Ulf Grenander – författare

Pattern Theory provides a comprehensive and accessible overview of the modern challenges in signal, data, and pattern analysis in speech recognition, computational linguistics, image analysis and computer vision. Aimed at graduate students in biomedical engineering, mathematics, computer science, and electrical engineering with a good background in mathematics and probability, the text includes numerous exercises and an extensive bibliography. Additional resources including extended proofs, selected solutions and examples are available on a companion website.The book commences with a short overview of pattern theory and the basics of statistics and estimation theory. Chapters 3-6 discuss the role of representation of patterns via condition structure. Chapters 7 and 8 examine the second central component of pattern theory: groups of geometric transformation applied to the representation of geometric objects. Chapter 9 moves into probabilistic structures in the continuum, studying random processes and random fields indexed over subsets of Rn. Chapters 10 and 11 continue with transformations and patterns indexed over the continuum. Chapters 12-14 extend from the pure representations of shapes to the Bayes estimation of shapes and their parametric representation. Chapters 15 and 16 study the estimation of infinite dimensional shape in the newly emergent field of Computational Anatomy. Finally, Chapters 17 and 18 look at inference, exploring random sampling approaches for estimation of model order and parametric representing of shapes.

The aim of pattern theory is to create mathematical knowledge representations of complex systems, analyse the mathematical properties of the resulting regular structures, and to apply them to practically occuring patterns in nature and the man-made world. Starting from an algebraic formulation of such representations they are studied in terms of their topological, dynamical and probabilistic aspects. Patterns are expressed through their typical behaviour as well as through their variability around their typical form. Employing the representations (regular structures) algorithms are derived for the understanding, recognition, and restoration of observed patterns. The algorithms are investigated through computer experiments.

Pattern Theory: From Representation to Inference provides a comprehensive and accessible overview of the modern challenges in signal, data and pattern analysis in speech recognition, computational linguistics, image analysis and computer vision. Aimed at graduate students in biomedical engineering, mathematics, computer science and electrical engineering with a good background in mathematics and probability, the text includes numerous exercises and an extensive bibliography. Additional resources including extended proofs, selected solutions and examples are available on a companion website. The book commences with a short overview of pattern theory and the basics of statistics and estimation theory. Chapters 3-6 discuss the role of representation of patterns via conditioning structure and Chapters 7 and 8 examine the second central component of pattern theory: groups of geometric transformation applied to the representation of geometric objects. Chapter 9 moves into probabilistic structures in the continuum, studying random processes and random fields indexed over subsets of Rn, and Chapters 10, 11 continue with transformations and patterns indexed over the continuum. Chapters 12-14 extend from the pure representations of shapes to the Bayes estimation of shapes and their parametric representation. Chapters 15 and 16 study the estimation of infinite dimensional shape in the newly emergent field of Computational Anatomy, and finally Chapters 17 and 18 look at inference, exploring random sampling approaches for estimation of model order and parametric representing of shapes.

This book arose out of a number of different contexts, and numerous persons have contributed to its conception and development. It had its origin in a project initiated jointly with the IBM Cambridge Scien­ tific Center, particularly with Dr. Rhett Tsao, then of that Center. We are grateful to Mr. Norman Rasmussen, Manager of the IBM Scientific Center Complex, for his initial support. The work is being carried on at Brown University with generous support from the Office of Computing Activities of the National Science Foundation (grants GJ-174 and GJ-7l0); we are grateful to Dr. John Lehmann of this Office for his interest and encouragement. Professors Donald McClure and Richard Vitale of the Division of Applied Mathematics at Brown University contributed greatly to the project and taught courses in its spirit. We are indebted to them and to Dr. Tore Dalenius of the University of Stockholm for helpful criticisms of the manuscript. The final stimulus to the book's completion came from an invLtation to teach a course at the IBM European Systems Research Institute at Geneva. We are grateful to Dr. J.F. Blackburn, Director of the Institute, for his invitation, and to him and his wife Beverley for their hospitality. We are greatly indebted to Mrs. Katrina Avery for her splendid secretarial and editorial work on the manuscript.

In this book a global shape model is developed and applied to the analysis of real pictures acquired with a visible light camera under varying conditions of optical degradation. Computational feasibility of the algorithms derived from this model is achieved by analytical means. The aim is to develop methods for image understanding based on structured restoration, for example automatic detection of abnormalities. We also want to find the limits of applicability of the algorithms. This is done by making the optical degradations more and more severe until the algorithms no longer succeed in their task. This computer experiment in pattern theory is one of several. The others, LEAVES, X-RAYS, and RANGE are described elsewhere. This book is suitable for an advanced undergraduate or graduate seminar in pattern theory, or as an accompanying book for applied probability, computer vision, or pattern recognition.

From the Preface to the First Edition (1957): The purpose of this book is two-fold. It is written in the terminology of the theoretical statistician because one of our objectives is to direct his attention to an approach to time series analysis that is essentially different from most of the techniques used by time series analysts in the past. The second objective is to present a unified treatment of methods that are being used increasingly in the physical sciences and technology. We hope that the book will be of considerable interest to research workers in these fields. Keeping the first objective in mind, we have given a rigorous mathematical discussion of these new topics in time series analysis.The existing literature in time series analysis is characterized with few exceptions by a lack of precision both in conception and in the mathematical treatment of the problems dealt with. To avoid this vagueness, we have devoted more space to rigorous proofs than may appear necessary to some readers, but we believe that a study of the proofs will furnish valuable clues to the practical validity of the results and be an important guide to intuition. We have tried to balance the formal proofs with intuitive remarks and comments on practical applications. While the regularity assumptions we have required in many cases may seem restrictive, appropriately interpreted they give an indication of the range in which the methods are practically valid. We have made such interpretations in the comments accompanying the formal proofs. Readers should have knowledge of statistics and basic probability. The second edition was printed with corrections.

From the Preface (1955): The first part of the present exposition is devoted to the theory of Toeplitz forms. The second part deals with applications, in particular to the calculus of probability and mathematical statistics. Neither part claims completeness in any way. Our purpose has been to elucidate the principal ideas of this remarkable chapter of modern analysis and to help the interested student of mathematical statistics to acquire a working knowledge of the subject. The somewhat protracted Chapter 1 explains not only the notation employed but contains also the definition of important auxiliary concepts and the exposition of basic results which will be used later. This arrangement avoids interruptions in the main text ... [It is assumed] that the reader is in possession of the fundamental facts of the theory of functions. ""In chapters 2 and 3 certain topics appear which were treated in the book on orthogonal polynomials by G. Szego. In view of the progress made in this subject since the publication of that book (1939) it was possible to bring some details in an improved setting. The other chapters contain partly old and partly more recent results, some older facts in a new setting, and finally some completely new results. Chapter 1-6 and Chapter 9 have been prepared by Szego, the other chapters by Grenander ...""