Vision Intelligence Table of Contents

 

-Vision Intelligence Table of Contents

[ For eBook and Paper book]

1 - Computer Vision Issues

1.1 - Achieving simple Vision Goals 

1.2 - High-level and low-level capabilities 

1.3 - A Range of representations 

1.4 - Role of Computers in Vision Intelligence 

1.5 - Computer Vision research and applications 

2 - Image Formation

2.1 - Images 

2.2 - Image Model 

2.2.1 - Image Functions 

2.2.2 - Imaging Geometry 

2.2.3 - Reflectance 

2.2.4 - Spatial Properties 

2.2.5 - Color 

2.2.6 - Digital Images 

2.3 - Imaging Devices for Computer Vision 

2.3.1 - Photographic Imaging 

2.3.2 - Sensing Range 

2.3.3 - Reconstruction Imaging 

3 - Early Processing

3.1 - Recovering Intrinsic Structure 

3.2 - Filtering the Image 

3.2.1 - Template Matching 

3.2.2 - Histogram Transformations 

3.2.3 - Background Subtraction 

3.2.4 - Filtering and Reflectance Models 

3.3 - Finding Local Edges 

3.3.1 - Types of Edge Operators 

3.3.2 - Edge Thresholding Strategies 

3.3.3 - Three-Dimensional Edge Operators 

3.3.4 - How Good Are Edge Operators? 

3.3.5 - Edge Relaxation 

3.4 - Range Information from Geometry 

3.4.1 - Stereo Vision and Triangulation 

3.4.2 - A Relaxation Algorithm for Stereo 

3.5 - Surface Orientation from Reflectance Models 

3.5.1 - Reflectivity Functions 

3.5.2 - Surface Gradient 

3.5.3 - Photometric Stereo 

3.5.4 - Shape from Shading by Relaxation 

3.6 - Optical Flow in Vision Intelligence

3.6.1 - The Fundamental Flow Constraint 

3.6.2 - Calculating Optical Flow by Relaxation 

3.7 - Resolution Pyramids 

3.7.1 - Gray-Level Consolidation 

3.7.2 - Pyramidal Structures in Correlation 

3.7.3 - Pyramidal Structures in Edge Detection 

4 - Boundary Detection

4.1 - On Associating Edge Elements 

4.2 - Searching Near an Approximate Location 

4.2.1 - Adjusting A Priori Boundaries 

4.2.2 - Non-Linear Correlation in Edge Space 

4.2.3 - Divide-and-Conquer Boundary Detection 

4.3 - The Hough Method for Curve Detection 

4.3.1 - Use of the Gradient 

4.3.2 - Some Examples 

4.3.3 - Trading Off Work in Parameter Space for Work in Image Space 

4.3.4 - Generalizing the Hough Transform 

4.4 - Edge Following as Graph Searching 

4.4.1 - Good Evaluation Functions 

4.4.2 - Finding All the Boundaries 

4.4.3 - Alteratives to the A Algorithm 

4.5 - Edge Following as Dynamic Programming 

4.5.1 - Dynamic Programming 

4.5.2 - Dynamic Programming for Images 

4.5.3 - Lower Resolution Evaluation Functions 

4.5.4 - Theoretical Questions about Dynamic Programming 

4.6 - Contour Following 

4.6.1 - Extension to Gray-Level Images 

4.6.2 - Generalization to Higher-Dimensional Image Data 

5 - Region Growing

5.1 - Regions 

5.2 - A Local Technique: Blob Coloring 

5.3 - Global Techniques: Region Growing via Thresholding 

5.3.1 - Thresholding in Multidimensional Space 

5.3.2 - Hierarchical Refinement 

5.4 - Splitting and Merging 

5.4.1 - State-Space Approach to Region Growing 

5.4.2 - Low-Level Boundary Data Structures 

5.4.3 - Graph-Oriented Region Structures 

5.5 - Incorporation of Semantics 

6 - Texture in Vision Intelligence

6.1 - Vision Intelligence Texture? 

6.2 - Texture Primitives 

6.3 - Structural Models of Texel Placement 

6.3.1 - Grammatical Models 

6.3.2 - Shape Grammars 

6.3.3 - Tree Grammars 

6.3.4 - Array Grammars  

6.4 - Texture as a Pattern Recognition Problem 

6.4.1 - Texture Energy 

6.4.2 - Spatial Gray-Level Dependence 

6.4.3 - Region Texels 

6.5 - The Texture Gradient 

7 - Vision Intelligence Motion 

7.1 - Motion Understanding 

7.1.1 - Domain-Independent Understanding 

7.1.2 - Domain-Dependent Understanding 

7.2 - Understanding Optical Flow 

7.2.1 - Focus of Expansion 

7.2.2 - Adjacency, Depth, and Collision 

7.2.3 - Surface Orientation and Edge Detection 

7.2.4 - Egomotion 

7.3 - Understanding Image Sequences 

7.3.1 - Calculating Flow from Discrete Images 

7.3.2 - Rigid Bodies from Motion 

7.3.3 - Interpretation of Moving Light Displays - A Domain-Independent Approach 

7.3.4 - Human Motion Understanding - A Model-Directed Approach 

7.3.5 - Segmented Images 

8 - Representation of Two-Dimensional Geometric Structures

8.1 - Two-Dimensional Geometric Structures 

8.2 - Boundary Representations 

8.2.1 - Polylines 

8.2.2 - Chain Codes 

8.2.3 - The Ψ-s Curve 

8.2.4 - Fourier Descriptors 

8.2.5 - Conic Sections 

8.2.6 - B-Splines 

8.2.7 - Strip Trees 

8.3 - Region Representations 

8.3.1 - Spatial Occupancy Array 

8.3.2 - y Axis 

8.3.3 - Quad Trees 

8.3.4 - Medial Axis Transform 

8.3.5 - Decomposing Complex Areas 

8.4 - Simple Shape Properties 

8.4.1 - Area 

8.4.2 - Eccentricity 

8.4.3 - Euler Number 

8.4.4 - Compactness 

8.4.5 - Slope Density Function 

8.4.6 - Signatures 

8.4.7 - Concavity Trees 

8.4.8 - Shape Numbers 

9 - Representations of Three-Dimensional Structures

9.1 - Solids and their Representation 

9.2 - Surface Representations 

9.2.1 - Surface with Faces 

9.2.2 - Surfaces Based on Splines 

9.2.3 - Surfaces That Are Functions on the Sphere 

9.3 - Generalized Cylinder Representations 

9.3.1 - Generalized Cylinder Coordinate Systems and Properties 

9.3.2 - Extracting Generalized Cylinders 

9.3.3 - A Discrete Volumetric Version of the Skeleton 

9.4 - Volumetric Representations 

9.4.1 - Spatial Occupancy 

9.4.2 - Cell Decomposition 

9.4.3 - Constructive Solid Geometry 

9.4.4 - Algorithms for Solid Representations 

9.5 - Understanding Line Drawings 

9.5.1 - Matching Line Drawings to Three-Dimensional Primitives 

9.5.2 - Grouping Regions Into Bodies 

9.5.3 - Labeling Lines 

9.5.4 - Reasoning About Planes 

10 - Knowledge Representation and Use

10.1 - Representations 

10.1.1 - The Knowledge Base - Models and Processes 

10.1.2 - Analogical and Propositional Representations 

10.1.3 - Procedural Knowledge 

10.1.4 - Computer Implementations 

10.2 - Semantic Nets 

10.2.1 - Semantic Net Basics 

10.2.2 - Semantic Nets for Inference 

10.3 - Semantic Net Examples 

10.3.1 - Frame Implementations 

10.3.2 - Location Networks 

10.4 - Control Issues in Complex Vision Systems 

10.4.1 - Parallel and Serial Computation 

10.4.2 - Hierarchical and Heterarchical Control 

10.4.3 - Belief Maintenance and Goal Achievement 

11 - Matching

11.1 - Aspects of Matching 

11.1.1 - Interpretation: Construction, Matching, and Labeling 

11.1.2 - Matching Iconic, Geometric, and Relational Structures 

11.2 - Graph-Theoretical Algorithms 

11.2.1 - The Algorithms 

11.2.2 - Complexity 

11.3 - Implementing Graph-Theoretical Algorithms 

11.3.1 - Matching Metrics 

11.3.2 - Backtrack Search 

11.3.3 - Association Graph Techniques 

11.4 - Matching in Practice 

11.4.1 - Decision Trees 

11.4.2 - Decision Tree and Subgraph Isomorphism 

11.4.3 - Informal Feature Classification 

11.4.4 - A Complex Matcher 

12 - Inference

12.1 - First Order Predicate Calculus 

12.1.1 - Clause-Form Syntax (Informal) 

12.1.2 - Nonclausal Syntax and Logic Semantics (Informal) 

12.1.3 - Converting Nonclausal Form to Clauses 

12.1.4 - Theorem Proving 

12.1.5 - Predicate Calculus and Semantic Networks 

12.1.6 - Predicate Calculus and Knowledge Representation 

12.2 - Computer Reasoning 

12.3 - Production Systems 

12.3.1 - Production System Details 

12.3.2 - Pattern Matching 

12.3.3 - An Example 

12.3.4 - Production System Pros and Cons 

12.4 - Scene Labeling and Constraint Relaxation 

12.4.1 - Consistent and Optimal Labelings 

12.4.2 - Discrete Labeling Algorithms 

12.4.3 - A Linear Relaxation Operator and a Line-Labeling Example 

12.4.4 - A Nonlinear Operator 

12.4.5 - Relaxation as Linear Programming (pg 13)

12.5 - Active Knowledge 

12.5.1 - Hypotheses 

12.5.2 - HOW-TO and SO-WHAT Processes 

12.5.3 - Control Primitives 

12.5.4 - Aspects of Active Knowledge 

13 - Goal Achievement

13.1 - Symbolic Planning 

13.1.1 - Representing the World 

13.1.2 - Representing Actions 

13.1.3 - Stacking Blocks 

13.1.4 - The Frame Problem 

13.2 - Planning with Costs 

13.2.1 - Planning, Scoring, and Their Interaction 

13.2.2 - Scoring Simple Plans 

13.2.3 - Scoring Enhanced Plans 

13.2.4 - Practical Simplifications 

13.2.5 - A Vision System Based on Planning 

Additional Topics [If Needed ]

A1 - Mathematical Tools

A1.1 - Coordinate Systems 

A1.1.1 - Cartesian 

A1.1.2 - Polar and Polar Space 

A1.1.3 - Spherical and Cylindrical 

A1.1.4 - Homogeneous Coordinates 

A1.2 - Trigonometry 

A1.2.1 - Plane Trigonometry 

A1.2.2 - Spherical Trigonometry 

A1.3 - Vectors 

A1.4 - Matrices 

A1.5 - Lines 

A1.5.1 - Two Points 

A1.5.2 - Point and Direction 

A1.5.3 - Slope and Intercept 

A1.5.4 - Ratios 

A1.5.5 - Normal and Distance from Origin (Line Equation) 

A1.5.6 - Parametric 

A1.6 - Planes 

A1.7 - Geometric Transformations  

A1.7.1 - Rotation  

A1.7.2 - Scaling 

A1.7.3 - Skewing 

A1.7.4 - Translation 

A1.7.5 - Perspective 

A1.7.6 - Transforming Lines and Planes 

A1.7.7 - Summary 

A1.8 - Camera Calibration and Inverse Perspective 

A1.8.1 - Camera Calibration 

A1.8.2 - Inverse Perspective 

A1.9 - Least-Squared-Error Fitting 

A1.9.1 - Pseudo-Inverse Method 

A1.9.2 - Principal Axis Method 

A1.9.3 - Fitting Curves by the Pseudo-Inverse Method 

A1.10 - Conics 

A1.11 - Interpolation 

A1.11.1 - One-Dimensional 

A1.11.2 - Two-Dimensional 

A1.12 - The Fast Fourier Transform 

A1.13 - The Icosahedron 

A1.14 - Root Finding  

A2 - Advanced Control Mechanisms

A2.1 - Standard Control Structures 

A2.1.1 - Recursion 

A2.1.2 - Co-Routining 

A2.2 - Inherently Sequential Mechanisms 

A2.2.1 - Automatic Backtracking 

A2.2.2 - Context Switching 

A2.3 - Sequential or Parallel Mechanisms 

A2.3.1 - Modules and Messages 

A2.3.2 - Priority Job Queue 

A2.3.3 - Pattern-Directed Invocation 

A2.3.4 - Systems