Издательство Thomson Publishing, 1999, -446 pp.Unfortunately the preface has to start sadly. Prof. Besslich died during the preparation stage of this book. As with all his colleagues, I deeply regret that he did not live to see his work completed. I have done my best to realize the project as he would have wished.
The project's name is Ad Oculos which means 'to the eyes'. It is intended to give any interested beginner a clear demonstration of digital image processing which is used for such tasks as the enhancement of noisy images, for identifying characters, for checking the measurements of workpieces, for inspecting the garnishment of cookies, for analyzing cellular substances (for example, biopsies), for detecting environmental pollution from aerial photographs, and so on.
Since the mid-sixties such digital image processing has become an increasingly important subject of scientific research. Due to the enormous computing power required by image processing algorithms, for several years it remained a research area of very few specialists but with the dramatic increase in desk top computing power over the last few years, more and more people have become interested in 'making the computer see'. Thus, the number of publications concerning image processing has risen considerably in recent years. This development has been paralleled by the entry of digital image processing into the industrial arts: the number of companies trying to make money from industrial image processing has risen sharply due to optimistic market prognoses. Nevertheless, digital image processing still seems to be a 'playground' for specialists, when compared to other new techniques such as GAD. Consequently, it is not at all easy for beginners to enter this discipline. The aim of this book is to give a helping hand to those who want to get started.
The central idea behind this book is to demonstrate classical image processing algorithms regarding: their basic idea from an application-oriented point of view, their realization from the programmers point of view, and (if necessary) their background from a theoretical point of view.
Consequently the basic intention of this work is by no means to be complete or scientifically up-to-date but to give the reader the seed for fruitful applications. The prerequisite for a good harvest is to fully grasp the subject material. It is no coincidence that to grasp means to touch it and to understand it. Thus the book is accompanied by Ad Oculos, a Windows 3.1 program which contains the algorithms discussed in the book. Due to its graphical user interface it allows you to grasp the individual algorithms easily and to combine these algorithms to form complex processing chains. Since this book aims to be relevant for practical utilization it was very important to illustrate the gap between the theoretical possibilities of digital image processing and the harsh realities of industrial image processing. For this purpose the appendix contains a chapter illustrating examples of industrial applications.Introduction
Point operations
Local operations
Global operations
Region-oriented segmentation
Contour-oriented segmentation
Hough transform
Morphological image processing
Texture analysis
Pattern recognition
Image sequence analysis
A: Examples of industrial image processing
B: General purpose procedures
C: Calculus of variations
D: Rules for integration
E: Taylor series expansion/total differential
F: Gauss-Seidel iteration
F: Gauss-Seidel iteration
F: Gauss-Seidel iteration
