Intro to Biometrics

Τεχνίτη Νοημοσύνη και Ρομποτική

17 Νοε 2013 (πριν από 4 χρόνια και 7 μήνες)

94 εμφανίσεις

Biometrics

W. A. Barrett, cmpe dept., SJSU

vs. 2.0

Theory in a nutshell

Segmentation

Recognition

Verification

Fingerprints/Face/Iris/Speaker recognition

Logface matching

Theory in a Nutshell

Capture
images of objects (usually persons)

Segment a view

Compress views to
biometric codes.

Compare two biometric codes, yielding a
biometric difference.

When two differences are small enough (less
than some threshold), the corresponding
objects are considered the same.

Otherwise the objects are considered
different.

A Sample Database

distance j

=

square
(D[j,i]
-

C[i])/var[i]

where

D[j,i] = database j’s component i, 1 <= j <= 5 (rows)

C[i] = candidate component i, 1 <= i <= 6 (columns)

var[i]= variance from database, component i

NAME
BIOCODE
Bill Barrett
5.9.6.30.7.6
Dave Matthews
3.9.4.25.9.7
Mike Sanders
5.8.4.33.6.5
Fred Friendly
8.4.2.28.7.3
Bill Clinton
2.6.3.30.6.7
Distance Calculation

local.xls

or
web.xls

NAME
BIOCODE
B I O C O D E S
Bill Barrett
5.9.6.30.7.6
5
9
6
30
7
6
Dave Matthews
3.9.4.25.9.7
3
9
4
25
9
7
Mike Sanders
5.8.4.33.6.5
5
8
4
33
6
5
Fred Friendly
8.4.2.28.7.3
8
4
2
28
7
3
Bill Clinton
2.6.3.30.6.7
2
6
3
30
6
7
AVERAGE
4.6
7.2
3.8
29.2
7
5.6
VARIANCE
5.3
4.7
2.2
8.7
1.5
2.8
Candidate:
4.8
7.9
3.2
31
6.2
5.1
DISTANCE
D I S T A N C E S
Bill Barrett
4.66
0.01
0.26
3.56
0.11
0.43
0.29
Dave Matthews
11.81
0.61
0.26
0.29
4.14
5.23
1.29
Mike Sanders
0.79
0.01
0
0.29
0.46
0.03
0
Fred Friendly
8.86
1.93
3.24
0.65
1.03
0.43
1.58
Bill Clinton
3.7
1.48
0.77
0.02
0.11
0.03
1.29
Segmentation

Image typically contains background noise

Segmentation

is isolating a biometric view
from the image

Motion segmentation uses video to reject static
background pixels

Two or more cameras yield distance measures

Given a static image, segmentation requires
heuristic methods

Static segmentation may be the most difficult
design challenge of a biometric system

Recognition

Form an
enrolled

database

of biometric codes

each entry represents a different
candidate

each candidate is associated with a biometric

Capture a view of a candidate and compute its
biometric code
C.

Compare
C with all candidates in the database.

Form a list of database candidates, ordered by
increasing biometric distance.

Front of the list should be the matching candidates.

Recognition (2)

If the top candidate has a small
-
enough
biometric
distance
, we say that we have
recognized

the candidate.

If the top candidate's biometric distance is
too large, then the candidate has
not

been
recognized.

This implies a
threshold level

has been
determined for biometric differences

Recognition
--

Four Cases

(good) Top candidate's biocode is small
enough, and is the
correct

person.

(bad) Top candidate's biocode is small
enough, but is the
wrong

person (false
acceptance)

(good) Top candidate's biocode is too large,
and this is the
wrong

person.

(bad) Top candidate's biocode is too large,
yet this is the
correct

person (false rejection)

Recognition Goals

Maximize

correct matching of a candidate to
the database

Minimize

false acceptance and false
recognition

Verification

Candidate presents biometric image PLUS
identification information, such as a credit
card plus PIN

System locates candidate in the database
through the credit card/PIN data

One biometric distance is computed
--

if
small enough, the candidate is verified.

Can still have a false acceptance or false
rejection!

Authentics
-
Imposters

Biometric quality is measured statistically by
acquiring two distributions
--

Authentics

--

distribution of biometric
distances of the
same

persons, but with
different

images

Imposters

--

distribution of biometric
distances of images of pairs of
different

persons

These should be widely separated, but often
aren't

Authentics
-

Imposters

Authentics
-
Imposters

The two distributions will
overlap

in general

The extent of the overlap relative to the two areas
provides a measure of the quality of this biometric
measure

Small overlap
--

good biometric

Large overlap
--

poor biometric

Best viewed through the
accumulated

distribution

shows probability of correct identification

local.xls
or

web.xls

for a model

Authentics
-
Imposters

Choice of Threshold

At the crossover of the A
-
I curves, we have a
threshold that makes

false acceptance rate == false rejection rate

Assumes that the relative number of attempts
is balanced

Moving the threshold to the
left

means more
false rejections, but fewer false acceptances

Moving the threshold to the
right

means
fewer false rejections, more false
acceptances

Quality Measure

The quality of a biometric measure can be
estimated from these two curves

use a good representative sample of
measurements (not easily done!)

find the crossover point

FARR = % at crossover point

FARR: False Acceptance
-
Recognition Ratio

View Compression

biometric code

from a
view

Fast Fourier transform

Gabor wavelet transform

Legendre moments

Chebyshev moments

pseudo
-
Zernike moments

The choice should:

eliminate unwanted view variations (scale,
rotation, translation, avg intensity, etc.)

produce maximum discrimination, i.e. smallest
possible FARR

Legendre Moments

f(x,y) is the image intensity vector

P
0
(x) = 1, P
1
(x) = x

1
1
1
1
)
,
(
)
(
)
(
4
)
1
2
)(
1
2
(
dy
dx
y
x
f
y
P
x
P
q
p
L
q
p
pq

n
x
P
n
x
P
x
n
x
P
n
n
n
)
(
)
1
(
)
(
)
1
2
(
)
(
2
1

Legendre Moments

Are orthogonal and complete

the view can be reconstructed, given enough
(p,q) pairs

Are translation invariant

the translation component is in (0,0)

Are
not

scale invariant

face: need to rescale to a normal view, typically
done by finding the eyes, etc.

Are
not

rotation invariant

pseudo
-
Zernike Moments

Much more complex set of polynomials

Are orthogonal and complete

Not scale or translation invariant

Certain functions of the moments are rotation
invariant

most human biometrics don't need this

Useful for logface biometrics

Face Recognition

Many methods have been proposed

eigenfaces (Alex Pentland, MIT)

feature extraction (Joseph Attick, Identix)

some are proprietary

Discrimination depends critically on

uniform lighting conditions

full frontal face
--

no side views

“plain” expression

no attempt at disguise

good segmentation, centering the eyes

Best results FARR = 1
-
5%

Face Recognition

Relatively high FARR means restricted use:

verification under controlled conditions (disguise
can be used to evade detection, but difficult to
fake a verification trial)

sifting out a small number of candidates from a
larger set

NOT indicated for

recognition

critical applications

Fingerprints

For digital prints, the FBI routinely finds
persons in their large national database from
prints sent through the internet (AFIS)

Statistics are unknown, but believed to have
a FARR less than 1
E
-
5

Fingerprint analysis for forensic purposes has a
much smaller FARR

Small or smudged prints (typical of crime
scenes) are likely to result in identification errors.

Iris Scanning

Iris Scanning

Image capture requires telephoto camera

Daugman recommends infrared light

Locate pupil (heuristic)

Daugman uses a circle
-
finding algorithm

Locate sclera

surrounds pupil

Locate upper and lower eyelids

Form biocode from iris patterns

Daugman uses 8 bands and a Gabor filtering to
yield a 256 byte code

Distance measure

Daugman uses a Hamming distance measure

Iris Scan A
-
I distribution

from John Daugman's patent

Iris Issues

Pupil finding is difficult

Background light sources reflected in pupil

Eyelashes sometimes obscure iris

Eyes may be partly closed

Eye movements are rapid, may cause image
capture failure

Telephoto centering and autofocus important

Capture system can be expensive

Sensar’s manufacturing cost ~\$2,000

Recognition failure rate fairly large ~1
-
5%

Sensar, Inc.

A New Jersey startup, 1990
-
2000 period

Used the Daugman iris patent

Developed extraordinary optics system

two cameras, one wide
-
angle, the other a
telephoto with autofocus and angular tracking

system could accurately identify a person as
he/she approached an ATM machine

tested in a Fort Worth bank system

Sensar failed for various reasons

Speaker Recognition

Starts with an audio sample of a human
voice

Typically, person is prompted to repeat
certain phrases

Speech fragment compressed by FFT or
wavelet transforms

Identification/verification similar to other
biometrics

FARR ~ 1E
-
2 at best

Forest Service Project

Goal
--

Match a cut log face to its mating
stump

U. S. Forest Service interested in combating
theft of timber from national forests

find stump face in a collection of photographs of
faces taken at various sawmills

use biometrics to filter out the most likely
candidates

use forensic tools to indict and prosecute thieves

Logface System Features

Color images input by digital camera, many
supported image formats

Semi
-
automatic segmentation of log faces

operator segmentation needed

Uses pseudo
-
Zernike polynomials to obtain a
rotation
-
invariant biometric code

Database
mysql

employed under Linux

Friendly user environment for locating
matching faces from a database

Logface Results

Selected Bibliography

http://www.biometrics.org

--

Biometrics web site

http://www.identix.com

--

Face recognition, fingerprint vendor

http://www.iritech.com

--

Daugman’s iris scanning company, patent holder

John Daugman, patent no. 5,291,560, Iris scanning patent

Wechsler
et al,
editors,
Face Recognition

Maltoni, Maio, Jain & Prabhakar,
Handbook of Fingerprint Recognition,

Springer, 2003.

Mukundan & Ramakrishnan,
Moment Functions in Image Analysis,

World Scientific, 2003

Duda & Hart,
Pattern Classification and Scene Analysis, Wiley Interscience

Fukunaga,
Introduction to Statistical Pattern Recognition, Academic Press

Theodoridis & Koutroumbas,

Summary

Biometrics is an established discipline

...though research is ongoing

Mechanism is

compressing an image into a biocode

comparing pairs of biocodes with a distance
measure
d(I1, I2)

forming a database of enrollees

locating or verifying a candidate against the
database with the distance measure

Summary

FARR = equal false acceptance and false
rejection ratio

Most popular human biometrics

digital fingerprints, with FARR ~ 1E
-
5

forensic fingerprints (non
-
digital), FARR < 1E
-
7

face, with FARR ~ 1E
-
2 at best

iris, with FARR < 1E
-
7

speaker recognition, with FARR < 1E
-
2

Other applications

Draws upon pattern recognition theory