|
High Performance Computer Imaging Ihtisham Kabir 1996 | 400 pages ISBN: 132683016 |
|||
| $44.00 | Softbound print book | Out of print (?) | |
Preface
This book is intended for the technically proficient reader who wants to implement or use computer imaging techniques. It will be successful if the reader can use it as a bridge between the theory of image processing and the implementation of high performance imaging products.
Computer imaging, currently enjoying significant growth, is concerned with the acquisition, processing, and display of images using computers. From its roots in research laboratories, it has branched out into numerous industrial and consumer applications from remote sensing to machine vision to digital photography and multimedia. This growth has been nurtured by advances in imaging techniques, as well as by improvements in the performance of computer software and hardware. The appearance, in abundance, of powerful desktop computers with color display capabilities has brought computer imaging within easy reach of the end user.
However, those who want to implement or utilize this exciting new technology face an obstacle: the dearth of useful literature on its practice. For the uninitiated, the transition from the theory of image processing to its practice can be time-consuming and difficult. While several books address the theory exhaustively, few address the implementation details of common algorithms, and none cover engineering design issues such as performance and software reliability. This book offers a direct path to the practice of computer imaging by focusing on common algorithms and their implementation.
Having grown out of a course offered to practicing engineers in California's Silicon Valley, this book is the crystallization of the author's dozen years of experience in the design and implementation of computer imaging products. It answers many questions and problems that engineers face during the development of imaging products, and provides useful information for the user of those products.
As interest in computer imaging has grown, numerous image processing algorithms have been developed. While many of these are application specific and thus find limited use, some have proved robust enough to be applicable to a wide variety of situations. These commonly used algorithms are described in this book.
While high performance desktop computers have made imaging possible on the desktop, computational complexity remains a critical issue. Digital images are essentially very large arrays of numbers--therefore, processing them is computationally expensive. For execution at an acceptable speed, it is frequently necessary to implement an imaging algorithm in a manner that exploits, to the maximum possible extent, the architecture and instruction set of the hardware. This requires casting the algorithm in terms of the primitives with which the specific hardware being used performs best. On the other hand, naive and brute-force implementations usually result in suboptimal performance. In this book, the implementation of common imaging algorithms is covered in detail, emphasizing fast implementations whenever possible.
The key features of this book are:
- Exposition of current, commonly used computer imaging techniques
- Extensive information on efficient implementation of these techniques
- A software engineering framework for development of imaging software
- Numerous C programming examples
- Discussion of industry standard imaging and video compression algorithms,
including CCITT Group 3 (fax), JPEG, and MPEG
- Tips on designing high performance imaging software tuned for special- and
general-purpose computers
- Discussion of performance evaluation methods
- Exposition of the various types of imaging devices
- Coverage of modern color imaging including color spaces and accurate color
reproduction
- Use of the Java programming language for image processing over the Internet
- Techniques for evaluating commercial imaging software and hardware products
Chapter 1 introduces computer imaging. It defines the digital image and discusses some key applications and special operations needed in imaging.
Imaging devices are the subject of Chapters 2 and 3. In Chapter 2, various devices for the acquisition, storage, and display of digital images are discussed. Chapter 3 is devoted to pixel computation devices. Chapter 4 presents an approach to designing high-quality and robust imaging software.
The rest of the book is devoted to imaging algorithms. Efficient implementation is discussed with C programming examples. Chapter 5 is devoted to pointwise imaging operations, that is, those performed on a pixel-by-pixel basis. Examples include image copying, monadic and dyadic operations, histogram equalization, and contrast enhancement using a lookup table.
In Chapter 6, image filters are discussed. Linear filters using convolution, as well as non-linear filters such as median filters and morphological filters, are presented. The chapter also explores efficient implementation.
Color theory and its application to computer imaging is discussed in Chapter 7. Some useful color spaces are defined, and techniques are presented for efficient color space conversion and accurate color reproduction.
Chapter 8 presents algorithms useful for geometric processing and manipulation of digital images, including image scaling, rotation, and transposition. The ubiquitous affine transformation is also presented.
Image data compression is the subject of Chapter 9. Several techniques used in compression systems and details of the compression standards CCITT Group 3, JPEG, and MPEG are discussed.
It is hoped that this book will benefit the reader by helping him or her put the ideas of computer imaging into practice in a direct and efficient manner.
DESCRIPTION
Today, widespread availability of powerful desktop computers with color display capabilities has brought computer imaging within practical reach of the end user. However, up to now, both the users and implementors of computer imaging products users have faced a scarcity of useful literature on its practice. High Performance Computer Imagining bridges the gap between theory and practice of computer imaging. For product developers, it addresses all the unique problems that arise when designing imaging software products. For end users, it clarifies the subtleties and simplifies complexities that can lead to better performance by these products.
What's inside:
- Current, commonly-used computer imaging algorithms and hardware
- A software engineering framework for imaging software development
- Numerous C programming examples
- Industry standard imaging and video compression algorithms
- Design methodology for high-performance imaging products
- Performance evaluation methods
- A review of current hardware imaging devices and trends
- Modern color imaging techniques
- Java and its implications for imaging
- Techniques to evaluate current imaging software and hardware products
WHAT THE EXPERTS SAY ABOUT THIS BOOK...
"... a great source on optimized implementations for high
performance computer imaging."
-William Pratt, author, Digital Image Processing
ABOUT THE AUTHOR...
Ihtisham Kabir is manager of the Imaging and Video software group at Sun Microsystems Computer Company. He has over twelve years of engineering experience in building high performance computer imaging products.