A key problem in practical image processing is the detection of specific features in a noisy image. Analysis of variance (ANOVA) techniques can be very effective in such situations, and this book gives a detailed account of the use of ANOVA in statistical image processing. The book begins by describing the statistical representation of images in the various ANOVA models. The authors present a number of computationally efficient algorithms and techniques to deal with such problems as line, edge, and object detection, as well as image restoration and enhancement. By describing the basic principles of these techniques, and showing their use in specific situations, the book will facilitate the design of new algorithms for particular applications. It will be of great interest to graduate students and engineers in the field of image processing and pattern recognition.

The European Conference on Computer Vision (ECCV) has established itself as a major event in this exciting and very active field of research and development. These refereed two-volume proceedings include the 123 papers accepted for presentation at the 4th ECCV, held in Cambridge, UK, in April 1996; these papers were selected from a total of 328 submissions and together give a well-balanced reflection of the state of the art in computer vision.
The papers in volume II are grouped in sections on color vision and shading; image features; motion; medical applications; tracking; applications and recognition; calibration, focus, and optics; applications; and structure from motion.

It is probably an overstatement to say that the discipline of telecommunication systems is becoming an application of digital signal processing (DSP). However, there is no doubt that by the mid-I980s integrated circuit technology has advanced to such an extent that revolutionary advances in telecommunications are fostered by the introduction of new and poweiful DSP algorithms. Actually, DSP has been recently playing a major role in the development of telecommuni cations systems: to name just one of the most widespread applications where this interaction has been most effective, we may mention the use of intelligent DSP to improve the peiformance of transmission systems by allowing sophisticated algorithm to be implemented in radio transmitters and receivers for personal communications. Other areas have equally benefited by the latest advances of DSP: speech coding and synthesis, speech recognition and enhancement, radar, sonar, digital audio, and remote sensing, just to cite afew. With this in mind, when choosing the topic for the 7th Tyrrhenian Workshop on Digital Communications, whose contributions are collected in this book, we aimed at focusing on the state of the art and the perspectives of the interaction between DSP and telecommunications, two disciplines that are becoming increasingly intertwined. Although by no means exhaustive of all the applications of DSP to telecommu nications, we believe that the material presented in this book pinpoints the most interesting among them, and hence it will be considered as a useful tool for investigating this complex and highly challenging field."

This book contains the invited reviewed session papers solicited for presentation as key papers at the Second Asian Conference on Computer Vision, ACCV '95, held in December 1995 in Singapore.
The fifteen leading researchers acting as session organizers at the ACCV '95 solicited and reviewed a unique collection of 56 invited papers presented in this volume. Together, these papers competently reflect the state-of-the-art developments in computer vision and image processing; the author index of the volume reads like a who-is-who in the field. The papers are devoted to both theoretical and practical issues relevant to researchers and professionals in the area.

This book constitutes the refereed proceedings of the 6th International Workshop on Discrete Geometry for Computer Imagery, DGCI'96, held in Lyon, France, in November 1996.
Computer imaging essentially depends on discrete models for coding, processing, recognition, representation, etc. The volume presents 24 revised full papers selected from 41 submissions together with 3 invited contributions and a tutorial paper, which bridges the gap between theory and practice. The issues addressed are topology, geometry, shape representation, 3D surfaces and volumes, models for discrete space, image transformation and generation.

This book presents the proceedings of the 8th International Conference on Image Analysis and Processing, ICIAP '95, held in Sanremo, Italy in September 1995 under the sponsorship of the International Association of Pattern Recognition IAPR.
The volume presents 108 papers selected from more than 180 submissions together with six invited contributions. The papers are written by a total of 265 contributing authors and give a comprehensive state-of-the-art report on all current issues of image analysis and processing. Theoretical aspects are addressed as well as systems design and advanced applications, particularly in medical imaging.

This book presents the proceedings of the Sixth International Conference on Computer Analysis of Images and Patterns, CAIP '95, held in Prague, Czech Republic in September 1995.
The volume presents 61 full papers and 75 posters selected from a total of 262 submissions and thus gives a comprehensive view on the state-of-the-art in computer analysis of images and patterns, research, design, and advanced applications. The papers are organized in sections on invariants, segmentation and grouping, optical flow, model recovery and parameter estimation, low level vision, motion detection, structure and matching, active vision and shading, human face recognition, calibration, contour, and sessions on applications in diverse areas.

Visualization is nowadays indispensable to get insight into the huge amounts of data pro duced by large scale simulations or advanced measurement devices. The use of com puter graphics for scientific purposes has become a well established discipline, known as Scientific Visualization. Many problems still have to be solved, and hence the field is a very active area for research and development. This book represents results of the sixth in a well established series of international workshops on Visualization in Scien tific Computing organized by the EUROGRAPHICS Association in collaboration with CRS4 (Center for Advanced Studies, Research and Development in Sardinia), held from May 3 to May 5,1995, in Chia, Italy. The thirteen contributions selected for this volume cover a wide range of topics, ranging from detailed algorithmic studies to searches for new metaphors. A rough di vision can be made into the parts interaction, irregular meshes, volume rendering, and applications. Interaction in three dimensions is a challenging area for research. The use of three dimensional user interfaces for more natural manipulation of three-dimensional data and their visualization is natural, but is far from trivial to realize. Pang et al. investigate the use of common objects such as spray cans and carving knives as metaphors for visualiza tion tools, in order to provide an intuitive and natural three dimensional user interface. Gibson uses a voxel-based data representation, not only for visualization, but also for physical modeling of objects. A prototype system under development for haptic explo ration is discussed."

Following five successful workshops in the previous five years, the Rendering Workshop is now well established as a major international forum and one of the most reputable events in the field of realistic image synthesis. Including the best 31 papers which were carefully evaluated out of 68 submissions the book gives an overview on hierarchical radiosity, Monte Carlo radiosity, wavelet radiosity, nondiffuse radiosity, and radiosity performance improvements. Some papers deal with ray tracing, reconstruction techniques, volume rendering, illumination, user interface aspects, and importance sampling. Also included are two invited papers by James Arvo and Alain Fournier. As is the style of the Rendering Workshop, the contributions are mainly of algorithmic nature, often demonstrated by prototype implementations. From these implementations result numerous color images which are included as appendix. The Rendering Workshop proceedings are certainly an obligatory piece of literature for all scientists working in the rendering field, but they are also very valuable for the practitioner involved in the implementation of state of the art rendering system certainly influencing the scientific progress in this field.

Recent developments in computer visualisation mean that it is now possible to combine computer-generated image sequences with real video, in real time, for broadcast quality production. This will not only revolutionise the broadcast industry, by making "electronic film sets" possible for example, but also has important implications for related fields such as virtual reality, multi-media, industrial vision, and medical image processing. This volume contains papers from the European Workshop on Combined Real and Synthetic Image Processing for Broadcast and Video Production, held in Hamburg, 23-24 November 1994. The papers cover three main aspects of research: hardware, image analysis, and image synthesis, and include several key contributions from the EU RACE II supported MONA LISA (MOdelling NAturaL Images for Synthesis and Animation) project. The resulting volume gives a comprehensive overview of this important area of research, and will be of interest to practitioners, researchers, and postgraduate students.

The European Conference on Computer Vision (ECCV) has established itself as a major event in this exciting and very active field of research and development. These refereed two-volume proceedings include the 123 papers accepted for presentation at the 4th ECCV, held in Cambridge, UK, in April 1996; these papers were selected from a total of 328 submissions and together give a well-balanced reflection of the state of the art in computer vision.
The papers in volume I are grouped in sections on structure from motion; recognition; geometry and stereo; texture and features; tracking; grouping and segmentation; stereo; and recognition, matching, and segmentation.

A realistic and comprehensive review of joint approaches to machine learning and signal processing algorithms, with application to communications, multimedia, and biomedical engineering systems Digital Signal Processing with Kernel Methods reviews the milestones in the mixing of classical digital signal processing models and advanced kernel machines statistical learning tools. It explains the fundamental concepts from both fields of machine learning and signal processing so that readers can quickly get up to speed in order to begin developing the concepts and application software in their own research. Digital Signal Processing with Kernel Methods provides a comprehensive overview of kernel methods in signal processing, without restriction to any application field. It also offers example applications and detailed benchmarking experiments with real and synthetic datasets throughout. Readers can find further worked examples with Matlab source code on a website developed by the authors. * Presents the necessary basic ideas from both digital signal processing and machine learning concepts * Reviews the state-of-the-art in SVM algorithms for classification and detection problems in the context of signal processing * Surveys advances in kernel signal processing beyond SVM algorithms to present other highly relevant kernel methods for digital signal processing An excellent book for signal processing researchers and practitioners, Digital Signal Processing with Kernel Methods will also appeal to those involved in machine learning and pattern recognition.

This work provides an instructive into applications and problems from the broad field of pattern recognition. It describes basic topics and the required mathematical background of image and speech processing. Algorithms and data structures for filtering, feature extraction, segmentation and classification are discussed, introducing and demonstrating different C++ concepts. The practice of object-oriented programming is illustrated by a step-wise development of a complete class library for image processing.

Digital signal processing has become more and more an integral part of observational seismology. While it offers unprecedented power in extracting information from seismic signals, it comes at the price of having to learn a variety of new skills. Dealing with digital seismic data requires at least a basic understanding of digital signal processing. Taking the calculation of true ground motion as the guiding problem, this course covers the basic theory of linear systems, the design and analysis of simple digital filters, the effect of sampling and A/D conversion and an introduction to spectral analysis of digital signals. It contains a number of examples and exercises that can be reproduced using the PITSA software package (Scherbaum and Johnson 1993) or similar programs.

This three-part monograph addresses topics in the areas of control systems, signal processing and neural networks. Procedures and results are determined which constitute the first successful synthesis procedure for associative memories by means of artificial neural networks with arbitrarily pre-specified full or partial interconnecting structure and with or without symmetry constraints for the connection matrix.

Over the last few years, multimedia hardware and applications have become widely available in the personal computer and workstation environments, and multimedia is rapidly becoming an integral part of stand-alone, single-user sy stems. In contrast, the problems encountered when moving to open, distributed environments are only just beginning to be identified. Examples are the trans mission of dynamic data (video, sound) over large distances and cooperative work. Following on from two successful workshops on multimedia, EG-MM '94 concentrates on topics related to multimedia/hypermedia in open, distributed environments. The goal of this symposium was not only to give a comprehensive overview of the current state of research, development, and standardisation in the field, but also to provide an opportunity for live demonstrations to experience directly the presented results. The symposium program consists of two invited keynote speeches, eight tech .nical sessions, one tutorial, and one demonstration session. A workshop following immediately after the symposium provides an opportunity for in-depth discussi ons of open problems among experts. It is intended to fill a gap often experienced at larger meetings and conferences: the lack of time to discuss in detail issues raised during the event, such as the characteristics of different approaches to a certain problem. It is intended that the results of the workshop be published as a Eurographics Technical Report."

Image sequence processing is becoming a tremendous tool to analyze spatio-temporal data in all areas of natural science. It is the key to studythe dynamics of of complex scientific phenomena. Methods from computer science and the field of application are merged establishing new interdisciplinary research areas. This monograph emerged from scientific applications and thus is an example for such an interdisciplinaryapproach. It is addressed both to computer scientists and to researchers from other fields who are applying methods of computer vision. The results presented are mostly from environmental physics (oceanography) but they will be illuminating and helpful for researchers applying similar methods in other areas.

Computational geometry is the part of theoretical computer science that concerns itself with geometrical objects; it aims to define efficient algorithms for problems involving points, lines, polygons, and so on. The field has gained popularity very rapidly during the last decade. This is partly due to the many application areas of computational geometry and partly due to the beauty of the field itself. This monograph focuses on three problems that arise in three-dimensional computational geometry. The first problem is the ray shooting problem: preprocess a set of polyhedra into a data structure such that the first polyhedron that is hit by a query ray can be determined quickly. The second problem is that of computing depth orders: we want to sort a set of polyhedra such thatif one polyhedron is (partially) obscured by another polyhedron then it comes first in the order. The third problem is the hidden surface removal problem: given a set of polyhedra and a view point, compute which parts of the polyhedra are visible from the view point. These three problems involve issues that are fundamental to three-dimensional computational geometry. The book also contains a large introductory part discussing the techniques used to tackle the problems. This part should interest not only those who need the background for the rest of the book but also anyone who wants to know more about some recent techniques in computational geometry.

This workshop on Graphics Modeling and Visualization in Scientific, Engineering and Technical Applications was held in Darmstadt, Germany, on 13-14 April 1992. Visualization is known as the key technology to control massive data sets and to achieve insight into these tera bytes of data. Graphics Modeling is the enabling technology for advanced interaction. This book contains the keynote papers from three internationally well-known invited speakers and a selection of papers submitted to this workshop. Due to the effective scientific contacts between German and Portuguese researchers, and the results from this cooperation, the workshop was also the 2nd Luso-German Meet- ing on Computer Graphics. The keynote papers present an excellent overview of the main topics of the workshop. Prof. Rae Earnshaw focuses on Scientific Visualization as an inter- disciplinary area of research and application, presenting state of the art concepts and works currently being developed, and the requirements of visualization sys- tems in the 90's. Prof. Frank-Lothar Krause presents Product Modeling as the key for information integration in industry, in particular to support CAD-CAM inte- gration and CIM. Dr. Rolf Iindner introduces the DEDICATED (DEvelopment of a new DImension in Computer Assisted Teaching and EDucation) project, a European DELTA project with a strong involvement of German and Portuguese partners.

There exists a large variety of image reconstruction methods proposed by different authors (see e. g. Pratt (1978), Rosenfeld and Kak (1982), Marr (1982)). Selection of an appropriate method for a specific problem in image analysis has been always considered as an art. How to find the image reconstruction method which is optimal in some sense? In this book we give an answer to this question using the asymptotic minimax approach in the spirit of Ibragimov and Khasminskii (1980a, b, 1981, 1982), Bretagnolle and Huber (1979), Stone (1980, 1982). We assume that the image belongs to a certain functional class and we find the image estimators that achieve the best order of accuracy for the worst images in the class. This concept of optimality is rather rough since only the order of accuracy is optimized. However, it is useful for comparing various image reconstruction methods. For example, we show that some popular methods such as simple linewise processing and linear estimation are not optimal for images with sharp edges. Note that discontinuity of images is an important specific feature appearing in most practical situations where one has to distinguish between the "image domain" and the "background" . The approach of this book is based on generalization of nonparametric regression and nonparametric change-point techniques. We discuss these two basic problems in Chapter 1. Chapter 2 is devoted to minimax lower bounds for arbitrary estimators in general statistical models.

Autonomous mobile systems (AMS) are systems capable of some mobility and equipped with advanced sensor devices in order to flexibly respond to changing environmental situations, thus achieving some degree of autonomy. The purpose of this book is to contribute to some essential topics in this broad research area related to sensing and control, but not to present a complete design of an AMS. Subjects conceming knowledge based control and decision, such as moving around obstacles, task planning and diagnosis are left for future publications in this series. Research in the area of AMS has grown rapidly during the last decade, see e.g. WAXMAN et al. 87], DICKMANNS, ZAPP 87]. The requirements of an AMS strongly depends on the desired tasks the system should execute, its operational environment and the expected speed of the AMS. For instance, road vehicles obtain velocities of 10 m/s and more, therefore the processing of sensor data such as video image sequences has to be very fast and simple, while indoor mobile robots deal with shorter distances and lower speeds, thus more sophistcated techniques are applicable and -as is done in our approach- additional sensors can be integrated to allow for multi sensor processing.

This volume contains the papers selected for presentation at the Second International Conference on Parallel Image Analysis (ICPIA '92), held in Ube, Japan, December 21-23, 1992. The conference topics are data structures, parallel algorithms and architectures, neural networks, computational vision, syntactic generation and recognition, and multidimensional models. The first meeting with these topics was theInternational Colloquium on Parallel Image Processing, which took place in Paris in June 1991. The aim of the meetings is to bring together specialistsfrom various countries who are interested in the topics and to stimulatetheoretical and practical research in the field of parallel image processingand analysis. The volume contains three invited papers, a summary of a tutorial lecture, and twenty selected and refereed communications.

This textbook gives details of recent developments in the field of image processing, machine vision and analysis. Based on the original book published in Czech, this English edition has been expanded to include 3D vision, neural networks and invariants.

This volume constitutes the proceedings of the 5th International Conference on Computer Analysis of Images and Patterns (CAIP'93), held in Budapest, Hungary, in September 1993. Formerly, the events in this biennial conference series were thought as a forum where East European researchers and professionals from academia and industry had an opportunity to discuss their results and ideas with Western colleagues active in image processing and pattern recognition. Now, CAIP'93 has a much more international scope, and in the future these conferences will not any longertake place only in East European countries, but roam throughout whole Europe. Besides invited talks by Belikova, Gimel'farb, Haralick and Roska, the volume contains 114 contributions, either presented as lectures or posters and carefully selected by a highly competent international program committee from a total of some 230 submissions; thus the book gives a thorough survey on recent research results and their applications in image processing and pattern recognition. The proceedings is organized in 20 sections, for example on image data structures, image processing, edges and contours, Hough transforms and related methods, shape, motion, 3-D vision, character recognition and document processing, biomedical applications, industrial applications, and neural networks.

This book contains the 61 papers that were accepted for presenta tion at the 1992 British Machine Vision Conference. Together they provide a snapshot of current machine vision research throughout the UK in 24 different institutions. There are also several papers from vision groups in the rest of Europe, North America and Australia. At the start of the book is an invited paper from the first keynote speaker, Robert Haralick. The quality of papers submitted to the conference was very high and the programme committee had a hard task selecting around half for presentation at the meeting and inclusion in these proceedings. It is a positive feature of the annual BMV A conference that the entire process from the submission deadline through to the conference itself and publication of the proceedings is completed in under 5 months. My thanks to members of the programme committee for their essential contribution to the success of the conference and to Roger Boyle, Charlie Brown, Nick Efford and Sue Nemes for their excellent local organisation and administration of the conference at the University of Leeds."