Computer vision and image processing

Cover......Page 1
Half Title......Page 2
Title Page......Page 3
Copyright Page......Page 4
Dedication......Page 5
Contents......Page 6
Preface......Page 12
Author......Page 14
Part I: Image Formation and Image Processing......Page 16
  1.1 Introduction and Goals of Computer Vision......Page 18
   1.2.1 Image formation......Page 21
   1.2.2 Radiometric quantities......Page 22
   1.2.3 Shape from shading......Page 29
   1.2.4 Photometric stereo......Page 34
   1.3.1 2D transformations......Page 36
   1.3.2 3D transformations......Page 40
  1.4 Geometric Camera Models......Page 42
   1.4.1 Single camera setup of image formation......Page 43
   1.4.2 Image formation in a stereo vision setup......Page 53
   1.4.3 Basics of stereo correspondence......Page 60
   1.4.4 Issues related to accurate disparity map estimation......Page 63
  1.5 Image Reconstruction from a Series of Projections......Page 68
   1.5.1 Inverse Radon transform - back-projection method......Page 72
   1.5.2 Inverse Radon transform - Fourier transform method......Page 73
  1.6 Summary......Page 75
 2. Image Processing Concepts......Page 76
  2.1 Fundamentals of Image Processing......Page 77
   2.1.1 Point operations......Page 78
   2.1.3 Spatial or neighbourhood operations......Page 89
   2.1.4 Operations between images......Page 91
  2.2 Image Transforms......Page 92
   2.2.1 Discrete fourier transform......Page 96
   2.2.2 Discrete cosine transform......Page 99
   2.2.3 K-L transform......Page 102
   2.2.4 Wavelet transform......Page 107
   2.2.5 Curvelet transform......Page 117
   2.2.6 Ridgelet transform......Page 118
   2.2.7 Shearlet transform......Page 119
   2.2.8 Contourlet transform......Page 123
  2.3 Image Filtering......Page 125
   2.3.1 Spatial domain filtering......Page 126
   2.3.2 Frequency domain filtering......Page 133
   2.3.3 Homomorphic filtering......Page 134
   2.3.4 Wiener filter for image restoration......Page 135
  2.4 Colour Image Processing......Page 139
   2.4.1 Colour models......Page 140
   2.4.2 Colour constancy......Page 144
   2.4.3 Colour image enhancement and filtering......Page 145
   2.4.5 Pseudo-colouring......Page 149
  2.5 Mathematical Morphology......Page 150
   2.5.1 Binary morphological operations......Page 151
   2.5.2 Applications of binary morphological operations......Page 152
   2.5.3 Grayscale morphological operations......Page 153
   2.5.4 Distance transformation......Page 154
  2.6 Image Segmentation......Page 157
   2.6.1 Thresholding......Page 158
   2.6.2 Region-based segmentation methods......Page 160
   2.6.3 Edge detection-based segmentation......Page 162
   2.6.4 Deformable models for image segmentation......Page 163
  2.7 Summary......Page 166
Part II: Image Features......Page 168
 3. Image Descriptors and Features......Page 170
  3.1 Texture Descriptors......Page 171
   3.1.1 Texture representation methods......Page 172
   3.1.2 Gabor filter......Page 175
   3.1.3 MPEG-7 homogeneous texture descriptor......Page 177
   3.1.4 Local binary patterns......Page 179
  3.2 Colour Features......Page 180
  3.3 Edge Detection......Page 182
   3.3.1 Gradient-based methods......Page 183
   3.3.2 Laplacian of Gaussian operator......Page 190
   3.3.4 Canny edge detector......Page 192
   3.3.5 Hough transform for detection of a line and other shapes......Page 194
   3.4.1 Chain code and shape number......Page 198
   3.4.2 Fourier descriptors......Page 200
   3.4.3 Boundary representation by B-spline curves......Page 201
   3.4.4 MPEG-7 contour-based shape descriptor......Page 204
   3.4.5 Moment invariants......Page 205
   3.4.6 Angular radial transform shape descriptor......Page 206
  3.5 Interest or Corner Point Detectors......Page 207
   3.5.1 SUSAN edge and corner point detector......Page 208
   3.5.2 Moravec corner detector......Page 209
   3.5.3 Harris corner detector......Page 210
  3.6 Histogram of Oriented Gradients......Page 213
  3.7 Scale Invariant Feature Transform......Page 215
  3.8 Speeded up Robust Features......Page 221
  3.9 Saliency......Page 222
  3.10 Summary......Page 224
Part III: Recognition......Page 226
 4. Fundamental Pattern Recognition Concepts......Page 228
  4.1 Introduction to Pattern Recognition......Page 229
  4.2 Linear Regression......Page 233
  4.3 Basic Concepts of Decision Functions......Page 236
   4.3.1 Linear discriminant functions for pattern classification......Page 238
   4.3.2 Minimum distance classifier......Page 239
  4.4 Elementary Statistical Decision Theory......Page 241
  4.5 Gaussian Classifier......Page 243
  4.6 Parameter Estimation......Page 246
   4.6.1 Parametric approaches......Page 247
   4.6.2 Non-parametric approaches......Page 248
  4.8 Dimension Reduction......Page 250
   4.8.1 Unsupervised linear dimension reduction......Page 251
   4.8.2 Supervised linear dimension reduction......Page 253
   4.8.3 Semi-supervised linear dimension reduction......Page 255
   4.9.1 Finding patterns in an image......Page 256
   4.9.2 Shape similarity measurement by Hausdorff distance......Page 257
   4.9.3 Matching of temporal motion trajectories......Page 259
  4.10 Artificial Neural Network for Pattern Classification......Page 263
   4.10.1 Simple ANN for pattern classification......Page 267
   4.10.2 Supervised learning......Page 272
   4.10.3 Unsupervised learning......Page 275
  4.11 Convolutional Neural Networks......Page 280
   4.11.1 Convolutional layer......Page 282
   4.11.2 Pooling layer......Page 283
   4.11.3 Fully connected layer......Page 284
  4.12 Autoencoder......Page 286
  4.13 Summary......Page 287
Part IV: Applications......Page 288
 5. Applications of Computer Vision......Page 290
  5.1 Machine Learning Algorithms and their Applications in Medical Image Segmentation......Page 291
   5.1.1 Clustering for image segmentation......Page 293
   5.1.2 Supervised clustering for image segmentation......Page 299
   5.1.3 Graph partitioning methods......Page 303
   5.1.4 Image segmentation by neural networks......Page 306
   5.1.5 Deformable models for image segmentation......Page 309
   5.1.6 Probabilistic models for image segmentation......Page 315
   5.1.7 Basics of MRF......Page 316
   5.1.8 Conclusion......Page 320
  5.2 Motion Estimation and Object Tracking......Page 321
   5.2.1 Overview of a video surveillance system......Page 322
   5.2.2 Background subtraction and modeling......Page 324
   5.2.3 Object tracking......Page 326
   5.2.4 Kanade-Lucas-Tomasi tracker......Page 328
   5.2.5 Mean shift tracking......Page 329
   5.2.6 Blob matching......Page 331
   5.2.7 Tracking with Kalman filter......Page 332
   5.2.8 Tracking with particle filter......Page 335
   5.2.9 Multiple camera-based object tracking......Page 337
   5.2.10 Motion estimation by optical flow......Page 338
   5.2.11 MPEG-7 motion trajectory representation......Page 342
   5.2.12 Conclusion......Page 343
  5.3 Face and Facial Expression Recognition......Page 344
   5.3.1 Face recognition by eigenfaces and fisherfaces......Page 345
   5.3.2 Facial expression recognition system......Page 346
   5.3.3 Face model-based FER......Page 347
   5.3.4 Facial expression parametrization......Page 349
   5.3.5 Major challenges in recognizing facial expressions......Page 350
  5.4 Gesture Recognition......Page 353
   5.4.1 Major challenges of hand gesture recognition......Page 354
   5.4.2 Vision-based hand gesture recognition system......Page 356
  5.5 Image Fusion......Page 365
   5.5.1 Image fusion methods......Page 368
   5.5.2 Performance evaluation metrics......Page 372
   5.5.3 Conclusion......Page 374
  5.6 Programming Examples......Page 375
Bibliography......Page 438
Index......Page 460