Skip to main content
main-content

Über dieses Buch

Biometrics, the science of using physical traits to identify individuals, is playing an increasing role in our security-conscious society and across the globe. Biometric authentication, or bioauthentication, systems are being used to secure everything from amusement parks to bank accounts to military installations. Yet developments in this field have not been matched by an equivalent improvement in the statistical methods for evaluating these systems. Compensating for this need, this unique text/reference provides a basic statistical methodology for practitioners and testers of bioauthentication devices, supplying a set of rigorous statistical methods for evaluating biometric authentication systems. This framework of methods can be extended and generalized for a wide range of applications and tests. This is the first single resource on statistical methods for estimation and comparison of the performance of biometric authentication systems. The book focuses on six common performance metrics: for each metric, statistical methods are derived for a single system that incorporates confidence intervals, hypothesis tests, sample size calculations, power calculations and prediction intervals. These methods are also extended to allow for the statistical comparison and evaluation of multiple systems for both independent and paired data. Topics and features: * Provides a statistical methodology for the most common biometric performance metrics: failure to enroll (FTE), failure to acquire (FTA), false non-match rate (FNMR), false match rate (FMR), and receiver operating characteristic (ROC) curves * Presents methods for the comparison of two or more biometric performance metrics * Introduces a new bootstrap methodology for FMR and ROC curve estimation * Supplies more than 120 examples, using publicly available biometric data where possible * Discusses the addition of prediction intervals to the bioauthentication statistical toolset * Describes sample-size and power calculations for FTE, FTA, FNMR and FMR Researchers, managers and decisions makers needing to compare biometric systems across a variety of metrics will find within this reference an invaluable set of statistical tools. Written for an upper-level undergraduate or master’s level audience with a quantitative background, readers are also expected to have an understanding of the topics in a typical undergraduate statistics course. Dr. Michael E. Schuckers is Associate Professor of Statistics at St. Lawrence University, Canton, NY, and a member of the Center for Identification Technology Research.

Inhaltsverzeichnis

Frontmatter

Introduction

Frontmatter

Chapter 1. Introduction

Abstract
This chapter is the introduction to this book.
Michael E. Schuckers

Chapter 2. Statistical Background

Abstract
This chapter is a summary of the statistical methods and theory that underlie the rest of this book.
Michael E. Schuckers

Primary Matching and Classification Measures

Frontmatter

Chapter 3. False Non-Match Rate

Abstract
The focus of this chapter is the false non-match rate. The false non-match rate (FNMR) is the rate at which a biometric matcher miscategorizes two signals from the same individual as being from different individuals. In this chapter, we focus on statistical methods for estimation and hypothesis testing of FNMR rates. We start with the notation and the correlation structure that we will use throughout the chapter. We then present statistical methods for estimating and comparing FNMR’s. This includes confidence interval and hypothesis testing for a single FNMR as well as for comparing two or more FNMR’s. These methods are done using both large sample as well as non-parametric methods. A discussion of sample size and power calculation follows that section. We conclude this chapter with a section on prediction intervals and a discussion section.
Michael E. Schuckers

Chapter 4. False Match Rate

Abstract
False match rates are an important measure of bioauthentication system performance. The false match rate (FMR) is the rate at which a biometric process mismatches biometric signals from two distinct individuals as coming from the same individual. Statistical methods for that rate are the focus of this chapter. We begin with an introduction to false match rates and the notation that we’ll use throughout this chapter. That is followed by a section on the correlation structure for the two-instance false match rate. In that section, we also discuss estimation of parameters in the general correlation structure as well as some simplifications of that general correlation structure. Section 4.2 contains a description of the two-instance bootstrap for estimation on an FNMR. The two-instance bootstrap is a new methodology for estimation of the sampling variability in an FNMR. We then turn to statistical methods for a single FNMR as well as for multiple FNMR’s. Large sample as well as bootstrap and randomization approaches to confidence intervals and hypothesis tests are given. This is followed by a section on sample size and power calculations for an FNMR. Prediction intervals for the FNMR are the focus on the next section. Lastly, we provide a brief discussion of the statistical methods for the FNMR in this chapters.
Michael E. Schuckers

Chapter 5. Receiver Operating Characteristic Curve and Equal Error Rate

Abstract
This chapters provides methodology for statistical inference concerning receiver operating characteristic (ROCs) curves and equal error rates (EER’). We begin with an introduction to the ROC with special focus on a polar coordinates representation of the ROC. We then propose a new bootstrap methodology for estimation of the variability in a sample ROC. Next we discuss our methodology for making curvewise confidence regions for the ROC. This methodology forms the basis for our approach to inference for the rest of the chapter. Having presented our methodology for a single ROC, we move to methods for comparing two ROC’s. We do this both for the case when the ROC’s are collected independently as well as when the matching scores are collected in a paired manner. Comparisons of three or more ROC’s whether paired or independent is the last topic on ROC’s in this chapter. Our focus then moves to statistical methods for equal error rates (EER’s). Our organization for the part of this chapter on EER’s is similar to our structure for ROC inference. We start with estimation for a single EER. This is followed by methodology for comparing two EER’s and then comparing three or more EER’s. Both sections have descriptions of comparisons for independent and paired data collection. This chapter ends with a further discussion of some of these topics.
Michael E. Schuckers

Biometric Specific Measures

Frontmatter

Chapter 6. Failure to Enrol

Abstract
The failure to enrol is an important metric for evaluating the performance of a bioauthentication system. The failure to enrol (FTE) rate is the percent of individuals who are unable to be enrolled in a biometric system. (Note that the spelling ‘enrol’ is typical in biometrics rather than the alternative spelling ‘enroll.’) Thus, failure to enrol rate or FTE is one measure of a system’s performance. This chapter lays out statistical methods for estimation and comparison of FTE’s. We begin by describing the correlation structure of failure to enrol data as well as the notation that we will use in the rest of the chapter. Moving to statistical methods, we start with tools for a single enrolment process. This is followed by a discussion of methods for comparing multiple FTE’s. Both large sample and bootstrap methods are used for implementing these methodologies. Sample size and power calculations, along with prediction interval are presented next. We conclude with a discussion of the work presented here. For the methodologies that we present in this chapter, we have provided at least one example of each; where possible we have used ‘real’ observed data from biometric collections.
Michael E. Schuckers

Chapter 7. Failure to Acquire

Abstract
This chapter focuses on statistical methods for the failure to acquire rate. We begin this chapter by presenting the notation that we will use in this chapter as well as the basic correlation structure for acquisition decisions. We then discuss statistical methods for a single FTA, γ. Both confidence intervals and hypothesis tests are given. Methods for comparing two and then three or more FTA’s are given in thereafter. We present large sample, bootstrap and randomization approaches to inference about multiple FTA’s. Sample size calculations and power calculations are presented next. We end this chapter with a section on prediction intervals for an FTA and a discussion section reiterating the results for this chapter. For each methodology, we provide examples of that application of that technique to ‘real’ biometric authentication acquisition decisions.
Michael E. Schuckers

Additional Topics and Appendices

Frontmatter

Chapter 8. Additional Topics and Discussion

Abstract
In this chapter, we provide a summational discussion of the topics covered in this book as well as a brief description of some topics not covered by this book.
Michael E. Schuckers

Chapter 9. Tables

Abstract
In this chapter we provide some statistical tables useful to the reader and practitioner of the methods given in this book.
Michael E. Schuckers

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise