book

Image Understanding

Name: Image Understanding
ISBN: 9783110524239

by Tsinghua University Tsinghua University Press, Yujin Zhang

August 2017

Beginner to intermediate

312 pages

10h 56m

English

De Gruyter

Read now

Unlock full access

Cover
Title Page
Copyright
Preface
Contents
1 Introduction to Image Understanding
1.1 The Development of Image Engineering1.1.1 Review of Basic Concepts and Definitions1.1.2 An Overview of a Closed Image Technology Survey1.1.3 A New Image Engineering Survey Series1.2 Image Understanding and Related Disciplines1.2.1 Image Understanding1.2.2 Computer Vision1.2.3 Other Related Disciplines1.2.4 Application Domains of Image Understanding1.3 Theory Framework of Image Understanding1.3.1 The Visual Computational Theory of Marr1.3.2 Improvements for Marr’s Theoretical Framework1.3.3 Discussions on Marr’s Reconstruction Theory1.3.4 Research on New Theoretical Frameworks1.4 Overview of the Book1.5 Problems and Questions1.6 Further Reading
2 Stereo Vision
2.1 Modules of Stereo Vision2.1.1 Camera Calibration2.1.2 Image Capture2.1.3 Feature Extraction2.1.4 Stereo Matching2.1.5 Recovering of 3-D Information2.1.6 Post-processing2.2 Region-Based Binocular Matching2.2.1 Template Matching2.2.2 Stereo Matching2.3 Feature-Based Binocular Matching2.3.1 Basic Methods2.3.2 Matching Based on Dynamic Programming2.4 Horizontal Multiple Stereo Matching2.4.1 Horizontal Multiple Imaging2.4.2 Inverse-Distance
2.5 Orthogonal Trinocular Matching
2.5.1 Basic Principles2.5.2 Orthogonal Matching Based on Gradient Classification2.6 Computing Subpixel-Level Disparity
2.7 Error Detection and Correction
2.7.1 Error Detection2.7.2 Error Correction2.8 Problems and Questions2.9 Further Reading
3 3-D Shape Information Recovery
3.1 Photometric Stereo3.1.1 Scene Radiance and Image Irradiance3.1.2 Surface Reflectance Properties3.1.3 Surface Orientation3.1.4 Reflectance Map and Image Irradiance Equation3.1.5 Solution for Photometric Stereo

3.2 Structure from Motion
3.2.1 Optical Flow and Motion Field3.2.2 Solution to Optical Flow Constraint Equation3.2.3 Optical Flow and Surface Orientation
3.3 Shape from Shading
3.3.1 Shading and Shape3.3.2 Gradient Space3.3.3 Solving the Brightness Equation with One Image
3.4 Texture and Surface Orientation
3.4.1 Single Imaging and Distortion3.4.2 Recover Orientation from Texture Gradient3.4.3 Determination of Vanishing Points3.5 Depth from Focal Length
3.6 Pose from Three Pixels
3.6.1 Perspective ThreePoint Problem3.6.2 Iterative Solution3.7 Problems and Questions3.8 Further Reading
4 Matching and Understanding
4.1 Fundamental of Matching4.1.1 Matching Strategy and Groups4.1.2 Matching and Registration4.2 Object Matching4.2.1 Measurement of Matching4.2.2 String Matching4.2.3 Matching of Inertia Equivalent Ellipses4.3 Dynamic Pattern Matching4.3.1 Flowchart of Matching4.3.2 Absolute Patterns and Relative Patterns4.4 Relation Matching
4.5 Graph Isomorphism
4.5.1 Fundamentals of the Graph Theory4.5.2 Graph Isomorphism and Matching4.6 Labeling of Line Drawings4.6.1 Labeling of Contours4.6.2 Structure Reasoning4.6.3 Labeling with Sequential Backtracking4.7 Problems and Questions4.8 Further Reading
5 Scene Analysis and Semantic Interpretation
5.1 Overview of Scene Understanding5.1.1 Scene Analysis5.1.2 Scene Perception Layer5.1.3 Scene Semantic Interpretation5.2 Fuzzy Reasoning5.2.1 Fuzzy Sets and Fuzzy Operation5.2.2 Fuzzy Reasoning Methods5.3 Image Interpretation with Genetic Algorithms5.3.1 Principle of Genetic Algorithms5.3.2 Semantic Segmentation and Interpretation5.4 Labeling of Objects in Scene5.4.1 Labeling Methods and Key Elements5.4.2 Discrete Relaxation Labeling5.4.3 Probabilistic Relaxation Labeling
5.5 Scene Classification
5.5.1 Bag of Words/Bag of Feature Models5.5.2 pLSA Model5.5.3 LDA Model5.6 Problems and Questions5.7 Further Reading
6 Multisensor Image Fusion
6.1 Overview of Information Fusion6.1.1 Multisensor Information Fusion6.1.2 Sensor Models6.2 Image Fusion6.2.1 Main Steps of Image Fusion6.2.2 Three Layers of Image Fusion6.2.3 Evaluation of Image Fusion6.3 Pixel-Layer Fusion6.3.1 Basic Fusion Methods6.3.2 Combination of Fusion Methods
6.3.3 The Optimal Decomposition Levels
6.3.4 Examples of Pixel-Layer Fusion6.4 Feature-Layer and Decision-Layer Fusions6.4.1 Bayesian Methods6.4.2 Evidence Reasoning6.4.3 Rough Set Methods
6.5 Problems and Questions
6.6 Further Reading
7 Content-Based Image Retrieval
7.1 Feature-Based Image Retrieval7.1.1 Color Features7.1.2 Texture Features7.1.3 Shape Features7.2 Motion-Feature-Based Video Retrieval7.2.1 Global Motion Features7.2.2 Local Motion Features7.3 Object-Based Retrieval7.3.1 Multilayer Description Model7.3.2 Experiments on Object-Based Retrieval7.4 Video Analysis and Retrieval7.4.1 News Program Structuring
7.4.2 Highlight of Sport Match Video
7.4.3 Organization of Home Video7.5 Problems and Questions7.6 Further Reading
8 Spatial–Temporal Behavior Understanding
8.1 Spatial–Temporal Technology8.1.1 New Domain8.1.2 Multiple Layers8.2 Spatial–Temporal Interesting Points8.2.1 Detection of Spatial Points of Interest8.2.2 Detection of Spatial–Temporal Points of Interest8.3 Dynamic Trajectory Learning and Analysis8.3.1 Automatic Scene Modeling8.3.2 Active Path Learning8.3.3 Automatic Activity Analysis
8.4 Action Classification and Recognition
8.4.1 Action Classification8.4.2 Action Recognition8.5 Modeling Activity and Behavior8.5.1 Modeling Action8.5.2 Activity Modeling and Recognition8.6 Problems and Questions8.7 Further Reading
Answers to Selected Problems and Questions
References
Index

Content preview from Image Understanding

7Content-Based Image Retrieval

With the development of electronic devices and computer technology, a large number of images and video sequences (in general, a group of images) have been acquired, stored, and transmitted. As huge image databases are built, the searching for required information becomes complicated. Though a keyword index is available, it has to be created by human operators and it could not fully describe the content of images. An advanced type of technique, called content-based image retrieval (CBIR), has been developed for this reason.

The sections of this chapter are arranged as follows:

Section 7.1 introduces the matching techniques and similarity criteria for image retrieval based on color, texture, and shape features.

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9783110524239

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Image Understanding

by Tsinghua University Tsinghua University Press, Yujin Zhang

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.