GUIDE TO BIOMETRICS (HB 2003)

Starting with fingerprints more than a hundred years ago, there has been ongoing research in biometrics. Within the last forty years face and speaker recognition have emerged as research topics. However, as recently as a decade ago, biometrics itself did not exist as an independent field. Each of the biometric-related topics grew out of different disciplines. For example, the study of fingerprints came from forensics and pattern recognition, speaker recognition evolved from signal processing, the beginnings of face recognition were in computer vision, and privacy concerns arose from the public policy arena. One of the challenges of any new field is to state what the core ideas are that define the field in order to provide a research agenda for the field and identify key research problems. Biometrics has been grappling with this challenge since the late 1990s. With the matu­ ration of biometrics, the separate biometrics areas are coalescing into the new discipline of biometrics. The establishment of biometrics as a recognized field of inquiry allows the research community to identify problems that are common to biometrics in general. It is this identification of common problems that will define biometrics as a field and allow for broad advancement.

In today’s globally connected world there is increasing interest in using biometrics (personal physical attributes such as fingerprints, facial images, voice patterns, iris codes, and hand geometry) for human verification, identification, and "screening" applications. Biometrics are attractive because they cannot be "forgotten," are not easily stolen, and provide a direct, undeniable link between a user and a transaction.

This is a complete technical guide aimed at presenting the core ideas that underlie the area of biometrics. It explains the definition and measurement of performance and examines the factors involved in choosing between different biometrics. It also delves into practical applications and covers a number of topics critical for successful system integration. These include recognition accuracy, total cost of ownership, acquisition and processing speed, intrinsic and system security, privacy and legal requirements, and user acceptance.

Features & Benefits:

*State-of-the-art coverage of biometric theory, research, and implementation

*Provides a broad orientation for a wide class of readers, yet has tightly integrated topical organization

*Debunks myths and candidly confronts problems associated with biometrics research

*Details relevant issues in choosing between biometrics, as well as defining and measuring performance

*Defines and explains how to measure the performance of both verification and identification systems

*Addresses challenges in managing tradeoffs between security and convenience

This up-to-date and detailed resource is an extensive survey of the principles, methods, and technologies used in biometric authentication systems. Security and financial administrators, computer science professionals, and biometric systems developers will all benefit from an enhanced understanding of this important technology.

Key Topics:

> Authentication protocols

> Probabilistic error analysis

> Performance comparison methodology

> Selecting a biometric

> Maintaining database integrity

> Logistics of large-scale systems

> Thwarting attacks

> Intrinsic individuality

> Error bounds and confidence intervals

> APIs, standards, & public databases

The authors are leading international experts at the IBM T. J. Watson Research Center (Hawthorne, NY). They have published extensively in scientific journals and have been invited speakers at major conferences covering biometrics, security, pattern recognition, and related technologies.

_________________________________________

-- Security / Pattern Recognition

-- Beginning-Intermediate Level