No label defined (Q4040635)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 193837
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 193837

      Statements

      0 references
      0 references
      0 references
      5 June 1993
      0 references
      Haar factorization
      0 references
      geometric point processes
      0 references
      Haar measures on topological groups
      0 references
      combinatorial integral geometry
      0 references
      stereology
      0 references
      isoperimetric inequalities
      0 references
      random chord length distributions
      0 references
      Palm distribution
      0 references
      The volume actually consists of two independent parts, and each part can be considered as a separate book. The second part is a translation from the German of the last two authors' book ``Stochastische Geometrie. Eine Einführung'' [Berlin: Akademie-Verlag (1983; Zbl 0539.60019)]. Therefore we will review only part I of this volume, ``Geometrical measures and processes'', due to the first author. Taken as a whole, this part presents the first systematic attempt to application of the fundamental notion of Haar factorization to classical integral geometry and to the theory of geometric point processes. The bulk of results obtained by Haar factorization in Ambartzumyan's book is the main specific feature different from all other texts on integral geometry and geometric probability where this simple and powerful tool was almost completely ignored. The Haar factorization is a corollary of the theorem of uniqueness of Haar measures on topological groups. In integral geometry, invariant measures on product spaces fail to be unique and after separating the Haar measure they yield non-trivial factor measures which are of considerable interest. A special case of Haar factorization is the Lebesgue factorization (in which case the Lebesgue measure is separated). These factorizations are efficiently used by Ambartzumyan to obtain simpler derivations of various established results as well as to obtain numerous new ones. The significance of these non-trivial measure factors is made very clear in the chapters devoted to the study of random point processes. Namely, it is shown that the probabilistic distribution of shapes (both Euclidean and affine) of randomly chosen aggregates of points from Poisson and related point process realizations are governed by these measure factors. This theme is one of the two main probabilistic topics of the book and it is carefully developed along with many auxiliary concepts and various ramifications. Another probabilistic topic consists in the application of the results of combinatorial integral geometry to the theory of general geometrical processes. Following the general self-contained style of the book here again the author presents the necessary apparatus of combinatorial integral geometry in a separate chapter. The latter contains the necessary essentials of the author's monograph ``Combinatorial integral geometry. With applications to mathematical stereology'' [Chichester etc.: John Wiley (1982; Zbl 0492.53043)], and also presents some novel elements of the theory, namely the so-called flag representations of convex bodies (which still await for stochastic applications). The book is saturated with previously unpublished results or results which failed to gain the attention they deserved. Among them, affine shapes seem to be a highly successful invention of the author. Typical is the situation with Euclidean shapes. Many facts of the theory of Euclidean shapes can be found in the papers of R. E. Miles. In his papers, however, the Haar factorization roots remained obscure. Chapter 1-4 of the book can be considered as an introduction to the theory of Haar factorization. In particular, they contain simple rigorous proofs of uniqueness of the standard invariant measures in the space of lines, planes, etc. as well as proofs of certain product characterizations in cases where the invariant measures are no longer unique. An application of the factorization methods leads to position- size-shape factorization with both Euclidean and affine shapes. In chapter 4, this method is applied to the study of the Haar measure on the group of Lebesgue measure-preserving affine transformations of \(R^d\) \((d=2,3)\) and, in particular, on the important subgroup of `affine deformations'. The chapter culminates in the solution of the modified J. J. Sylvester problem for four points in \(R^2\) and five points in \(R^3\) and also gives a complete description of the space of planar affine shapes. In chapter 5, the author starts with the theme of combinatorial integral geometry (this theory finds important stochastic applications in chapter 10) and concentrates on the combinatorics of lines in \(R^ 2\) and of planes in \(R^3\). The results on planes are applied to the construction of flag representations of bounded centrally-symmetrical convex bodies in \(R^3\) which in turn have conspicuous connections with the zonoids. The last section of chapter 5 contains a brief synopsis of other ramifications of combinatorial integral geometry (a complete presentation can be found in the author's monograph cited above). Chapter 6 presents a mixed series of integrations aiming at one hand at complementing the preceding material, and on the other at providing tools for the study of geometrical processes. The chapter starts with calculations of several integrals with respect to the translation invariant measures, and the results generalize the formulae of classical integral geometry. There are sections on translational analysis of discrete points sets, on an extremal property of the uniform distribution in the space of directions ({\S}6.7), on kinematic analysis of sets of segments and mosaics, and on Pleijel identities and isoperimetric inequalities. A separate section contains calculations of integrals in terms of elementary functions. These refer to random chord length distributions for convex polygons as well as to triangular shapes. Chapter 7 (`Stochastic point processes') is an introduction to the theory of random point processes oriented to serve the subsequent needs. In particular, it contains a beautiful geometrical interpretation of a probability continuation theorem in the space of point process realizations as well as a number of concrete constructions of group- invariant lattice type point processes (including one based upon C. L. Siegel's theorem). Results concerning the types of nodes in segment processes and random mosaics in the plane are also included. In chapter 8 (Palm distributions of point processes in \(R^n)\), the Palm distribution for \(T_n\)-invariant finite intensity point processes \((T_n\) is the group of translations in \(R^n)\) is defined with the help of Haar factorization. Then the author concentrates on related analytical tools, especially the `Palm formulae' and considers certain equations that relate the distribution of a point process to its Palm distribution. Versions of these equations reappear in stochastic geometrical context in chapters 9 and 10. In chapter 9 (`Typical configurations in Poisson processes') an application of Haar factorization leads to the concept of relative Palm distribution of a group-invariant geometrical process. In particular, a number of size- shape distributions of (or stochastic constructions for) typical configurations generated by Poisson processes are obtained. In the concluding sections of chapter 9, Voronoi mosaics and mean values for random polygons are considered. In chapter 10 (`Sections through planar geometrical processes'), which the author calls `synthetic', many results which where obtained in the previous chapters are used in the study of sections of geometrical processes in the plane. The results essentially belong to `mathematical stereology', to use the term coined by the author in his monograph cited before. The author often treats the cases \(R^2\) and \(R^3\) separately in order to make the material accessible for broader circles of readers. However this tendency of simplification has its disadvantages. For example, the framework built by the author in chapter 10 seems to be sufficient for treating processes of general convex domains (rather than merely disc processes as the author actually does) without significant extra complications. Recently he published a review of the book of the last two authors of this volume and \textit{W. S. Kendall} [Stochastic geometry and its applications. Berlin: Akademie-Verlag (1987; Zbl 0622.60019)], (which is an extended version of part II of the volume we now discuss) in which he strongly advocated the integral-geometrical approach to the problems of stochastic geometry. The present book illustrates these ideas in detail in a rather convincing way. We hope that the book will stimulate further work in this applicable kind of integral geometry which branch may one day provide answers to many problems in the modern sciences where comprehension of complicated geometrical patterns and structures is required. It is a pleasure to inform the reader that the extended English version of part I of this book will appear in 1990 in Cambridge University Press [Factorization calculus and geometric probability. Cambridge etc. (1990; Zbl 0715.53049)]
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references