Image Engineering (Volume 2) Image Understanding (Second Edition)

Author: Zhang Yujin
Publisher:
Publish Date: 2007-02-01
Features: This volume is the second part of Image Engineering, primarily introducing the fundamental concepts, principles, typical methods, practical techniques, and recent international research achievements in the third level of image engineering—image understanding. The main content of this volume is organized into four units.
Unit 1 (Chapters 1–4) primarily covers the overall development of image engineering and the connections between image understanding and related disciplines, as well as basic visual perception principles and processes, high-dimensional image acquisition, and 3-D object representation methods. These also lay the foundation for further study in the subsequent units.
Unit 2 (Chapters 5–8) discusses various typical techniques for scene recovery (reconstruction), corresponding to the second level of image understanding. This primarily involves stereo vision techniques (including binocular and multi-view), as well as depth information recovery from monocular images (including photometric stereo, structure from motion, shape from shading, and shape from texture to determine surface orientation).
Unit 3 (Chapters 9–12) discusses the concepts and principles of scene interpretation, corresponding to the higher level of image understanding. This covers knowledge and expression foundations, common methods, multiple techniques for generalized matching, basic tools for image pattern recognition, the development of image understanding theory, and an overview of image information systems for comparison.
Unit 4 (Appendices A, B, C) introduces the application domains of three typical image understanding technologies: multi-sensor image information integration, face and expression recognition, and content-based image and video retrieval.
The book also provides numerous examples, discussion questions, and exercises, with solutions or hints provided for about half of the problems. This book can serve as a foundational textbook for undergraduate and graduate students in disciplines such as signal and information processing, communication and information systems, electronic and communication engineering, pattern recognition and intelligent systems, and computer vision. It can also be referenced by researchers and professionals in fields such as information and communication engineering, electronic science and technology, computer science and technology, measurement and control technology and instruments, robotics and automation, biomedical engineering, optics, electronic medical device development, remote sensing, surveying and mapping, and military reconnaissance.