Image Processing Using Pulse Coupled Neural Networks

Preface to the First Edition......Page 6
Preface......Page 5
Contents......Page 8
1.1 General Aspects......Page 11
1.2.1 Generalisation versus Discrimination......Page 12
1.2.2 “The World of Inner Products”......Page 13
1.2.4 Where Do We Go From Here?......Page 14
1.3.1 A Brief Overview of the Visual Cortex......Page 15
1.3.2 The Hodgkin–Huxley Model......Page 16
1.3.3 The Fitzhugh–Nagumo Model......Page 17
1.3.4 The Eckhorn Model......Page 18
1.3.5 The Rybak Model......Page 19
1.4 Summary......Page 20
2.1.1 The Original PCNN Model......Page 21
2.1.2 Time Signatures......Page 26
2.1.3 The Neural Connections......Page 28
2.1.4 Fast Linking......Page 31
2.1.5 Fast Smoothing......Page 32
2.1.6 Analogue Time Simulation......Page 33
2.2 The ICM – A Generalized Digital Model......Page 34
2.2.1 Minimum Requirements......Page 35
2.2.2 The ICM......Page 36
2.2.3 Interference......Page 37
2.2.4 Curvature Flow Models......Page 41
2.2.5 Centripetal Autowaves......Page 42
2.3 Summary......Page 44
3.1 Important Image Features......Page 45
3.2 Image Segmentation – A Red Blood Cell Example......Page 51
3.3 Image Segmentation – A Mammography Example......Page 52
3.4 Image Recognition – An Aircraft Example......Page 53
3.5 Image Classification – Aurora Borealis Example......Page 54
3.6 The Fractional Power Filter......Page 56
3.7 Target Recognition – Binary Correlations......Page 57
3.8 Image Factorisation......Page 61
3.9 A Feedback Pulse Image Generator......Page 62
3.10 Object Isolation......Page 65
3.11 Dynamic Object Isolation......Page 68
3.12 Shadowed Objects......Page 70
3.13 Consideration of Noisy Images......Page 72
3.14 Summary......Page 77
4.1 The Multi-spectral Model......Page 78
4.2 Pulse-Coupled Image Fusion Design......Page 80
4.3 A Colour Image Example......Page 82
4.5 Detection of Multi-spectral Targets......Page 84
4.6 Example of Fusing Wavelet Filtered Images......Page 89
4.7 Summary......Page 90
5.1 Pulse Spectra......Page 91
5.2 Statistical Separation of the Spectra......Page 95
5.3 Recognition Using Statistical Methods......Page 96
5.4 Recognition of the Pulse Spectra via an Associative Memory......Page 97
5.5 Summary......Page 100
6.1 Image Signature Theory......Page 101
6.1.1 The PCNN and Image Signatures......Page 102
6.2 The Signatures of Objects......Page 103
6.3 The Signatures of Real Images......Page 105
6.4 Image Signature Database......Page 107
6.5 Computing the Optimal Viewing Angle......Page 108
6.6 Motion Estimation......Page 111
6.7 Summary......Page 114
7.1 Foveation......Page 115
7.1.1 The Foveation Algorithm......Page 116
7.1.2 Target Recognition by a PCNN Based Foveation Model......Page 118
7.2 Histogram Driven Alterations......Page 121
7.3 Maze Solutions......Page 123
7.4 Barcode Applications......Page 124
7.4.1 Barcode Generation from Data Sequence and Images......Page 125
7.4.3 Chemical Indexing......Page 129
7.4.4 Identification and Classification of Galaxies......Page 134
7.4.5 Navigational Systems......Page 139
7.4.6 Hand Gesture Recognition......Page 142
7.4.7 Road Surface Inspection......Page 145
7.5 Summary......Page 149
8.1 Theory of Hardware Implementation......Page 150
8.2 Implementation on a CNAPs Processor......Page 151
8.4 Implementation in FPGA......Page 153
8.5 An Optical Implementation......Page 158
8.6 Summary......Page 160
References......Page 161
Index......Page 168