Video- and Audio-based Biometric Person Authentication

collarlimabeansSecurity

Feb 23, 2014 (3 years and 6 months ago)

82 views



Video
-

and Audio
-
based Biometric Person
Authentication



Motivation: Applications.


Modalities and their characteristics.


Characterization of a biometric authentication system.


Problem formulation: authentication and identification.


Structure of a biometric system.




Biometrics is the science of identifying or verifying the identity of a person based on physiological or
behavioral characteristics
.

Methods of Identity Verification


Is their any need for automatic biometric identification?

Possessions

(what you have)


keys, cards,

badges



-
lost or stolen

-
can be shared

-
can be duplicated

-
no remote or


over
-
the
-
phone use

Knowledge

(what you know)

passwords
,

PIN, user id

-
can be forgotten

-
can be shared

-
can be guessed

Biometric

(what you are)

fingerprints,
face, iris,




?

Applications of Automatic Identification and
Authentication

1.
physical access control
(computer, car, house, room, bank, border
crossing, boarding planes, … )

2.
access to services, facilities, information

(computer,
bank account, database)

3.
user identity monitoring for authorized use

(is it still
the user who did the login?, authorized car driver? )

4.
surveillance, security, crime prevention

5.
many other situations
(authorization to collect a child)


Note: applications 3. and 4. cannot be handled by classical identity
verification procedure (3. requires continuous operation, 4. is covert)

Market size
is huge
…..

Biometric
: any measurement of person’s
physiological traits or behavior



fingerprint



face



iris



retinal scan



ear

shape



hand geometry


infrared

(face, body parts)




speech



handwriting



signature



lip movements



keystroke dynamics



gait



odor

Physiological:

Behavior:

Genetic:


tissue sample

Market share

hand geometry

iris scan

fingerprint

face recognition

and more …

Characterization of a biometric system


Cost


Recognition performance (false accept, false reject, identification precision)
measured by:
FAR and FAR
, or
ROC curves
,


User acceptability (friendliness)


depends on many factors: ease of use, ease of
enrolment, prejudice, invasive/non invasive, speed of authentication …


(retinal scan: low, face recognition: high, widely used on ID cards)


Covert/Overt


(covert: face recognition, gait; overt: signature, iris scan)



One shot/continuous


(keystrokes, face recognition can verify identity over time)


Contact/Remote/Tele


(Tele: can identity by verified over the phone?, Remote: physical contact needed?)


Non
-
Repudiation (denial should not be possible)




Attended / Unattended


Subterfuge resistance (e.g. is just a recording presented? )


Human verifiable? (can be used semi
-
automatically)?

Properties of individual biometrics

Performance of selected biometrics

Problems:

Identification vs. Authentication

Identification

(1:N matching):


the system uses the biometric to determine the corresponding person
from a database containing many people, or decides that the person is
not enrolled in the database.

Authentication

(1:1 matching, identity verification):
the system matches the input biometric against a single biometric
record. The latter could be stored on a card presented at the transaction
time, or looked up in a database with the help of a key such as an
account number or employee ID. The output of the match is either
“Yes” if the two biometrics match or “No” otherwise. Often during the
enrollment process, we need to employ an identification system to
make sure the person is not already enrolled and the subsequent usages
are authentication instances using the unique number assigned to the
user during the enroll.


Structure of a biometric system

Literature:


Biometrics: Personal Identification in a Networked
Society
, Jain, Bolle, Pankanti, Eds., Kluwer Press.
ISBN 0792383451



Biometrics: Advanced Identify Verification: The
Complete Guide
, by Julian D. M. Ashbourn, 352
pages Bk&Cd
-
Rom edition (October 2000), Springer
Verlag; ISBN: 1852332433



Biometrics: Identity Verification in a Networked
World
, by Samir Nanavati, 224 pages 1 edition (March
15,2002) John Wiley & Sons; ISBN: 0471099457