Azriel Rosenfeld - Böcker

The rapid rate at which the field of digital picture processing has grown in the past five years had necessitated extensive revisions and the introduction of topics not found in the original edition.

The technical ability to generate volumes of digital multimedia data is becoming increasingly "mainstream" in today's electronic world. Online services create volumes of primarily textual information, such as news reports, product reviews, and e-mail chronicles. Advances in digital video technology have given organizations the capability to amass visual records and produce collections of surveillance monitoring data streams. With this ability to generate and archive volumes of data comes the potential of deriving or recalling information and knowledge from these data histories. To effectively utilize the growing number of multimedia data repositories, there is a convergence in technologies from large-scale data management, semantic-oriented media (text, image, and video) understanding, and multi-source trend analysis. This convergence is not straightforward and introduces a significant challenge in construction solutions that offer scalable deployment with semantically rich quality.The Microelectronics and Computer Technology Corporation (MCC) and its member companies carried out a study in 1997 to investigate the state of the art in technologies for annotating and manipulating large-scale networks of multimedia information objects with content-based concepts. This book documents the study's technology assessment and identifies shortcomings where further research and integration of technologies are needed to meet anticipated application requirements. The major points highlighted in this book can be used as cornerstones for defining advanced research and development directions, and opportunities to exploit the content available in networks of large-scale multi-media sources. Based on the results of the study, MCC initiated the Content-Based Access to Multimedia (CBAM) Information project to investigate semantically-oriented access to large-scale image and video repositories. The project focuses on concept extraction, annotation, and collection principles applied in and across large-scale image and video repositories.It demonstrates proof-of-concept environments where multimedia objects acquire semantic content annotations and become elements exploited in distributed information-gathering applications.

Biological visual systems employ parallel processing to perform real-world visual tasks in real time. A key to this performance seems to be that biological systems construct representations of their visual image data at multiple scales. This text describes a multiscale, or "pyramid", approach to vision, including its theoretical foundations, a set of pyramid-based modules for image processing, object detection, texture discrimination, contour detection and processing, feature detection and description, and motion detection and tracking. It also shows how these modules can be efficiently implemented on hypercube-connected processor networks. The text is intended for both students of vision and vision system designers. It provides a general approach to vision systems design as well as a set of robust, efficient vision modules.

This text aims to be a reference for the practitioners and academicians in the fields of multimedia search engines. Half a terabyte or 9000 hours of motion pictures are produced around the world every year. Furthermore, 3000 television stations broadcasting for twenty-four hours a day produce eight million hours per year, amounting to 24,000 terabytes of data. Although some of the data is labelled at the time of production, an enormous portion remains unindexed. For practical access to such huge amounts of data, there is a great need to develop efficient tools for browsing and retrieving content of interest, so that producers and end users can quickly locate specific video sequences in this ocean of audio-visual data. This text describes the main techniques being developed by the major players in industry and academic research to address this problem.

Video Mining is an essential reference for the practitioners and academicians in the fields of multimedia search engines. Half a terabyte or 9,000 hours of motion pictures are produced around the world every year. Furthermore, 3,000 television stations broadcasting for twenty-four hours a day produce eight million hours per year, amounting to 24,000 terabytes of data. Although some of the data is labeled at the time of production, an enormous portion remains unindexed. For practical access to such huge amounts of data, there is a great need to develop efficient tools for browsing and retrieving content of interest, so that producers and end users can quickly locate specific video sequences in this ocean of audio-visual data. Video Mining is important because it describes the main techniques being developed by the major players in industry and academic research to address this problem. It is the first time research from these leaders in the field developing the next-generation multimedia search engines is being described in great detail and gathered into a single volume.Video Mining will give valuable insights to all researchers and non-specialists who want to understand the principles applied by the multimedia search engines that are about to be deployed on the Internet, in studios' multimedia asset management systems, and in video-on-demand systems.

Biological visual systems employ massively parallel processing to perform real-world visual tasks in real time. A key to this remarkable performance seems to be that biological systems construct representations of their visual image data at multiple scales. A Pyramid Framework for Early Vision describes a multiscale, or `pyramid', approach to vision, including its theoretical foundations, a set of pyramid-based modules for image processing, object detection, texture discrimination, contour detection and processing, feature detection and description, and motion detection and tracking. It also shows how these modules can be implemented very efficiently on hypercube-connected processor networks. A Pyramid Framework for Early Vision is intended for both students of vision and vision system designers; it provides a general approach to vision systems design as well as a set of robust, efficient vision modules.

In the past five years, the field of electrostatic discharge (ESD) control has under­ gone some notable changes. Industry standards have multiplied, though not all of these, in our view, are realistic and meaningful. Increasing importance has been ascribed to the Charged Device Model (CDM) versus the Human Body Model (HBM) as a cause of device damage and, presumably, premature (latent) failure. Packaging materials have significantly evolved. Air ionization techniques have improved, and usage has grown. Finally, and importantly, the government has ceased imposing MIL-STD-1686 on all new contracts, leaving companies on their own to formulate an ESD-control policy and write implementing documents. All these changes are dealt with in five new chapters and ten new reprinted papers added to this revised edition of ESD from A to Z. Also, the original chapters have been augmented with new material such as more troubleshooting examples in Chapter 8 and a 20-question multiple-choice test for certifying operators in Chapter 9. More than ever, the book seeks to provide advice, guidance, and practical ex­ amples, not just a jumble of facts and generalizations. For instance, the added tailored versions of the model specifications for ESD-safe handling and packaging are actually in use at medium-sized corporations and could serve as patterns for many readers.

The generic term "graph-grammars" refers to a variety of methods for specifying (possibly infinite) sets of graphs or sets of maps. The area of graph-grammars originated in the late 60s motivated by considerations concerning pattern recognition - since then the list of areas which have interacted with the development of graph-grammars has grown quite impressively. It includes pattern recognition, software specification and development, VLSI layout schemes, data bases, lambda-calculus, analysis of concurrent systems, massively parallel computer architectures, incremental compilers, computer animation, complexity theory, developmental biology, music composition, representation of physical solids, and many others. This volume is based on the contributions presented at the third international workshop on graph-grammars and their applications, held in Warrenton, Virginia, USA in December 1986. Aiming at the best possible representation of the field not all of the papers presented at the meeting appear in this volume and some of the papers from this volume were not presented at the workshop. The volume consists of two parts: Part I presents tutorial introductions to a number of basic graph and map rewriting mechanisms. Part II contains technical contributions. This collection of papers provides the reader with an up-to-date overview of current trends in graph-grammars.

This book constitutes the refereed proceedings of the Third International Conference on Machine Learning and Data Mining in Pattern Recognition, MLDM 2003, held in Leipzig, Germany, in July 2003. The 33 revised full papers presented together with two invited papers were carefully reviewed and selected from 75 submissions. The papers are organized in topical sections on decision trees; clustering and its applications; support vector machines; case-based reasoning; classification, retrieval, and feature Learning; discovery of frequent or sequential patterns; Bayesian models and methods; association rule mining; and applications.

This book constitutes the refereed proceedings of the 6th International Workshop on Structural and Syntactical Pattern Recognition, SSPR '96, held in Leipzig, Germany in August 1996.The 36 revised full papers included together with three invited papers were carefully selected from a total of 52 submissions. The papers are organized in topical sections on grammars and languages; morphology and mathematical approaches to pattern recognition; semantic nets, relational models and graph-based methods; 2D and 3D shape recognition; document image analysis and recognition; and handwritten and printed character recognition.

A NATO advanced Study Institute took place at Bonas from June 14th to June 25th 1976 on "Digital Image Processing and Analysis". This book is the lasting result of a successful meeting, where the best specialists of the field could exchange their ideas and results. The papers are arranged so as to present first the more general and tutorial articles and then the more specific ones on applications. The general topics cover two dimensional transforms, techniques of image restoration, recursive filters, segmentation and analysis of image parts, some points of view from psychology and physiology, and problems of software and processing. The application fields concerned are remote sensing, medical applications, TV image compression,and optical character recognition. The editors wish to thank the Scientific Affairs Division of NATO for the edition of this book. Acknowledgment: This ASI has been made possible by the financial support of the NATO Scientific Affairs Division and D. R. M. E. and the material support of IRIA and the Institut de Programmation. VII TABLE OF CONTENTS William K. Pratt Two dimensional unitary transforms 1 T. S. Huang Two-dimensional Fourier transform 23 T. S. Huang Algebraic methods of image restoration 41 S. Castan Image enhancement and restoration 47 T. S. Huang Film grain noise 63 K. G. Beauchamp Two-dimensional recursive digital filtering 69 S. Attasi A new approach to 2D-recursive filtering 81 V. Cappellini Some efficient two-dimensional recursive digital filters 87 T. S. Durrani and C. E.

The research and development of multi-dimensional pattern recognition, scene analysis, computer vision and image processing have progressed very rapidly in recent years. Among various models employed for pattern representation and analysis, the array grammar has attracted more and more attention because it has several advantages over others.This special volume, perhaps the first time ever in the literature, is a collection of 14 papers by prominent professionals and experts, aimed at promoting array grammars, patterns and recognizers. They are grouped in the following categories: (1) Array grammars and pattern generation, (2) Array pattern recognizers, (3) Coordinate grammars and L-systems, and (4) Hexagonal grids, tilings and encryption.