Biometric System and Data Analysis: Design, Evaluation, and Data Mining

ISBN-13: 9780387776255 / Angielski / Twarda / 2008 / 268 str.

Dunstone Ted; Neil Yager

Opis

Biometric systems are being used in more places and on a larger scale than ever before. As these systems mature, it is vital to ensure the practitioners responsible for development and deployment, have a strong understanding of the fundamentals of tuning biometric systems. The focus of biometric research over the past four decades has typically been on the bottom line: driving down system-wide error rates. In doing so, powerful recognition algorithms have been developed for a number of biometric modalities. These algorithms operate exceedingly well under test conditions. Books on biometrics tend to focus on biometric systems and their components, and differentiate between the various biometric modalities. Biometric System and Data Analysis: Design, Evaluation, and Data Mining brings together aspects of statistics and machine learning to provide a comprehensive guide to evaluating, interpreting and understanding biometric data. This professional book naturally leads to topics including data mining and prediction, which have been widely applied to other fields but not rigorously to biometrics, to be examined in detail. Biometric System and Data Analysis: Design, Evaluation, and Data Mining places an emphasis on the various performance measures available for biometric systems, what they mean, and when they should and should not be applied. The evaluation techniques are presented rigorously, however are always accompanied by intuitive explanations that can be used to convey the essence of the statistical concepts to a general audience. This last point is an important one for the increased acceptance of biometrics among non-technical decision makers, and ultimately the general public. Biometric System and Data Analysis: Design, Evaluation, and Data Mining is designed for a professional audience composed of practitioners and researchers in industry. This book is also suitable as a reference for advanced-level students in computer science and engineering.

Recenzje

From the reviews: "This book brings together a discussion on biometric systems, statistics, and machine learning. Its aim is to help us understand biometric data, measurements of such systems, and the ability to interpret them is highly desirable. ... The book suits the needs of many practitioners, those involved with data in all its forms, and also would make useful reading for those who are studying systems, cybernetics, and management in the many and varied courses now being made available worldwide." (W. R. Howard, Kybernetes, Vol. 38 (9), 2009)

Spis treści

An Overview of Biometrics.- An Introduction to Biometric Data Analysis.- Biometric Matching Basics.- Biometric Data.- Multimodal Systems.- Performance Testing and Reporting.- Definitions.- The Biometric Performance Hierarchy.- System Evaluation: The Statistical Basis of Biometric Systems.- Individual Evaluation: The Biometric Menagerie.- Group Evaluation: Data Mining for Biometrics.- Special Topics in Biometric Data Analysis.- Proof of Identity.- Covert Surveillance Systems.- Vulnerabilities.

Blurb

Biometric  System and Data Analysis: Design, Evaluation, and Data Mining brings together aspects of statistics and machine learning to provide a comprehensive guide to evaluate, interpret and understand biometric data. This professional book naturally leads to topics including data mining and prediction, widely applied to other fields but not rigorously to biometrics.

This volume places an emphasis on the various performance measures available for biometric systems, what they mean, and when they should and should not be applied. The evaluation techniques are presented rigorously, however are always accompanied by intuitive explanations that convey the essence of the statistical concepts in a general manner.

Designed for a professional audience composed of practitioners and researchers in industry, Biometric  System and Data Analysis: Design, Evaluation, and Data Mining is also suitable as a reference for advanced-level students in computer science and engineering.