Welcome to the Second Edition of the Handbook of Fingerprint Recognition!

The first edition of this monography (Springer, 2003) received many positive feedbacks from readers and colleagues; the book also received the prestigious 2003 PSP award for the "Computer Science" category given by the Association of American Publishers. These accolades motivated us in our efforts to prepare this new edition of the book, which is not a simple retouch of the first version. In fact, over 500 papers on fingerprint recognition were published in the last 5 years (2003 to 2008) alone and a major update was necessary to track recent advances. The total length of the handbook grew from about 350 to about 500 pages and the number of references increased from about 600 to about 1,200. The DVD included with the book now also contains the databases used in the 2004 Fingerprint Verification Competition (FVC2004). For more details on the changes you can check the book preface (subsection "From the First to the Second Edition").

Detailed Table of Contents

Preface

Overview

Objectives

Organization and Features

From the First to the Second Edition

Contents of the DVD

Intended Audience

Acknowledgments

1 Introduction

1.1 Introduction

1.2 Biometric Recognition

1.3 Biometric Systems

1.4 Comparison of Traits

1.5 System Errors

1.5.1 Reasons behind system errors

1.5.2 Capture module errors

1.5.3 Feature extraction module errors

1.5.4 Template creation module errors

1.5.5 Matching module errors

1.5.6 Verification error rates

1.5.7 Identification error rates

1.6 System Evaluation

1.7 Applications of Fingerprint Systems

1.7.1 Application characteristics

1.7.2 Application categories

1.7.3 Barriers to adoption

1.8 History of Fingerprints

1.9 Formation of Fingerprints

1.10 Individuality of Fingerprints

1.11 Fingerprint Sensing and Storage

1.12 Fingerprint Representation and Feature Extraction

1.13 Fingerprint Matching

1.14 Fingerprint Classification and Indexing

1.15 Synthetic Fingerprints

1.16 Biometric Fusion

1.17 System Integration and Administration Issues

1.18 Securing Fingerprint Systems

1.19 Privacy Issues

1.20 Summary and Future Prospects

1.21 Image Processing and Pattern Recognition Background

1.21.1 Image processing books

1.21.2 Pattern recognition books

1.21.3 Journals

2 Fingerprint Sensing

2.1 Introduction

2.2 Off-Line Fingerprint Acquisition

2.3 Live-Scan Fingerprint Sensing

2.3.1 Optical sensors

2.3.2 Solid-state sensors

2.3.3 Ultrasound sensors

2.4 Touch Versus Sweep

2.4.1 Image reconstruction from slices

2.5 Fingerprint Images and Their Parameters

2.6 Image Quality Specifications for Fingerprint Scanners

2.7 Operational Quality of Fingerprint Scanners

2.8 Examples of Fingerprint Scanners

2.9 Dealing with Small Area Sensors

2.10 Storing and Compressing Fingerprint Images

2.11 Summary

3 Fingerprint Analysis and Representation

3.1 Introduction

3.2 Local Ridge Orientation

3.2.1 Gradient-based approaches

3.2.2 Slit- and projection-based approaches

3.2.3 Orientation estimation in the frequency domain

3.2.4 Other approaches

3.2.5 Orientation image regularization

3.2.6 Global models of ridge orientations

3.3 Local Ridge Frequency

3.4 Segmentation

3.5 Singularity and Core Detection

3.5.1 Poincaré index

3.5.2 Methods based on local characteristics of the orientation image

3.5.3 Partitioning-based methods

3.5.4 Methods based on a global model of the orientation image

3.5.5 Core detection and registration

3.5.6 Miscellanea

3.6 Enhancement

3.6.1 Pixel-wise enhancement

3.6.2 Contextual filtering

3.6.3 Multi-resolution enhancement

3.6.4 Crease detection and removal

3.6.5 Miscellanea

3.7 Minutiae Detection

3.7.1 Binarization-based methods

3.7.2 Direct gray-scale extraction

3.7.3 Minutiae encoding standards

3.8 Minutiae Filtering

3.8.1 Structural post-processing

3.8.2 Minutiae filtering in the gray-scale domain

3.9 Estimation of Ridge Count

3.10 Estimation of Fingerprint Quality

3.10.1 Global quality estimation

3.10.2 Local quality estimation

3.11 Summary

4 Fingerprint Matching

4.1 Introduction

4.2 Correlation-Based Techniques

4.3 Minutiae-Based Methods

4.3.1 Problem formulation

4.3.2 Similarity score

4.3.3 Point pattern matching

4.3.4 Some simple algebraic geometry methods

4.3.5 Hough transform-based approaches for minutiae matching

4.3.6 Minutiae matching with pre-alignment

4.3.7 Avoiding alignment

4.3.8 Miscellanea

4.4 Global Versus Local Minutiae Matching

4.4.1 The earlier approaches

4.4.2 Local structure matching through invariant distances and angles

4.4.3 Evolution of local structure matching

4.4.4 Consolidation

4.4.5 Asymmetrical local matching

4.5 Dealing with Distortion

4.5.1 Tolerance boxes

4.5.2 Warping

4.5.3 Multiple-registration and clustering

4.5.4 Triangulation and incremental expansion

4.5.5 Normalization

4.5.6 Fingerprint distortion models

4.6 Non-Minutiae Feature-Based Matching Techniques

4.6.1 Global and local texture information

4.6.2 Geometrical attributes and spatial relationship of the ridge lines

4.6.3 Level 3 features

4.7 Comparing the Performance of Matching Algorithms

4.7.1 Fingerprint database

4.7.2 Fingerprint evaluation campaigns

4.7.3 Interoperability of fingerprint recognition algorithms

4.7.4 Further notes on performance evaluation

4.8 Summary

5 Fingerprint Classification and Indexing

5.1 Introduction

5.2 Classification Techniques

5.2.1 Rule-based approaches

5.2.2 Syntactic approaches

5.2.3 Structural approaches

5.2.4 Statistical approaches

5.2.5 Neural network-based approaches

5.2.6 Multiple classifier-based approaches

5.2.7 Miscellanea

5.3 Performance of Fingerprint Classification Techniques

5.3.1 Results on NIST DB4

5.3.2 Results on NIST DB14

5.4 Fingerprint Indexing and Retrieval

5.4.1 Fingerprint sub-classification

5.4.2 Continuous classification and other indexing techniques

5.4.3 Retrieval strategies

5.4.4 Performance of fingerprint retrieval techniques

5.5 Summary

6 Synthetic Fingerprint Generation

6.1 Introduction

6.2 Background

6.3 The SFinGe Method

6.4 Generation of a Master Fingerprint

6.4.1 Fingerprint area generation

6.4.2 Orientation image generation

6.4.3 Frequency image generation

6.4.4 Ridge pattern generation

6.5 Generation of Synthetic Fingerprint Impressions

6.5.1 Variation in ridge thickness

6.5.2 Fingerprint distortion

6.5.3 Perturbation and global translation/rotation

6.5.4 Background generation

6.6 Validation of the Synthetic Generator

6.7 Automatic Generation of Ground Truth Features

6.8 SFinGe Software Tool

6.9 Summary

7 Biometric Fusion

7.1 Introduction

7.2 Performance Improvement from Fusion

7.3 Application-specific Considerations

7.4 Sources of Information

7.4.1 Fusion of multiple traits

7.4.2 Multi-finger fusion

7.4.3 Fusion of multiple samples of a finger: different sensors

7.4.4 Fusion of multiple samples of a finger: same sensor

7.4.5 Fusion of multiple representation and matching algorithms

7.5 Level of Detail of Information in Fusion

7.6 Image-Level Fusion

7.7 Feature-Level Fusion

7.8 Rank-Level Fusion

7.9 Score-Level Fusion

7.9.1 Score normalization methods

7.9.2 Bayesian framework for score fusion

7.9.3 Density-based methods

7.9.4 Classifier-based methods

7.10 Decision-Level Fusion

7.11 Summary

8 Fingerprint Individuality

8.1 Introduction

8.2 Background

8.3 Uniform Minutiae Placement Model

8.3.1 The model

8.3.2 Parameter estimation

8.3.3 Experimental evaluation

8.4 Finite Mixture Minutiae Placement Model

8.4.1 The model

8.4.2 Model fitting

8.4.3 Experimental evaluation

8.5 Other Recent Approaches

8.6 Summary

9 Securing Fingerprint Systems

9.1 Introduction

9.2 Types of Failures in Fingerprint Systems

9.3 Methods of Obtaining Fingerprint Data and Countermeasures

9.3.1 Obtaining fingerprint data of a specific user

9.3.2 Obtaining generic fingerprint data

9.4 Methods of Injecting Fingerprint Data and Countermeasures

9.4.1 Injecting a fake finger at the scanner

9.4.2 Injecting fingerprint in a communication channel or in the template storage