Book Image

Identity Management with Biometrics

By : Lisa Bock
Book Image

Identity Management with Biometrics

By: Lisa Bock

Overview of this book

Biometric technologies provide a variety of robust and convenient methods to securely identify and authenticate an individual. Unlike a password or smart card, biometrics can identify an attribute that is not only unique to an individual, but also eliminates any possibility of duplication. Identity Management with Biometrics is a solid introduction for anyone who wants to explore biometric techniques, such as fingerprint, iris, voice, palm print, and facial recognition. Starting with an overview of biometrics, you’ll learn the various uses and applications of biometrics in fintech, buildings, border control, and many other fields. You’ll understand the characteristics of an optimal biometric system and then review different types of errors and discover the benefits of multi-factor authentication. You’ll also get to grips with analyzing a biometric system for usability and accuracy and understand the process of implementation, testing, and deployment, along with addressing privacy concerns. The book outlines the importance of protecting biometric data by using encryption and shows you which factors to consider and how to analyze them before investing in biometric technologies. By the end of this book, you’ll be well-versed with a variety of recognition processes and be able to make the right decisions when implementing biometric technologies.
Table of Contents (20 chapters)
1
Section 1 –Understanding Biometric Authentication
6
Section 2 – Applying Biometric Technologies
12
Section 3 – Deploying a Large-Scale Biometric System

Enrolling and matching

As with any biometric, the process begins with enrolling the subject. Once the subject is enrolled, the system can obtain the voice sample of the subject and determine whether or not there is a match.

In this section, we’ll step through the process of what happens from enrollment to matching. We will compare the different types of systems, either text dependent or text independent. Then, we’ll look at matching methods that use either template matching or feature analysis.

Let’s start with the enrollment process.

Registering with a system

A voice recognition system goes through the phases shown in the following screenshot:

Figure 8.9 – Voice recognition process

To enroll in a voice recognition system, the process goes through the following steps:

  1. The subject is prompted to repeat a short passphrase or sequence of numbers.
  2. The voice must pass through an ADC, and then the algorithm...