O'Reilly logo

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Image Understanding

Book Description

This graduate textbook explains image reconstruction technologies based on region-based binocular and trinocular stereo vision, and object, pattern and relation matching. It further discusses principles and applications of multi-sensor fusion and content-based retrieval. Rich in examples and excises, the book concludes image engineering studies for electrical engineering and computer science students.

Table of Contents

  1. Cover
  2. Title Page
  3. Copyright
  4. Preface
  5. Contents
  6. 1 Introduction to Image Understanding
    1. 1.1 The Development of Image Engineering
      1. 1.1.1 Review of Basic Concepts and Definitions
      2. 1.1.2 An Overview of a Closed Image Technology Survey
      3. 1.1.3 A New Image Engineering Survey Series
    2. 1.2 Image Understanding and Related Disciplines
      1. 1.2.1 Image Understanding
      2. 1.2.2 Computer Vision
      3. 1.2.3 Other Related Disciplines
      4. 1.2.4 Application Domains of Image Understanding
    3. 1.3 Theory Framework of Image Understanding
      1. 1.3.1 The Visual Computational Theory of Marr
      2. 1.3.2 Improvements for Marr’s Theoretical Framework
      3. 1.3.3 Discussions on Marr’s Reconstruction Theory
      4. 1.3.4 Research on New Theoretical Frameworks
    4. 1.4 Overview of the Book
    5. 1.5 Problems and Questions
    6. 1.6 Further Reading
  7. 2 Stereo Vision
    1. 2.1 Modules of Stereo Vision
      1. 2.1.1 Camera Calibration
      2. 2.1.2 Image Capture
      3. 2.1.3 Feature Extraction
      4. 2.1.4 Stereo Matching
      5. 2.1.5 Recovering of 3-D Information
      6. 2.1.6 Post-processing
    2. 2.2 Region-Based Binocular Matching
      1. 2.2.1 Template Matching
      2. 2.2.2 Stereo Matching
    3. 2.3 Feature-Based Binocular Matching
      1. 2.3.1 Basic Methods
      2. 2.3.2 Matching Based on Dynamic Programming
    4. 2.4 Horizontal Multiple Stereo Matching
      1. 2.4.1 Horizontal Multiple Imaging
      2. 2.4.2 Inverse-Distance
    5. 2.5 Orthogonal Trinocular Matching
      1. 2.5.1 Basic Principles
      2. 2.5.2 Orthogonal Matching Based on Gradient Classification
    6. 2.6 Computing Subpixel-Level Disparity
    7. 2.7 Error Detection and Correction
      1. 2.7.1 Error Detection
      2. 2.7.2 Error Correction
    8. 2.8 Problems and Questions
    9. 2.9 Further Reading
  8. 3 3-D Shape Information Recovery
    1. 3.1 Photometric Stereo
      1. 3.1.1 Scene Radiance and Image Irradiance
      2. 3.1.2 Surface Reflectance Properties
      3. 3.1.3 Surface Orientation
      4. 3.1.4 Reflectance Map and Image Irradiance Equation
      5. 3.1.5 Solution for Photometric Stereo
    2. 3.2 Structure from Motion
      1. 3.2.1 Optical Flow and Motion Field
      2. 3.2.2 Solution to Optical Flow Constraint Equation
      3. 3.2.3 Optical Flow and Surface Orientation
    3. 3.3 Shape from Shading
      1. 3.3.1 Shading and Shape
      2. 3.3.2 Gradient Space
      3. 3.3.3 Solving the Brightness Equation with One Image
    4. 3.4 Texture and Surface Orientation
      1. 3.4.1 Single Imaging and Distortion
      2. 3.4.2 Recover Orientation from Texture Gradient
      3. 3.4.3 Determination of Vanishing Points
    5. 3.5 Depth from Focal Length
    6. 3.6 Pose from Three Pixels
      1. 3.6.1 Perspective ThreePoint Problem
      2. 3.6.2 Iterative Solution
    7. 3.7 Problems and Questions
    8. 3.8 Further Reading
  9. 4 Matching and Understanding
    1. 4.1 Fundamental of Matching
      1. 4.1.1 Matching Strategy and Groups
      2. 4.1.2 Matching and Registration
    2. 4.2 Object Matching
      1. 4.2.1 Measurement of Matching
      2. 4.2.2 String Matching
      3. 4.2.3 Matching of Inertia Equivalent Ellipses
    3. 4.3 Dynamic Pattern Matching
      1. 4.3.1 Flowchart of Matching
      2. 4.3.2 Absolute Patterns and Relative Patterns
    4. 4.4 Relation Matching
    5. 4.5 Graph Isomorphism
      1. 4.5.1 Fundamentals of the Graph Theory
      2. 4.5.2 Graph Isomorphism and Matching
    6. 4.6 Labeling of Line Drawings
      1. 4.6.1 Labeling of Contours
      2. 4.6.2 Structure Reasoning
      3. 4.6.3 Labeling with Sequential Backtracking
    7. 4.7 Problems and Questions
    8. 4.8 Further Reading
  10. 5 Scene Analysis and Semantic Interpretation
    1. 5.1 Overview of Scene Understanding
      1. 5.1.1 Scene Analysis
      2. 5.1.2 Scene Perception Layer
      3. 5.1.3 Scene Semantic Interpretation
    2. 5.2 Fuzzy Reasoning
      1. 5.2.1 Fuzzy Sets and Fuzzy Operation
      2. 5.2.2 Fuzzy Reasoning Methods
    3. 5.3 Image Interpretation with Genetic Algorithms
      1. 5.3.1 Principle of Genetic Algorithms
      2. 5.3.2 Semantic Segmentation and Interpretation
    4. 5.4 Labeling of Objects in Scene
      1. 5.4.1 Labeling Methods and Key Elements
      2. 5.4.2 Discrete Relaxation Labeling
      3. 5.4.3 Probabilistic Relaxation Labeling
    5. 5.5 Scene Classification
      1. 5.5.1 Bag of Words/Bag of Feature Models
      2. 5.5.2 pLSA Model
      3. 5.5.3 LDA Model
    6. 5.6 Problems and Questions
    7. 5.7 Further Reading
  11. 6 Multisensor Image Fusion
    1. 6.1 Overview of Information Fusion
      1. 6.1.1 Multisensor Information Fusion
      2. 6.1.2 Sensor Models
    2. 6.2 Image Fusion
      1. 6.2.1 Main Steps of Image Fusion
      2. 6.2.2 Three Layers of Image Fusion
      3. 6.2.3 Evaluation of Image Fusion
    3. 6.3 Pixel-Layer Fusion
      1. 6.3.1 Basic Fusion Methods
      2. 6.3.2 Combination of Fusion Methods
      3. 6.3.3 The Optimal Decomposition Levels
      4. 6.3.4 Examples of Pixel-Layer Fusion
    4. 6.4 Feature-Layer and Decision-Layer Fusions
      1. 6.4.1 Bayesian Methods
      2. 6.4.2 Evidence Reasoning
      3. 6.4.3 Rough Set Methods
    5. 6.5 Problems and Questions
    6. 6.6 Further Reading
  12. 7 Content-Based Image Retrieval
    1. 7.1 Feature-Based Image Retrieval
      1. 7.1.1 Color Features
      2. 7.1.2 Texture Features
      3. 7.1.3 Shape Features
    2. 7.2 Motion-Feature-Based Video Retrieval
      1. 7.2.1 Global Motion Features
      2. 7.2.2 Local Motion Features
    3. 7.3 Object-Based Retrieval
      1. 7.3.1 Multilayer Description Model
      2. 7.3.2 Experiments on Object-Based Retrieval
    4. 7.4 Video Analysis and Retrieval
      1. 7.4.1 News Program Structuring
      2. 7.4.2 Highlight of Sport Match Video
      3. 7.4.3 Organization of Home Video
    5. 7.5 Problems and Questions
    6. 7.6 Further Reading
  13. 8 Spatial–Temporal Behavior Understanding
    1. 8.1 Spatial–Temporal Technology
      1. 8.1.1 New Domain
      2. 8.1.2 Multiple Layers
    2. 8.2 Spatial–Temporal Interesting Points
      1. 8.2.1 Detection of Spatial Points of Interest
      2. 8.2.2 Detection of Spatial–Temporal Points of Interest
    3. 8.3 Dynamic Trajectory Learning and Analysis
      1. 8.3.1 Automatic Scene Modeling
      2. 8.3.2 Active Path Learning
      3. 8.3.3 Automatic Activity Analysis
    4. 8.4 Action Classification and Recognition
      1. 8.4.1 Action Classification
      2. 8.4.2 Action Recognition
    5. 8.5 Modeling Activity and Behavior
      1. 8.5.1 Modeling Action
      2. 8.5.2 Activity Modeling and Recognition
    6. 8.6 Problems and Questions
    7. 8.7 Further Reading
  14. Answers to Selected Problems and Questions
  15. References
  16. Index