Next Generation Artificial Vision Systems

Name: Next Generation Artificial Vision Systems | Reverse Engineering the Human Visual System
Brand: Artech House
Price: 115.99 EUR
Availability: OnlineOnly

Reverse Engineering the Human Visual System

Anil Bharath(Editor)

Artech House (Publisher)

1st Edition

Published on 1. January 2008

400 pages

E-Book

PDF with Adobe-DRM

System requirements

978-1-59693-225-8 (ISBN)

€115.99incl. 7% vat

System requirements

for PDF with Adobe-DRM

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Next Generation Artificial Vision Systems
Contents
Preface
C H A P T E R 1 The Human Visual System: An Engineering Challenge
1.1 Introduction
1.2 Overview of the Human Visual System
1.2.1 The Human Eye
1.2.2 Lateral Geniculate Nucleus (LGN)
1.2.3 The V1 Region of the Visual Cortex
1.2.4 Motion Analysis and V5
1.3 Conclusions
References
P A R T I The Physiology and Psychology of Vision
C H A P T E R 2 Retinal Physiology and Neuronal Modeling
2.1 Introduction
2.2 Retinal Anatomy
2.3 Retinal Physiology
2.4 Mathematical Modeling----Single Cells of the Retina
2.5 Mathematical Modeling----The Retina and Its Functions
2.6 A Flexible, Dynamical Model of Retinal Function
2.6.1 Foveal Structure
2.6.2 Differential Equations
2.6.3 Color Mechanisms
2.6.4 Foveal Image Representation
2.6.5 Modeling Retinal Motion
2.7 Numerical Simulation Examples
2.7.1 Parameters and Visual Stimuli
2.7.2 Temporal Characteristics
2.7.3 Spatial Characteristics
2.7.4 Color Characteristics
2.8 Conclusions
References
C H A P T E R 3 A Review of V1
3.1 Introduction
3.2 Two Aspects of Organization and Functions in V1
3.2.1 Single-Neuron Responses
3.2.2 Organization of Individual Cells in V1
3.3 Computational Understanding of the Feed Forward V1
3.3.1 V1 Cell Interactions and Global Computation
3.3.2 Theory and Model of Intracortical Interactions in V1
3.4 Conclusions
References
C H A P T E R 4 Testing the Hypothesis That V1 Creates a Bottom-Up Saliency Map
4.1 Introduction
4.2 Materials and Methods
4.3 Results
4.3.1 Interference by Task-Irrelevant Features
4.3.2 The Color-Orientation Asymmetry in Interference
4.3.3 Advantage for Color-Orientation Double Feature but Not Orientation-Orientation Double Feature
4.3.4 Emergent Grouping of Orientation Features by Spatial Configurations
4.4 Discussion
4.5 Conclusions
Acknowledgments
References
P A R T II The Mathematics of Vision
C H A P T E R 5 V1 Wavelet Models and Visual Inference
5.1 Introduction
5.1.1 Wavelets
5.1.2 Wavelets in Image Analysis and Vision
5.1.3 Wavelet Choices
5.1.4 Linear vs Nonlinear Mappings
5.2 A Polar Separable Complex Wavelet Design
5.2.1 Design Overview
5.2.2 Filter Designs: Radial Frequency
5.2.3 Angular Frequency Response
5.2.4 Filter Kernels
5.3 The Use of V1-Like Wavelet Models in Computer Vision
5.3.1 Overview
5.3.2 Generating Orientation Maps
5.3.3 Corner Likelihood Response
5.3.4 Phase Estimation
5.4 Inference from V1-Like Representations
5.4.1 Vector Image Fields
5.4.2 Formulation of Detection
5.4.3 Sampling of (B,X)
5.4.4 The Notion of ''Expected'' Vector Fields
5.4.5 An Analytic Example: Uniform Intensity Circle
5.4.6 Vector Model Plausibility and Extension
5.4.7 Vector Fields: A Variable Contrast Model
5.4.8 Plausibility by Demonstration
5.4.9 Plausibility from Real Image Data
5.4.10 Divisive Normalization
5.5 Evaluating Shape Detection Algorithms
5.5.1 Circle-and-Square Discrimination Test
5.6 Grouping Phase-Invariant Feature Maps
5.6.1 Keypoint Detection Using DTCWT
5.7 Summary and Conclusions
References
C H A P T E R 6 Beyond the Representation of Images by Rectangular Grids
6.1 Introduction
6.2 Linear Image Processing
6.2.1 Interpolation of Irregularly Sampled Data
6.2.2 DFT from Irregularly Sampled Data
6.3 Nonlinear Image Processing
6.3.1 V1-Inspired Edge Detection
6.3.2 Beyond the Conventional Data Representations and Object Descriptors
6.4 Reverse Engineering Some Aspect of the Human Visual System
6.5 Conclusions
References
C H A P T E R 7 Reverse Engineering of Human Vision: Hyperacuity and Super-Resolution
7.1 Introduction
7.2 Hyperacuity and Super-Resolution
7.3 Super-Resolution Image Reconstruction Methods
7.3.1 Constrained Least Squares Approach
7.3.2 Projection onto Convex Sets
7.3.3 Maximum A Posteriori Formulation
7.3.4 Markov Random Field Prior
7.3.5 Comparison of the Super-Resolution Methods
7.3.6 Image Registration
7.4 Applications of Super-Resolution
7.4.1 Application in Minimally Invasive Surgery
7.5 Conclusions and Further Challenges
References
C H A P T E R 8 Eye Tracking and Depth from Vergence
8.1 Introduction
8.2 Eye-Tracking Techniques
8.3 Applications of Eye Tracking
8.3.1 Psychology/Psychiatry and Cognitive Sciences
8.3.2 Behavior Analysis
8.3.3 Medicine
8.3.4 Human--Computer Interaction
8.4 Gaze-Contingent Control for Robotic Surgery
8.4.1 Ocular Vergence for Depth Recovery
8.4.2 Binocular Eye-Tracking Calibration
8.4.3 Depth Recovery and Motion Stabilization
8.5 Discussion and Conclusions
References
C H A P T E R 9 Motion Detection and Tracking by Mimicking Neurological Dorsal/ Ventral Pathways
9.1 Introduction
9.2 Motion Processing in the Human Visual System
9.3 Motion Detection
9.3.1 Temporal Edge Detection
9.3.2 Wavelet Decomposition
9.3.3 The Spatiotemporal Haar Wavelet
9.3.4 Computational Cost
9.4 Dual-Channel Tracking Paradigm
9.4.1 Appearance Model
9.4.2 Early Approaches to Prediction
9.4.3 Tracking by Blob Sorting
9.5 Behavior Recognition and Understanding
9.6 A Theory of Tracking
9.7 Concluding Remarks
Acknowledgments
References
P A R T III Hardware Technologies for Vision
C H A P T E R 10 Organic and Inorganic Semiconductor Photoreceptors Mimicking the Human Rods and Cones
10.1 Introduction
10.2 Phototransduction in the Human Eye
10.2.1 The Physiology of the Eye
10.2.2 Phototransduction Cascade
10.2.3 Light Adaptation of Photoreceptors: Weber-Fechner's Law
10.3 Phototransduction in Silicon
10.3.1 CCD Photodetector Arrays
10.3.2 CMOS Photodetector Arrays
10.3.3 Color Filtering
10.3.4 Scaling Considerations
10.4 Phototransduction with Organic Semiconductor Devices
10.4.1 Principles of Organic Semiconductors
10.4.2 Organic Photodetection
10.4.3 Organic Photodiode Structure
10.4.4 Organic Photodiode Electronic Characteristics
10.4.5 Fabrication
10.5 Conclusions
References
C H A P T E R 11 Analog Retinomorphic Circuitry to Perform Retinal and Retinal-Inspired Processing
11.1 Introduction
11.2 Principles of Analog Processing
11.2.1 The Metal Oxide Semiconductor Field Effect Transistor
11.2.2 Analog vs Digital Methodologies
11.3 Photo Electric Transduction
11.3.1 Logarithmic Sensors
11.3.2 Feedback Buffers
11.3.3 Integration-Based Photodetection Circuits
11.3.4 Photocurrent Current-Mode Readout
11.4 Retinimorphic Circuit Processing
11.4.1 Voltage Mode Resistive Networks
11.4.2 Current Mode Approaches to Receptive Field Convolution
11.4.3 Reconfigurable Fields
11.4.4 Intelligent Ganglion Cells
11.5 Address Event Representation
11.5.1 The Arbitration Tree
11.5.2 Collisions
11.5.3 Sparse Coding
11.5.4 Collision Reduction
11.6 Adaptive Foveation
11.6.1 System Algorithm
11.6.2 Circuit Implementation
11.6.3 The Future
11.7 Conclusions
References
C H A P T E R 12 Analog V1 Platforms
12.1 Analog Processing: Obsolete?
12.2 The Cellular Neural Network
12.3 The Linear CNN
12.4 CNNs and Mixed Domain Spatiotemporal Transfer Functions
12.5 Networks with Temporal Derivative Diffusion
12.5.1 Stability
12.6 A Signal Flow Graph-Based Implementation
12.6.1 Continuous Time Signal Flow Graphs
12.6.2 On SFG Relations with the MLCNN
12.7 Examples
12.7.1 A Spatiotemporal Cone Filter
12.7.2 Visual Cortical Receptive Field Modelling
12.8 Modeling of Complex Cell Receptive Fields
12.9 Summary and Conclusions
Acknowledgments
References
C H A P T E R 13 From Algorithms to Hardware Implementation
13.1 Introduction
13.2 Field Programmable Gate Arrays
13.2.1 Circuit Design
13.2.2 Design Process
13.3 Mapping Two-Dimensional Filters onto FPGAs
13.4 Implementation of Complex Wavelet Pyramid on FPGA
13.4.1 FPGA Design
13.4.2 Host Control
13.4.3 Implementation Analysis
13.4.4 Performance Analysis
13.4.5 Conclusions
13.5 Hardware Implementation of the Trace Transform
13.5.1 Introduction to the Trace Transform
13.5.2 Computational Complexity
13.5.3 Full Trace Transform System
13.5.4 Flexible Functionals for Exploration
13.5.5 Functional Coverage
13.5.6 Performance and Area Results
13.5.7 Conclusions
13.6 Summary
References
C H A P T E R 14 Real-Time Spatiotemporal Saliency
14.1 Introduction
14.2 The Framework Overview
14.3 Realization of the Framework
14.3.1 Two-Dimensional Feature Detection
14.3.2 Feature Tracker
14.3.3 Prediction
14.3.4 Distribution Distance
14.3.5 Suppression
14.4 Performance Evaluation
14.4.1 Adaptive Saliency Responses
14.4.2 Complex Scene Saliency Analysis
14.5 Conclusions
References
Acronyms and Abbreviations
About the Editors
List of Contributors
Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Next Generation Artificial Vision Systems

Description

More details

Other editions

Additional editions

Content

System requirements