334
  4.1 Outdoor Dataset......Page 335
  4.3 Experimental Results......Page 336
  4.4 Comparison......Page 339
 References......Page 340
Comparison Between LGBP and DCLBP for Non-frontal Emotion Recognition......Page 343
 1 Introduction......Page 344
  3.1 Feature Extraction......Page 345
  3.2 Method......Page 348
 4 Result Analysis......Page 349
 References......Page 352
 1 Introduction......Page 354
 2 Related Work......Page 355
 3 Proposed Approach......Page 356
  3.3 Overall Methodology......Page 357
  4.1 Dataset......Page 360
  4.4 Results......Page 361
 References......Page 364
Hand Down, Face Up: Innovative Mobile Attendance System Using Face Recognition Deep Learning......Page 366
 1 Introduction......Page 367
 2 Literature Review......Page 368
  3.2 Technologies Used......Page 369
  4.2 Preprocessing......Page 370
  4.4 Face Recognition......Page 371
  5.1 Database......Page 372
  5.2 Performance Analysis......Page 373
  5.3 User Interface......Page 375
 6 Conclusions......Page 376
 References......Page 378
Trajectory Classification Using Feature Selection by Genetic Algorithm......Page 379
 1 Introduction......Page 380
  2.1 Feature Extraction......Page 381
  2.2 Genetic Framework......Page 383
  2.3 Trajectory Classification Using RF and SVM......Page 384
  3.1 Datasets and Ground Truths......Page 386
  3.2 Classification Results......Page 387
 References......Page 389
Action Recognition from Egocentric Videos Using Random Walks......Page 391
 1 Introduction......Page 392
 2 Related Work......Page 393
  3.1 PHOG Feature Extraction......Page 394
  3.2 Features from the Center-Surround Model......Page 396
  3.4 Random Walk on VSG......Page 398
  4.2 Tuning of the Parameters......Page 399
  4.3 Results on ADL Dataset......Page 400
 References......Page 403
A Bagaut]Mukherjee, Anindita of Constrainedaut]Sil, Jaya Visual Wordsaut]Chowdhury, Ananda S. Model for Image Representation......Page 405
 2 Proposed Model......Page 406
  2.2 Formulation of the Constraints......Page 407
 3 Experimental Results......Page 411
  3.2 Clustering Results......Page 412
  3.3 Retrieval Results......Page 414
 References......Page 416
 1 Introduction......Page 418
 2 Related Work......Page 419
  3.1 Dataset......Page 421
  3.2 Techniques Adopted......Page 422
 4 Results and Discussion......Page 425
  4.2 CNN-LSTM Model Performance on 360-Degree Videos......Page 426
 5 Conclusion......Page 428
 References......Page 429
 1 Introduction......Page 431
  2.1 All Phase Sine Biorthogonal Transform......Page 433
  2.2 Dynamic Stochastic Resonance......Page 434
  3.1 Embedding Process......Page 435
  3.2 Authentication Casting Process......Page 436
  3.3 Extraction Process......Page 437
  4.1 Performance of Authentication Process......Page 438
  4.2 Robustness Analysis......Page 439
 References......Page 443
Multi-scale Image Fusion Scheme Based on Gray-Level Edge Maps and Adaptive Weight Maps......Page 444
  1.2 Literature Review......Page 445
  1.3 Motivation and Contribution......Page 446
  2.1 Nonlinear Anisotropic Diffusion (NLAD)......Page 447
 3 Proposed Multi-focused Image Fusion Scheme......Page 448
  3.1 Adaptive Weight Map Calculation......Page 449
  3.2 Winner-Take-All Scheme......Page 450
  3.3 Proposed Multi-focused Image Fusion Architecture......Page 451
 4 Experimental Results and Discussion......Page 452
 5 Conclusions......Page 457
 References......Page 458
 1 Introduction......Page 459
 2 Theory......Page 460
  2.2 NNLS and Lasso Methods......Page 461
 3 Results and Discussion......Page 462
 References......Page 466
 1 Introduction......Page 468
 2 Overview of Evolutionary Techniques (ETs)......Page 469
 3 Problem Formulation......Page 470
  3.1 Formulation of an Objective Function......Page 471
  3.2 Design Procedure of FIR Filter Using CSA......Page 472
 4 Results and Discussion......Page 473
 5 Implementation of Filter on Arduino......Page 474
 6 Conclusion......Page 476
 References......Page 478
 1 Introduction......Page 480
 2 Proposed Methodology......Page 482
 4 Conclusion......Page 486
 References......Page 491
Author Index......Page 493