ppt on iris scan
DESCRIPTION
Iris recognizationTRANSCRIPT
Automated Iris Recognition Technology &
Iris Biometric System
CS 790Q Biometrics
Instructor: Dr G. Bebis
Presented by Chang Jia
Dec 9th, 2005
2
Overview
The Iris as a Biometrics: The iris is an overt body that
is available for remote assessment with the aid of a
machine vision system to do automated iris
recognition.
Iris recognition technology combines computer vision,
pattern recognition, statistical inference, and optics.
The spatial patterns that are apparent in the human iris
are highly distinctive to an individual. Clinical observations
Developmental biology
3
The structure of the human eyeThe structure of the iris seen in a transverse section
The structure of the iris seen in a frontal section
Overview
4
Its suitability as an exceptionally accurate biometric
derives from its extremely data-rich physical structure
genetic independence — no two eyes are the same
patterns apparently stable throughout life
physical protection by a transparent window (the
cornea), highly protected by internal organ of the eye
externally visible, so noninvasive — patterns imaged
from a distance
Overview
5
The disadvantages to use iris as a biometric
measurement are Small target (1 cm) to acquire from a distance (about 1
m)
Moving target
Located behind a curved, wet, reflecting surface
Obscured by eyelashes, lenses, reflections
Partially occluded by eyelids, often drooping
Deforms non-elastically as pupil changes size
Illumination should not be visible or bright
Overview
PART I:Iris Recognition: An Emerging Biometric Technology
R. Wildes, "Iris Recognition: An Emerging Biometric Technology", Proceedings of the IEEE, vol 85, no. 9, pp. 1348-1363, 1997.
CS 790Q Biometrics
7
Outline
Technical Issues ** Image Acquisition
Iris Localization
Pattern Matching
Systems and Performance
** (Throughout the discussion in this paper, the iris-recognition s
ystems of Daugman and Wildes et al. will be used to provide illu
strations.)
8
Technical Issues
Schematic diagram of iris recognition
9
I. Image Acquisition
Why important?
One of the major challenges of automated iris
recognition is to capture a high-quality image of the iris
while remaining noninvasive to the human operator.
Concerns on the image acquisition rigs Obtained images with sufficient resolution and sharpness
Good contrast in the interior iris pattern with proper
illumination
Well centered without unduly constraining the operator
Artifacts eliminated as much as possible
10
The Daugman image-acquisition rig
I. Image Acquisition - Rigs
11
The Wildes et al. image-acquisition rig
I. Image Acquisition - Rigs
12
Result Image from Wildes et al. rig -- capture the iris as part of a larger image that also contains data derived from the immediately surrounding eye region
I. Image Acquisition - Results
13
In common: Easy for a human operator to master Use video rate capture
Difference: Illumination
The Daugman’s system makes use of an LED-based point light source in conjunction with a standard video camera.
The Wildes et al. system makes use of a diffuse source and polarization in conjunction with a low-light level camera.
Operator self-position The Daugman’s system provides the operator with live video fee
dback The Wildes et al. system provides a reticle to aid the operator in
positioning
Discussion
14
Purpose: to localize that portion of the acquired imag
e that corresponds to an iris
In particular, it is necessary to localize that portion of
the image derived from inside the limbus (the border
between the sclera and the iris) and outside the pupil.
Desired characteristics of iris localization:
Sensitive to a wide range of edge contrast
Robust to irregular borders
Capable of dealing with variable occlusions
II. Iris Localization
15
The Daugman system fits the circular contours via gradie
nt ascent on the parameters so as to maximize
Where is a radial Gaussian,
and circular contours (for the limbic and pupillary bounda
ries) be parameterized by center location (xc,yc), and radi
us r (active contour fitting method)
II. Iris Localization
16
The Wildes et al. system performs its contour fitting in tw
o steps. (histogram-based approach) First, the image intensity information is converted into a binary e
dge-map
where
and
Second, the edge points vote to instantiate particular contour par
ameter values.
II. Iris Localization
17
The voting procedure of the Wildes et al. system is realized via Hough transforms on parametric definitions of the iris boundary contours.
II. Iris Localization
18
Illustrative Results of Iris Localization
Obtained by using the Wildes et al. system
only that portion of the image below the upper eyelid and above the lower
eyelid should be included
19
Both approaches are likely to encounter difficulties if r
equired to deal with images that contain broader regi
ons of the surrounding face than the immediate eye r
egion
Difference: the active contour approach avoids the inevitable thres
holding involved in generating a binary edge-map
the histogram-based approach to model fitting should a
void problems with local minima that the active contour
model’s gradient descent procedure might experience
Discussion
20
Four steps:
1) bringing the newly acquired iris pattern into spatial
alignment with a candidate data base entry;
2) choosing a representation of the aligned iris patterns
that makes their distinctive patterns apparent;
3) evaluating the goodness of match between the newly
acquired and data base representations;
4) deciding if the newly acquired data and the data base
entry were derived from the same iris based on the
goodness of match.
III. Pattern Matching
21
Purpose: to establish a precise correspondence
between characteristic structures across the two
images.
Both of the systems under discussion compensate for
image shift, scaling, and rotation.
For both systems, iris localization is charged with
isolating an iris in a larger acquired image and thereby
accomplishes alignment for image shift.
III. Pattern Matching -Alignment
22
The Daugman’s system uses radial scaling to compensate for overall size as well as a simple model of pupil variation based on linear stretching.
III. Pattern Matching -Alignment
while being constrained to capture a similarity transformation of image coordinates (x, y) to (x’, y’)
Map Cartesian image coordinates (x, y) to dimensionless polar (r, ө)image coordinates according to
The Wildes et al. system uses an image-registration technique to compensate for both scaling and rotation. The mapping function (u,v) is to minimize
23
The two methods for establishing correspondences
between acquired and data base iris images seem to
be adequate for controlled assessment scenarios
Improvements:
more sophisticated methods may prove to be
necessary in more relaxed scenarios
more complicated global geometric
compensations will be necessary if full perspective
distortions (e.g., foreshortening) become
significant
III. Pattern Matching -Alignment
24
The Daugman’s system uses demodulation with complex-valued 2D Gabor wavelets to encode the phase sequence of the iris pattern to an “IrisCode”.
III. Pattern Matching - Representation
25
In implementation, the Gabor filtering is performed via a relaxation algorithm, with quantization of the recovered phase information yielding the final representation.
III. Pattern Matching - Representation
Pictorial Examples of one IrisCode
26
The Wildes et al. system makes us of an isotropic bandpass decomposition derived from application of Laplacian of Gaussian filters to the image data.
In practice, the filtered image is realized as a Laplacian pyramid. This representation is defined procedurally in terms of a cascade of small Gaussian-like filters.
III. Pattern Matching - Representation
with σ the standard deviation of the Gaussian and ρ the radial distance of a point from the filter’s center
27
Result: Multiscale representation for iris pattern matching. Distinctive features of the iris are manifest across a range of spatial scales.
III. Pattern Matching - Representation
Obtained by using the Wildes et al. system
28
The Daugman system computes the normalized Hamming distance as
The result of this computation is then used as the goodness of match, with smaller values indicating better matches.
IV. Pattern Matching – Goodness of Match
29
The Wildes et al. system employs normalized correlation between the acquired and data base representations.
IV. Pattern Matching - Decision
30
IV. Pattern Matching - Decision
For the Daugman system, this amounts to choosing a separation point in the space of (normalized) Hamming distances between iris representations.
In order to calculate the cross-over point, sample populations of imposters and authentics were each fit with parametrically defined distributions.
31
For the Wildes et al. system, the decision-making process must combine the four goodness-of-match measurements that are calculated by the previous stage of processing (i.e., one for each pass band in the Laplacian pyramid representation) into a single accept/reject judgment.
IV. Pattern Matching - Decision
32
Both the enrollment and verification modes take under 1s to
complete.
Empirical test 1: 592 irises from 323 persons the system
exhibited no false accepts and no false rejects
Empirical test 2:
Phase1: 199 irises from 122 persons, 878 attempts in
identification mode over 8 days no false accepts and 89
false rejects (47 retry with still 16 rejected)
Phase2: 96 irises (among 199) with 403 entries for
identification no false accepts and no false rejects
Systems and Performance - The Daugman iris-recognition system
33
Both the enrollment and verification modes require ap
proximately 10s to complete.
Only one empirical test: 60 different irises with 10 ima
ges each (5 at the beginning and 5 about one month l
ater) from 40 persons no false accepts and no fals
e rejects.
Systems and Performance - The Wildes et al. iris-recognition system
34
Questions?
PART II:An Iris Biometric System for Public and Personal Use
M. Negin et al., "An Iris Biometric System for Public and Personal Use", IEEE Computer, pp. 70-75, February 2000.
CS 790Q Biometrics
36
Iris identification process
The system captures a digital image of one eye, encodes its iris pattern, then matches that file against the file stored in the database for that individual.
37
The public-use system
The public-use multiple-camera system for correctly positioning and imaging a subject’s iris.
Note: wide-field-of-view (WFOV) & narrow-field-of-view (NFOV) camera
38
The public-use optical platform
(a) left and right illuminator pods, gaze director, and optical filter
(b) a solid model of the platform’s internal components.
39
The user manually positions the camera three to four inches in front of the eye.
Make sure that the device’s LED centers within the aperture that superimposes the user’s line of sight and the camera’s optical axis.
The personal-use system
40
Identification Performance
Verification distributions of authentic results (in brown) and imposter results (in green).
41
Field Trial Experience
The first pilot program—with the Nationwide Building Societ
y in Swindon, England—ran for six months and included mo
re than 1,000 participants, before going into regular service
during the fourth quarter of 1998.
The field trial experience has been very positive: • 91 percent prefer iris identification to a PIN (personal identification nu
mber) or signature,
• 94 percent would recommend iris identification to friends and family,
• 94 percent were comfortable or very comfortable using the system.
The survey also found nearly 100 percent approval on three
areas of crucial importance to consumers: reliability, securit
y, and acceptability.
42
Thank You.
Questions?