生物识别技术基础：指纹、面部与虹膜识别

需积分: 10 168 浏览量更新于2024-07-18 1 收藏 20.38MB PDF 举报

"Introduction to Biometrics" 是一本针对生物识别技术的入门教材，由Anil K. Jain、Arun A. Ross和Karthik Nandakumar撰写，旨在为本科和研究生级别的计算机科学和电子工程学生介绍生物测量学的基础知识。本书深入探讨了指纹、面部和虹膜这三种主要的生物识别模式，并简要介绍了手部几何、耳形和步态等其他识别方式，同时也涵盖了多生物特征系统和生物识别系统的安全性等高级主题。每章附有练习题，帮助学生深化对主题的理解并实践生物识别应用的编程设计。该书的配套资源可在http://www.csee.wvu.edu/~ross/BiometricsTextBook/找到。在生物测量学领域，本书主要关注以下知识点： 1. **生物识别基础**：生物识别是一种通过物理或行为特征来确定个人身份的技术。这种技术的应用范围广泛，包括个人电脑的安全访问和国家的边境控制。 2. **主要生物识别模式**： - **指纹识别**：基于每个人的指纹独特性进行身份验证，是最早的生物识别技术之一。 - **面部识别**：利用人脸的特征如眼睛、鼻子和嘴巴的相对位置和形状进行识别。 - **虹膜识别**：虹膜的纹理细节非常独特，且不易改变，因此在安全要求高的场合广泛应用。 3. **其他生物识别模式**： - **手部几何**：通过测量手掌和手指的尺寸进行识别。 - **耳形识别**：耳朵的形状和结构也有其独特性，可以用于生物识别。 - **步态识别**：分析行走时的运动模式来识别个体。 4. **多生物特征系统**：结合两种或多种生物识别技术以提高识别的准确性和鲁棒性，减少误报和漏报。 5. **生物识别系统的安全性**：讨论了如何保护生物识别数据不被滥用，防止欺诈和数据泄露。 6. **教育用途**：本书不仅适用于学生，还适合研究人员和从事生物识别及计算机安全的专业人士。书中每章的练习题旨在帮助学生理论联系实际，提升编程能力。 7. **版权信息**：本书由Springer Science+Business Media出版，所有权利保留，未经许可不得翻译或复制。若要在评论或学术分析中引用，需遵循适当的版权法规。 8. **电子版和ISBN**：提供了电子版图书的ISBN和DOI，方便读者在线访问和引用。 "Introduction to Biometrics" 是一个全面介绍生物识别技术的教育资源，涵盖了从基本原理到高级应用的各个方面，对于理解这一领域的核心技术及其应用具有重要的指导价值。

Contents xv

3.5.3 Facemodeling.......................................132

3.6 Summary .................................................137

BibliographicalandHistoricalRemarks.............................137

References.....................................................138

4 Iris Recognition ................................................141

4.1 Introduction . . .............................................141

4.2 Design of an Iris Recognition System ..........................144

4.3 Image Acquisition . . ........................................146

4.4 IrisSegmentation...........................................151

4.4.1 Segmentationusingtheintegro-differentialoperator .......152

4.4.2 SegmentationusingGeodesicActiveContours(GAC) .....153

4.4.3 Generatingirismasks.................................159

4.5 IrisNormalization..........................................159

4.6 IrisEncodingandMatching..................................161

4.7 Iris Quality . . . .............................................164

4.7.1 Quality assessment techniques .........................164

4.8 PerformanceEvaluation .....................................169

4.9 Summary .................................................170

BibliographicalandHistoricalRemarks.............................171

References.....................................................172

5 Additional Biometric Traits .....................................175

5.1 Introduction . . .............................................175

5.2 Ear.......................................................176

5.2.1 Eardetection........................................177

5.2.2 Ear recognition . . . ...................................178

5.2.3 Challenges in ear recognition . .........................180

5.3 Gait......................................................182

5.3.1 Featureextractionandmatching........................183

5.3.2 Challenges in gait recognition . .........................185

5.4 HandGeometry............................................186

5.4.1 Imagecapture .......................................186

5.4.2 Handsegmentation...................................188

5.4.3 FeatureExtraction ...................................189

5.4.4 Featurematching ....................................189

5.4.5 Challenges in hand geometry recognition . ...............190

5.5 SoftBiometrics ............................................190

5.5.1 Periocular ..........................................191

5.5.2 Facemarks .........................................194

5.5.3 Tattoo..............................................201

5.6 Summary .................................................206

References.....................................................207

xvi Contents

6 MULTIBIOMETRICS ..........................................209

6.1 Introduction . . .............................................209

6.2 Sources of Multiple Evidence . . ..............................212

6.2.1 Multi-sensor systems . . . ..............................213

6.2.2 Multi-algorithm systems ..............................215

6.2.3 Multi-instance systems . ..............................217

6.2.4 Multi-sample systems . . ..............................218

6.2.5 Multimodal systems ..................................219

6.3 Acquisition and Processing Architecture . . . ....................221

6.3.1 Acquisition sequence . . . ..............................221

6.3.2 Processingsequence..................................222

6.4 FusionLevels..............................................224

6.4.1 Sensor-levelfusion...................................226

6.4.2 Feature-levelfusion ..................................227

6.4.3 Score-levelfusion....................................232

6.4.4 Rank-levelfusion ....................................246

6.4.5 Decision-levelfusion.................................250

6.5 Summary .................................................253

BibliographicalandHistoricalRemarks.............................254

References.....................................................256

7 SECURITY OF BIOMETRIC SYSTEMS ........................259

7.1 Introduction . . .............................................259

7.2 AdversaryAttacks..........................................264

7.2.1 Insiderattacks.......................................264

7.2.2 Infrastructureattacks .................................266

7.3 AttacksattheUserInterface .................................268

7.3.1 Impersonation.......................................268

7.3.2 Obfuscation.........................................269

7.3.3 Spooﬁng . . . ........................................269

7.3.4 Countermeasure: spoof detection . . . ....................272

7.4 AttacksonBiometricProcessing..............................278

7.4.1 Attacks on the system modules . ........................278

7.4.2 Attacksattheinterconnections.........................280

7.5 AttacksontheTemplateDatabase.............................283

7.5.1 Countermeasure: biometric template security . . . ..........284

7.6 Summary .................................................302

BibliographicalandHistoricalRemarks.............................302

References.....................................................304

Index .............................................................307

Chapter 1

INTRODUCTION

Sur un proc

e d’identiﬁcation permettant de retrouver le nom d’un r

ecidiviste au moyen de son

seul signalment, et pouvant servir de cadre pour une classiﬁcation de photographies

alapr

efecture

de police,

alasˆuret

erale, au minist

e de la justice, etc.

Alphonse Bertillon, 1881

About an identiﬁcation process that enables ﬁnding the name of a repeat offender based on

his description only, and that can be used in the context of a classiﬁcation of photographies in the

police headquarters, in the national security ofﬁce, at the ministry of justice, etc.

English translation

The ability to identify individuals uniquely and to associate personal attributes

(e.g., name, nationality, etc.) with an individual has been crucial to the fabric of hu-

man society. Humans typically use body characteristics such as face, voice, and gait

along with other contextual information (e.g., location and clothing) to recognize

one another. The set of attributes associated with a person constitutes their personal

identity. In the early days of civilization, people lived in small communities where

individuals could easily recognize each other. However, an explosion in popula-

tion growth accompanied by increased mobility in modern society has necessitated

the development of sophisticated identity management systems that can efﬁciently

record, maintain, and obliterate personal identities of individuals.

Identity management plays a critical role in a number of applications. Examples

of such applications include regulating international border crossings, restricting

physical access to important facilities like nuclear plants or airports, controlling log-

ical access to shared resources and information, performing remote ﬁnancial trans-

actions, or distributing social welfare beneﬁts. The proliferation of web-based ser-

vices (e.g., online banking) and the deployment of decentralized customer service

A. Bertillon, “Une application pratique de l’anthropom

etrie, Annals de d

emographie interna-

tionale, 1881”

A.K. Jain et al., Introduction to Biometrics, DOI 10.1007/978-0-387-77326-1_1,

2 1 INTRODUCTION

centers (e.g., credit cards) have led to the risk of identity theft

. Rising magnitude

of identity theft and heightened concerns about national security have reinforced the

need for reliable identity management systems.

1.1 Person Recognition

The fundamental task in identity management is to establish the association be-

tween an individual and his personal identity. One must be able to determine a

person’s identity or verify the identity claim of an individual whenever required.

This process is known as person recognition. A person can be recognized based on

the following three basic methods (see Figure 1.1): (a) what he knows, (b) what he

possesses extrinsically, and (c) who he is intrinsically. While the ﬁrst method relies

on the fact that the individual has exclusive knowledge of some secret information

(e.g., password, personal identiﬁcation number, or cryptographic key), the second

method assumes that the person has exclusive possession of an extrinsic token (e.g.,

identiﬁcation card, driver’s license, passport, physical key, or personal device such

as a mobile phone). The third method establishes the person’s identity based on

his inherent physical or behavioral traits and is known as biometric recognition.

Formally, biometric recognition can be deﬁned as the science of establishing the

identity of an individual based on the physical and/or behavioral characteristics of

the person either in a fully automated or a semi-automated manner.

Knowledge-based and token-based person recognition rely on surrogate repre-

sentations of identity such as passwords or ID cards, which can be easily for-

gotten/lost, guessed/stolen, or shared. Moreover, they cannot provide vital iden-

tity management functions like non-repudiation and detecting multiple enrollments

by the same person under different identities. For example, individuals can easily

deny (repudiate) using a service by claiming that their password had been stolen or

guessed. Individuals can also conceal their true identity by presenting forged or du-

plicate identiﬁcation documents. In addition, traditional mechanisms like passwords

and tokens do not provide strong evidence for post-event person recognition, such

as suspect identiﬁcation at a crime scene. Therefore, it is becoming increasingly ap-

parent that knowledge-based and token-based mechanisms alone are not sufﬁcient

for reliable identity management.

Biometric recognition, or simply biometrics

, offers a natural and more reliable

solution to the problem of person recognition. Since the biometric identiﬁers are in-

herent to an individual, it is more difﬁcult to manipulate, share, or forget these traits.

Hence, biometric traits constitute a strong and reasonably permanent link between

a person and his identity.

Identity theft or identity fraud occurs when a person usurps the identity of another individual or

claims a false identity in order to access resources or services to which he is not entitled.

The term biometric recognition is perhaps more appropriate than biometrics, because the latter

has been historically used in the ﬁeld of statistics to refer to the analysis of biological (particularly

medical) data. For the sake of brevity, we use these two terms interchangeably in this book.

剩余327页未读，继续阅读

kernelkoder

粉丝: 57
资源: 317

生物识别技术基础：指纹、面部与虹膜识别

IntroductiontoBiometrics.pdf 英文原版

有关于虹膜定位研究的纯理论外文文献

已知两张指纹图片，根据细节点匹配，匹配指标如何计算。相关文献

ifeq ($(wildcard vendor/qcom/proprietary/biometrics/fingerprint),)里的wildcard是什么意思

请简述生物特征识别(Biometrics)技术。

ifeq ($(wildcard vendor/qcom/proprietary/biometrics/fingerprint),)

联邦学习用于生物特征识别参考文献

win10找不到指纹

刷掌技术是是什么?, 代表文献？

写一篇关于通过基于纹线的方法提取指纹特征的文献综述，大概五百字，参考英文文献10篇左右，并给出参考文献来源

最新资源