Rosandich R.G. Intelligent Visual Inspection. Using Artificial Neural Networks

Chapman & Hall, 1996. — 311 p.Over the past several years a great deal of research has been carried out in the areas of artificial vision and neural networks. Although much of this research has been theoretical in nature, many of the techniques developed through these efforts are now mature enough to be used in practical applications.
Concurrently, the opening of worldwide markets to many products has forced manufacturing companies to compete on a global basis in recent years. This high level of competition between manufacturers has led to rapid developments in the areas of computer-integrated manufacturing, flexible manufacturing, agile manufacturing and intelligent manufacturing. These developments have in turn generated a need for intelligent sensing and decision making systems capable of automatically performing many tasks traditionally done by human beings. Visual inspection is one such task, and there is a need for effective automated visual inspection systems in today's competitive manufacturing environment.
In this book, the author seeks to accomplish two things. First, to collect and organize information in the artificial vision and neural network fields and to present that information in an informative, clear manner without excess mathematical detail. Second, to help bridge the gap between the theoretical research that has been done and the present needs of manufacturing companies by demonstrating how several recently developed techniques can be implemented via existing hardware to solve practical automated inspection problems. To this end the book is organized into four sections. Section One introduces and defines the problem of intelligent visual inspection in manufacturing environments.
Section Two reviews fundamental research in the areas of artificial vision and neural networks. Section Three covers in detail the practical design of artificial vision systems based on neural networks, and Section Four presents case studies which show how these systems can be applied to current and future real-world inspection problems. This book is intended to appeal to electrical, mechanical, industrial and manufacturing engineers, computer scientists, technicians, and managers employed by manufacturing companies that are interested in the potential of intelligent visual inspection systems. It should also appeal to manufacturers of machine vision systems, and to consulting engineers and system integrators engaged in the design and installation of such systems.
The book is intended for academic appeal as well. It should serve as a reference book for researchers and graduate students interested in the areas of artificial vision or neural networks, and it should also be a valuable reference book for courses with content related to these topics. Such courses could be part of a curriculum in computer science or electrical, mechanical, manufacturing, or industrial engineering. Finally, management programs may also be interested in the book as a reference for graduate courses in quality control.Intelligent manufacturing
Intelligent visual inspection
Fundamentals of Artificial Vision Systems
Biological vision systems
Artificial neural networks for pattern recognition
Artificial Vision Systems Design
Image acquisition and storage
Low-level image processing
Intermediate image processing
Computational approach to artificial vision
Connectionist approach to artificial vision
Experimental evaluation of the CAMERA vision model
Case Studies
Automated visual inspection systems
Future of automated visual inspection
A: Applications of machine vision systems for inspection in manufacturing environments
B: Proof of convergence of the learning algorithm employed in the HAVNET artificial neural network