Document image analysis is the automatic computer interpretation of images of printed and handwritten documents, including text, drawings, maps, music scores, etc. Research in this field supports a rapidly growing international industry. This is the first book to offer a broad selection of state-of-the-art research papers, including authoritative critical surveys of the literature, and parallel studies of the architectureof complete high-performance printed-document reading systems. A unique feature is the extended section on music notation, an ideal vehicle for international sharing of basic research. Also, the collection includes important new work on line drawings, handwriting, character and symbol recognition, and basic methodological issues. The IAPR 1990 Workshop on Syntactic and Structural Pattern Recognition is summarized, including the reports of its expert working groups, whose debates provide a fascinating perspective on the field. The book is an excellent text for a first-year graduate seminar in document image analysis, and is likely to remain a standard reference in the field for years.

Thirty years ago pattern recognition was dominated by the learning machine concept: that one could automate the process of going from the raw data to a classifier. The derivation of numerical features from the input image was not considered an important step. One could present all possible features to a program which in turn could find which ones would be useful for pattern recognition. In spite of significant improvements in statistical inference techniques, progress was slow. It became clear that feature derivation was a very complex process that could not be automated and that features could be symbolic as well as numerical. Furthennore the spatial relationship amongst features might be important. It appeared that pattern recognition might resemble language analysis since features could play the role of symbols strung together to form a word. This led. to the genesis of syntactic pattern recognition, pioneered in the middle and late 1960's by Russel Kirsch, Robert Ledley, Nararimhan, and Allan Shaw. However the thorough investigation of the area was left to King-Sun Fu and his students who, until his untimely death, produced most of the significant papers in this area. One of these papers (syntactic recognition of fingerprints) received the distinction of being selected as the best paper published that year in the IEEE Transaction on Computers. Therefore syntactic pattern recognition has a long history of active research and has been used in industrial applications.

A sharp increase in the computing power of modern computers has triggered the development of powerful algorithms that can analyze complex patterns in large amounts of data within a short time period. Consequently, it has become possible to apply pattern recognition techniques to new tasks. The main goal of this book is to cover some of the latest application domains of pattern recognition while presenting novel techniques that have been developed or customized in those domains.

This book presents novel graph-theoretic methods for complex computer vision and pattern recognition tasks. It presents the application of graph theory to low-level processing of digital images, presents graph-theoretic learning algorithms for high-level computer vision and pattern recognition applications, and provides detailed descriptions of several applications of graph-based methods to real-world pattern recognition tasks.

Many business decisions are made in the absence of complete information about the decision consequences. Credit lines are approved without knowing the future behavior of the customers; stocks are bought and sold without knowing their future prices; parts are manufactured without knowing all the factors affecting their final quality; etc. All these cases can be categorized as decision making under uncertainty. Decision makers (human or automated) can handle uncertainty in different ways. Deferring the decision due to the lack of sufficient information may not be an option, especially in real-time systems. Sometimes expert rules, based on experience and intuition, are used. Decision tree is a popular form of representing a set of mutually exclusive rules. An example of a two-branch tree is: if a credit applicant is a student, approve; otherwise, decline. Expert rules are usually based on some hidden assumptions, which are trying to predict the decision consequences. A hidden assumption of the last rule set is: a student will be a profitable customer. Since the direct predictions of the future may not be accurate, a decision maker can consider using some information from the past. The idea is to utilize the potential similarity between the patterns of the past (e.g., "most students used to be profitable") and the patterns of the future (e.g., "students will be profitable").

Networks have become nearly ubiquitous and increasingly complex, and their support of modern enterprise environments has become fundamental. Accordingly, robust network management techniques are essential to ensure optimal performance of these networks. This monograph treats the application of numerous graph-theoretic algorithms to a comprehensive analysis of dynamic enterprise networks. Network dynamics analysis yields valuable information about network performance, efficiency, fault prediction, cost optimization, indicators and warnings. Based on many years of applied research of generic network dynamics, this work covers a number of elegant applications (including many new and experimental results) of traditional graph theory algorithms and techniques to computationally tractable network dynamics analysis to motivate network analysts, practitioners and researchers alike. The material is also suitable for graduate courses addressing state-of-the-art applications of graph theory in analysis of dynamic communication networks, dynamic databasing, and knowledge management.

This book constitutes the thoroughly refereed post-proceedings of the International Workshop on Sensor Based Intelligent Robots held in Dagstuhl Castle, Germany, in October 2000.The 20 revised full papers were carefully reviewed and improved for inclusion in this book. Addressing a broad variety of aspects of the highly interdisciplinary field of robotics, the book presents three topical sections on sensing, robotics, and intelligence.

Many business decisions are made in the absence of complete information about the decision consequences. Credit lines are approved without knowing the future behavior of the customers; stocks are bought and sold without knowing their future prices; parts are manufactured without knowing all the factors affecting their final quality; etc. All these cases can be categorized as decision making under uncertainty. Decision makers (human or automated) can handle uncertainty in different ways. Deferring the decision due to the lack of sufficient information may not be an option, especially in real-time systems. Sometimes expert rules, based on experience and intuition, are used. Decision tree is a popular form of representing a set of mutually exclusive rules. An example of a two-branch tree is: if a credit applicant is a student, approve; otherwise, decline. Expert rules are usually based on some hidden assumptions, which are trying to predict the decision consequences. A hidden assumption of the last rule set is: a student will be a profitable customer. Since the direct predictions of the future may not be accurate, a decision maker can consider using some information from the past. The idea is to utilize the potential similarity between the patterns of the past (e.g., "most students used to be profitable") and the patterns of the future (e.g., "students will be profitable").

Robotics is a highly interdisciplinary research topic, that requires integ- tion of methods for mechanics, control engineering, signal processing, planning, graphics, human-computer interaction, real-time systems, applied mathematics, and software engineering to enable construction of fully operational systems. The diversity of topics needed to design, implement, and deploy such systems implies that it is almost impossible for individual teams to provide the critical mass required for such endeavours.To facilitate interaction and progresson s- sor based intelligent robotics the organisation of inter-disciplinary workshops is necessarythrough which in-depth discussion can be used for cross dissemination between di?erent disciplines. The Dagstuhlfoundation has organiseda number of workshopsonModelling and Integration of Sensor Based Intelligent Robot Systems. The Dagstuhl s- inars are all organised over a full week in a beautiful setting in the Saarland in Germany. The setting provides an ideal environment for in-depth presen- tions and rich interactions between the participants. This volume contains the papers presented during the third workshopheld over the period September 28 - October 2, 1998. All papers have been reviewed by one-three reviewers over a relativly short period. We wish to thank all the reviewers for their invaluable help in making this a high quality selection of papers. We gratefully acknowledge the support of the Schloss Dagstuhl Foundation and the sta? at Springer-Verlag. Without their support the production of this volume would not have been possible.

Dieses Buch bietet eine systematische Einfuhrung in dieses neue Arbeitsfeld der automatischen Bildanalyse. Es behandelt samtliche wichtigen Teilaspekte, beginnend mit der Gewinnung von Tiefenbildern durch passive und aktive Verfahren uber die Extraktion charakteristischer Flachenmerkmale und die Segmentierung bis hin zur modellbasierten Objekterkennung. Daneben werden konkrete Anwendungen der Tiefenbildanalyse vorgestellt. Die Didaktik des Buches erlaubt es Forschern und Praktikern, sich selbstandig in dieses Gebiet einzuarbeiten. Das teilweise schwer zugangliche Material wurde in einheitlicher Notation und verstandlicher Form aufbereitet. Die beschriebenen Verfahren koennen damit leicht auf dem Computer implementiert werden. Die Literaturhinweise geben einen vollstandigen UEberblick uber die aktuelle Forschung.

This book presents novel graph-theoretic methods for complex computer vision and pattern recognition tasks. It presents the application of graph theory to low-level processing of digital images, presents graph-theoretic learning algorithms for high-level computer vision and pattern recognition applications, and provides detailed descriptions of several applications of graph-based methods to real-world pattern recognition tasks.

Dieses Buch bietet eine systematische Einfuhrung in dieses neue Arbeitsfeld der automatischen Bildanalyse. Es behandelt samtliche wichtigen Teilaspekte, beginnend mit der Gewinnung von Tiefenbildern durch passive und aktive Verfahren uber die Extraktion charakteristischer Flachenmerkmale und die Segmentierung bis hin zur modellbasierten Objekterkennung. Daneben werden konkrete Anwendungen der Tiefenbildanalyse vorgestellt. Die Didaktik des Buches erlaubt es Forschern und Praktikern, sich selbstandig in dieses Gebiet einzuarbeiten. Das teilweise schwer zugangliche Material wurde in einheitlicher Notation und verstandlicher Form aufbereitet. Die beschriebenen Verfahren koennen damit leicht auf dem Computer implementiert werden. Die Literaturhinweise geben einen vollstandigen UEberblick uber die aktuelle Forschung.

Der Band enthAlt die VortrAge, die auf dem 10. DAGM-Symposium Ende September 1988 in ZA1/4rich gehalten wurden. Die DAGM veranstaltet seit 1978 jAhrlich an verschiedenen Orten ein wissenschaftliches Symposium mit dem Ziel, Aufgabenstellungen, Denkweisen und Forschungsergebnisse aus verschiedenen Gebieten der Mustererkennung vorzustellen, den Erfahrungs- und Ideenaustausch zwischen den Fachleuten anzuregen und den Nachwuchs zu fArdern. Die BeitrAge zum Symposium kommen aus dem gesamten deutschen Sprachraum und darA1/4ber hinaus. Die DAGM ist Mitglied der International Association for Pattern Recognition (IAPR).

Der vorliegende Band behandelt Verfahren der Kunstlichen Intelligenz (KI) in der Bild- und Sprachanalyse, also in einem Teilgebiet der Mustererkennung (ME). Die Definition und Abgrenzung von Begriffen wie KI und ME wird in der Literatur nicht einheitl ich gehandhabt; es ist aber wichtig daran zu erinnern, dass beide aus ihrer Fruhzeit gemeinsame Wurzeln haben. Die Fahigkeit zur Erkennung von Mustern, und ganz allgemein zur Wahrnehmung der Umwelt mit geeigneten Sensoren, wurde als wesentl i che Vorau ssetzung fur autonom agi erende "i nte 11 i gente" tech nische Systeme angesehen. Einerseits wurde in ersten Veroffentlichungen uber KI immer wieder die ME als ein zu losendes Problem genannt, andererseits wurde als wichtiges Problem in der ME die Einbeziehung von Wissen uber das Problem und die zu erkennenden Muster gefordert. Die Wichtigkeit der Wissensverarbeitung in der ME lasst sich in Forschungsantragen schon aus dem Ende der funfziger Jahre nach lesen. Zu dieser Zeit war der Stand der ME naturl ich noch nicht so weit ent wickelt wie heute, er stand praktisch auf der Stufe der Extraktion und Klassifi kation von Merkmalvektoren. Leider wird immer noch vielfach ME mit diesem An fangsstand der Technik verwechselt und nicht die generelle Aufgabe, namlich die automatische Verarbei tung, Auswertung und Interpretation sensorischer Informa tion gesehen. Inzwi schen sind auf dem Sektor der Wi ssensverarbeitung in der KI solche Fortschritte erzielt worden, dass man in der ME die Nutzung von Wi ssen nicht mehr nur fordern sondern tatsachl ich auch durchfuhren kann."

Wissensbasierte Verarbeitung von Daten, ein Teilgebiet der kunst lichen Intelligenz innerhalb der Informatik, gewinnt zunehmend an Bedeutung. Die Anfange wissensbasierter Systeme und kunstlicher Intelligenz reichen bis in die 50er Jahre zuruck und somit hat dieses Gebiet - gemessen an der Geschichte der Informatik - eine lange Tradition. Seit etwa 1980 ist jedoch eine starke Zunahme des Interesses an Fragen zu wissensbasierten Systemen festzustel len. Diese Entwicklung ist u.a. begrundet durch zahlreiche For schungsprogramme, die weltweit ins Leben gerufen wurden. Systeme zur modellgesteuerten Verarbeitung und Analyse von Bil dern stellen eine interessante und wichtige Teilklasse allgemei ner wissensbasierter Systeme dar. Derartige Systeme stehen im Mittelpunkt der vorliegenden Publikation. Das Buch gliedert sich in zwei Teile. Der erste Teil behandelt die wissensbasierte Bildanalyse unter allgemeinen Gesichtspunkten in breitem Rahmen. Hierbei wird auch auf Methoden zur Extraktion elementarer Bild bestandteile eingegangen, d.h. auf Vorverarbeitung und Segmen tierung. lm zweiten Teil erfolgt die detaillierte Beschreibung eines unter der Leitung des Autors an der Universitat Erlangen NUrnberg entwickelten wissensbasierten Systems zur Bildanalyse. Der Anwendungsbereich dieses Systems liegt in der Medizin, ge nauer gesagt in der Herzdiagnostik. Die verwendeten Methoden sind jedoch zum uberwiegenden Teil so allgemein, dass das vorge steIlte System als eine prototypische Realisierung eines allge meinen, wissensbasierten Bildanalysesystems verstanden werden kann."

This compendium is a completely revised version of an earlier book, Data Mining in Time Series Databases, by the same editors. It provides a unique collection of new articles written by leading experts that account for the latest developments in the field of time series and data stream mining.The emerging topics covered by the book include weightless neural modeling for mining data streams, using ensemble classifiers for imbalanced and evolving data streams, document stream mining with active learning, and many more. In particular, it addresses the domain of streaming data, which has recently become one of the emerging topics in Data Science, Big Data, and related areas. Existing titles do not provide sufficient information on this topic.

This book is concerned with a fundamentally novel approach to graph-based pattern recognition based on vector space embedding of graphs. It aims at condensing the high representational power of graphs into a computationally efficient and mathematically convenient feature vector.

This volume utilizes the dissimilarity space representation originally proposed by Duin and Pekalska to embed graphs in real vector spaces. Such an embedding gives one access to all algorithms developed in the past for feature vectors, which has been the predominant representation formalism in pattern recognition and related areas for a long time.

This book is a collection of scientific papers published during the last five years, showing a broad spectrum of actual research topics and techniques used to solve challenging problems in the areas of computer vision and image analysis. The book will appeal to researchers, technicians and graduate students.

An inadequate infrastructure for software testing is causing major losses to the world economy. The characteristics of software quality problems are quite similar to other tasks successfully tackled by artificial intelligence techniques. The aims of this book are to present state-of-the-art applications of artificial intelligence and data mining methods to quality assurance of complex software systems, and to encourage further research in this important and challenging area.

http://www.worldscientific.com/worldscibooks/10.1142/1960

The widespread use of bankchecks in daily life makes the development of check-reading systems of fundamental relevance to banks and other financial institutions. This will improve productivity and allow advanced customer services. Therefore, many industrial companies and academic research laboratories have recently been attracted to this field, which involves several aspects, like image acquisition and preprocessing, layout analysis, preprinted data identification and recognition, user-entered data extraction, recognition of handwritten characters and words, and signature verification.The contributions collected in this book present the state of the art in the field of complete systems for bankcheck recognition, and explore the most promising trends in key aspects of this research field.