Next Generation Artificial Vision Systems: Reverse Engineering the Human Visual System (Artech House Series Bioinformatics & Biomedical Imaging)

Next Generation Artificial Vision Systems......Page 2
Contents......Page 6
Preface......Page 14
1.1 Introduction......Page 16
1.2 Overview of the Human Visual System......Page 17
1.2.1 The Human Eye......Page 18
1.2.2 Lateral Geniculate Nucleus (LGN)......Page 25
1.2.3 The V1 Region of the Visual Cortex......Page 27
1.3 Conclusions......Page 30
References......Page 32
P A R T I The Physiology and Psychology of Vision......Page 34
2.2 Retinal Anatomy......Page 36
2.3 Retinal Physiology......Page 40
2.4 Mathematical Modeling----Single Cells of the Retina......Page 42
2.5 Mathematical Modeling----The Retina and Its Functions......Page 43
2.6 A Flexible, Dynamical Model of Retinal Function......Page 45
2.6.1 Foveal Structure......Page 46
2.6.2 Differential Equations......Page 47
2.6.3 Color Mechanisms......Page 49
2.6.4 Foveal Image Representation......Page 51
2.6.5 Modeling Retinal Motion......Page 52
2.7.1 Parameters and Visual Stimuli......Page 53
2.7.2 Temporal Characteristics......Page 54
2.7.3 Spatial Characteristics......Page 56
2.7.4 Color Characteristics......Page 58
2.8 Conclusions......Page 60
References......Page 61
3.1 Introduction......Page 66
3.2.1 Single-Neuron Responses......Page 67
3.2.2 Organization of Individual Cells in V1......Page 68
3.3 Computational Understanding of the Feed Forward V1......Page 73
3.3.1 V1 Cell Interactions and Global Computation......Page 74
3.3.2 Theory and Model of Intracortical Interactions in V1......Page 76
3.4 Conclusions......Page 77
References......Page 78
4.1 Introduction......Page 84
4.2 Materials and Methods......Page 88
4.3 Results......Page 90
4.3.1 Interference by Task-Irrelevant Features......Page 91
4.3.2 The Color-Orientation Asymmetry in Interference......Page 96
4.3.3 Advantage for Color-Orientation Double Feature but Not Orientation-Orientation Double Feature......Page 99
4.3.4 Emergent Grouping of Orientation Features by Spatial Configurations......Page 102
4.4 Discussion......Page 107
4.5 Conclusions......Page 113
References......Page 114
P A R T II The Mathematics of Vision......Page 118
5.1.1 Wavelets......Page 120
5.1.3 Wavelet Choices......Page 122
5.1.4 Linear vs Nonlinear Mappings......Page 127
5.2.1 Design Overview......Page 128
5.2.2 Filter Designs: Radial Frequency......Page 129
5.2.3 Angular Frequency Response......Page 131
5.2.4 Filter Kernels......Page 133
5.3.1 Overview......Page 135
5.3.2 Generating Orientation Maps......Page 136
5.3.4 Phase Estimation......Page 138
5.4 Inference from V1-Like Representations......Page 139
5.4.1 Vector Image Fields......Page 140
5.4.2 Formulation of Detection......Page 141
5.4.3 Sampling of (B,X)......Page 142
5.4.4 The Notion of ‘‘Expected’’ Vector Fields......Page 143
5.4.6 Vector Model Plausibility and Extension......Page 144
5.4.7 Vector Fields: A Variable Contrast Model......Page 145
5.4.8 Plausibility by Demonstration......Page 146
5.4.9 Plausibility from Real Image Data......Page 147
5.4.10 Divisive Normalization......Page 148
5.5.1 Circle-and-Square Discrimination Test......Page 150
5.6.1 Keypoint Detection Using DTCWT......Page 153
5.7 Summary and Conclusions......Page 155
References......Page 156
6.2 Linear Image Processing......Page 160
6.2.1 Interpolation of Irregularly Sampled Data......Page 161
6.2.2 DFT from Irregularly Sampled Data......Page 171
6.3 Nonlinear Image Processing......Page 172
6.3.1 V1-Inspired Edge Detection......Page 173
6.3.2 Beyond the Conventional Data Representations and Object Descriptors......Page 177
6.4 Reverse Engineering Some Aspect of the Human Visual System......Page 182
6.5 Conclusions......Page 183
References......Page 184
7.1 Introduction......Page 186
7.2 Hyperacuity and Super-Resolution......Page 187
7.3 Super-Resolution Image Reconstruction Methods......Page 188
7.3.1 Constrained Least Squares Approach......Page 189
7.3.2 Projection onto Convex Sets......Page 192
7.3.4 Markov Random Field Prior......Page 195
7.3.6 Image Registration......Page 198
7.4.1 Application in Minimally Invasive Surgery......Page 199
References......Page 203
8.1 Introduction......Page 206
8.2 Eye-Tracking Techniques......Page 207
8.3.1 Psychology/Psychiatry and Cognitive Sciences......Page 210
8.3.2 Behavior Analysis......Page 211
8.3.3 Medicine......Page 212
8.3.4 Human--Computer Interaction......Page 214
8.4 Gaze-Contingent Control for Robotic Surgery......Page 215
8.4.1 Ocular Vergence for Depth Recovery......Page 217
8.4.2 Binocular Eye-Tracking Calibration......Page 219
8.4.3 Depth Recovery and Motion Stabilization......Page 221
8.5 Discussion and Conclusions......Page 224
References......Page 225
9.1 Introduction......Page 232
9.2 Motion Processing in the Human Visual System......Page 233
9.3 Motion Detection......Page 234
9.3.1 Temporal Edge Detection......Page 236
9.3.2 Wavelet Decomposition......Page 239
9.3.3 The Spatiotemporal Haar Wavelet......Page 240
9.4 Dual-Channel Tracking Paradigm......Page 245
9.4.1 Appearance Model......Page 246
9.4.2 Early Approaches to Prediction......Page 247
9.4.3 Tracking by Blob Sorting......Page 248
9.5 Behavior Recognition and Understanding......Page 252
9.6 A Theory of Tracking......Page 254
9.7 Concluding Remarks......Page 256
References......Page 257
P A R T III Hardware Technologies for Vision......Page 264
10.1 Introduction......Page 266
10.2.1 The Physiology of the Eye......Page 268
10.2.2 Phototransduction Cascade......Page 270
10.2.3 Light Adaptation of Photoreceptors: Weber-Fechner’s Law......Page 273
10.3 Phototransduction in Silicon......Page 275
10.3.1 CCD Photodetector Arrays......Page 277
10.3.2 CMOS Photodetector Arrays......Page 278
10.3.3 Color Filtering......Page 280
10.3.4 Scaling Considerations......Page 283
10.4 Phototransduction with Organic Semiconductor Devices......Page 284
10.4.1 Principles of Organic Semiconductors......Page 285
10.4.2 Organic Photodetection......Page 286
10.4.3 Organic Photodiode Structure......Page 288
10.4.4 Organic Photodiode Electronic Characteristics......Page 289
10.4.5 Fabrication......Page 296
10.5 Conclusions......Page 300
References......Page 301
11.1 Introduction......Page 304
11.2 Principles of Analog Processing......Page 305
11.2.1 The Metal Oxide Semiconductor Field Effect Transistor......Page 307
11.3 Photo Electric Transduction......Page 311
11.3.1 Logarithmic Sensors......Page 312
11.3.3 Integration-Based Photodetection Circuits......Page 313
11.4 Retinimorphic Circuit Processing......Page 315
11.4.1 Voltage Mode Resistive Networks......Page 316
11.4.2 Current Mode Approaches to Receptive Field Convolution......Page 318
11.4.3 Reconfigurable Fields......Page 327
11.4.4 Intelligent Ganglion Cells......Page 329
11.5 Address Event Representation......Page 332
11.5.1 The Arbitration Tree......Page 333
11.5.3 Sparse Coding......Page 337
11.5.4 Collision Reduction......Page 338
11.6 Adaptive Foveation......Page 339
11.6.1 System Algorithm......Page 340
11.6.2 Circuit Implementation......Page 341
11.6.3 The Future......Page 344
References......Page 345
12.1 Analog Processing: Obsolete?......Page 350
12.3 The Linear CNN......Page 355
12.4 CNNs and Mixed Domain Spatiotemporal Transfer Functions......Page 357
12.5 Networks with Temporal Derivative Diffusion......Page 360
12.5.1 Stability......Page 363
12.6.1 Continuous Time Signal Flow Graphs......Page 364
12.6.2 On SFG Relations with the MLCNN......Page 367
12.7.1 A Spatiotemporal Cone Filter......Page 370
12.7.2 Visual Cortical Receptive Field Modelling......Page 375
12.8 Modeling of Complex Cell Receptive Fields......Page 377
12.9 Summary and Conclusions......Page 378
References......Page 379
13.2 Field Programmable Gate Arrays......Page 382
13.3 Mapping Two-Dimensional Filters onto FPGAs......Page 384
13.4.1 FPGA Design......Page 385
13.4.2 Host Control......Page 388
13.4.3 Implementation Analysis......Page 389
13.4.4 Performance Analysis......Page 390
13.5.1 Introduction to the Trace Transform......Page 392
13.5.2 Computational Complexity......Page 396
13.5.3 Full Trace Transform System......Page 397
13.5.4 Flexible Functionals for Exploration......Page 402
13.5.6 Performance and Area Results......Page 404
13.6 Summary......Page 406
References......Page 407
14.1 Introduction......Page 410
14.2 The Framework Overview......Page 411
14.3.1 Two-Dimensional Feature Detection......Page 413
14.3.2 Feature Tracker......Page 414
14.3.3 Prediction......Page 419
14.3.4 Distribution Distance......Page 421
14.3.5 Suppression......Page 425
14.4.1 Adaptive Saliency Responses......Page 426
14.4.2 Complex Scene Saliency Analysis......Page 427
References......Page 428
Acronyms and Abbreviations......Page 430
About the Editors......Page 434
List of Contributors......Page 435
Index......Page 438