This book is a concise introduction to the basic topics of statistical pattern recognition and as such makes a good reference work on the subject. It is not clear, however, that the book covers the material in enough depth to serve as a textbook, even though examples are scattered throughout the text and exercises appear at the end of each chapter.
The topics covered include probability theory for random vectors, hypothesis testing, feature extraction, quadratic and linear classifiers, both parametric and nonparametric estimation, sequential testing, relaxation and Markov random fields, and unsupervised learning. The material is covered in a crisp style and the presentation is good. Although the last three chapters describe newer and more advanced topics, it would have been useful for the author to give some minimal coverage of the relation between standard techniques and neural network models, as well as to discuss the relevant optimization techniques.
All in all, the book fulfills its basic purpose, which is to provide a self-contained yet adequate introduction to pattern recognition. It provides useful examples and bibliographic references.
Compared to other texts, the book is a reasonable complement to the book by Tou and Gonzalez [1], but is a subset of Duda and Hart’s book [2]. It could perhaps be used in conjunction with the former for a course on pattern recognition. The author gives only a few basic references, which appear at the end of each chapter. In summary, the book provides a clear and brief synopsis of some of the basic techniques in pattern recognition.