scientific article; zbMATH DE number 3860431

From MaRDI portal
Publication:3327745





The purpose of this monograph is to provide a concise summary of the major recent developments in pattern recognition. This book can be divided into two parts: basic techniques in pattern recognition and applications. The first one includes statistical techniques (in particular, the motion of pattern complexity), cluster analysis using the main clustering algorithms, syntactic techniques (in particular, high dimensional pattern grammars) and other recognition procedures. The second part deals with two-dimensional picture recognition, one- dimensional speech recognition and understanding. This part is devoted to the methods of measuring properties of pictures, extracting objects from pictures and their applications to picture recognition and descriptions of the picture. The descriptions include representation of the objects (or regions), properties of the objects (color, texture, shape etc.), relations between objects. A relational structure is represented by a labeled directed graph in which each node represents an object (or region) and is labeled with a list of property values for that object; arcs join pairs of related regions and they are labeled with lists of relation values. Generalized relational structures, in which the information is only partially specified, can be used as models for classes of pictures. Recently, in speech recognition systems attention has been focused on the problem of characterizing the difficulty of a recognizing task. The continuing growth of the field of the speech recognition is but one aspect of the recent spurt of activity in all fields of speech processing.












This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3327745)