Iris Recognition

1
Iris Recognition
BIOM 426 Biometrics Systems
Instructor Natalia Schmid
2
Outline

• Anatomy
• Iris Recognition System
• Image Processing (John Daugman)
  
  - iris localization
  -
  encoding
• Measure of Performance
• Results
• Other Algorithms
• Pros and Cons
• Ongoing Work at WVU
• References

3
Anatomy of the Human Eye

• Eye Camera
• Cornea bends, refracts, and focuses light.
• Retina Film for image projection (converts
  image into electrical signals).
• Optical nerve transmits signals to the brain.

4
Structure of Iris

• Iris Aperture
• Different types of muscles
• - the sphincter muscle (constriction)
• - radial muscles (dilation)
• Iris is flat
• Color pigment cells called melanin
• The color texture, and patterns are unique.

5
Individuality of Iris
Left and right eye irises have distinctive
pattern.
6
Iris Recognition System
7
Iris Imaging

• Distance up to 1 meter
• Near-infrared camera
• Mirror

8
Imaging Systems
http//www.iridiantech.com/
9
Imaging Systems
http//www.iridiantech.com/
10
Image Processing

• John Daugman (1994)
• Pupil detection circular edge detector
• Segmenting sclera

11
Rubbersheet Model
Each pixel (x,y) is mapped into polar pair (r,
).
Circular band is divided into 8 subbands of
equal thickness for a given angle .
Subbands are sampled uniformly in and in r.
Sampling averaging over a patch of pixels.
12
Encoding
2-D Gabor filter in polar coordinates
13
IrisCode Formation
Intensity is left out of consideration. Only
sign (phase) is of importance.
256 bytes 2,048 bits
14
Measure of Performance

• Off-line and on-line modes of operation.

Hamming distance standard measure for
comparison of binary strings.
x and y are two IrisCodes is the notation
for exclusive OR (XOR) Counts bits that
disagree.
15
Observations

• Two IrisCodes from the same eye form genuine
  pair gt genuine Hamming distance.
• Two IrisCodes from two different eyes form
  imposter pair gt imposter Hamming distance.
• Bits in IrisCodes are correlated (both for
  genuine pair and for imposter pair).
• The correlation between IrisCodes from the
  same eye is stronger.

Strong radial dependencies Some angular
dependencies
16
Observations
Read J. Daugmans statement with caution.
Interpret correctly.
The fact that this distribution is uniform
indicates that different irises do not
systematically share any common structure. For
example, if most irises had a furrow or crypt in
the 12-o'clock position, then the plot shown here
would not be flat.
URL http//www.cl.cam.ac.uk/users/jgd1000/indepe
ndence.html
17
Measure of Performance
Hamming distance standard measure for
comparison of binary strings.
x and y are two IrisCodes is the notation
for exclusive OR (XOR) Counts bits that
disagree.
XOR
Example 1 0 0 0 0 1 1 0 0 0 1 1 1 1 1
1 0 1 0 1 0 0 0 0 0 1 1 1 0 1
0 0 1 0 1 1 1 0 0 0 0 0 0 1 0
18
Training Sets
M users (2 iris images per user) ex.
M 10,000
Genuine Set (userm_iris1userm_iris2), m
1,,M. Compute M genuine Hamming distances.
Imposter Set Formed from combination of irises
from different users (userk_iris1,userl_iris1)
(userk_iris1,userl_iris2) (userk_iris2,userl_iri
s1) (userk_iris2,userl_iris2) k is not equal
to l, k,l 1,,M. Compute imposter Hamming
distances.
19
Degrees of Freedom
Imposter matching score - normalized histogram
- approximation curve - Binomial with 249
degrees of freedom Interpretation Given a
large number of imposter pairs. The average
number of distinctive bits is equal to 249.
20
Histograms of Matching Scores
Decidability Index d-prime d-prime 11.36
The cross-over point is 0.342 Compute FMR and
FRR for every threshold value.
21
Decision
The same eye distributions depend strongly on the
quality of imaging.
Non-ideal conditions
- motion blur - focus - noise - pose variation
- illumination
22
Decision
Ideal conditions
Imaging quality determines how much the same iris
distribution evolves and migrates leftwards.
d-prime for ideal imaging d-prime 14.1
d-prime for non-ideal imaging (previous slide)
d-prime 7.3
23
Error Probabilities
Biometrics Personal Identification in Networked
Society, p. 115
24
False Accept Rate
For large database search - FMR is used in
verification - FAR is used in identification
Adaptive threshold to keep FAR fixed
25
Test Results
The results of tests published in the period
from 1996 to 2003.
Be cautious about reading these numbers The
middle column shows the number of imposter pairs
tested (not the number of individuals per
dataset).
http//www.cl.cam.ac.uk/users/jgd1000/iristests.pd
f
26
Performance Comparison
UK National Physical Laboratory test report, 2001.
http//www.cl.cam.ac.uk/users/jgd1000/NPLsummary.g
if
27
Performance Comparison
Best-of-3 error rates
UK National Physical Laboratory test report, 2001.
28
Other Systems
R. Wildes et al. System 1. Image Acquisition
- 256 pixels across diameter - 20 cm
distance - diffuse source, circular
polarization, and a low-light level camera 2.
Iris Localization - image is transformed into
a binary edge-map - contour fitting using
Hough transforms 3. Pattern Matching -
alignment of two patterns - representation (a
Laplacian pyramid) - goodness of match
(estimate of correlation coefficient) -
Fishers linear discriminant Iris
Recognition An Emerging Biometric Technology,
Proc. of the IEEE, 1997.
29
Fraud Protection
1. Hippus steady-state small oscillations of
pupil size at about 0.5 Hz. 2. Tracking eyelid
movements. 3. Examining ocular reflections (4
optical surfaces - 4 reflections). 4. 2D
Fourier spectra (printers dot in artificial
irises).
30
References
1. J. Daugmans web site. URL
http//www.cl.cam.ac.uk/users/jgd1000/ 2. J.
Daugman, High Confidence Visual Recognition of
Persons by a Test of Statistical Independence,
IEEE Trans. on Pattern Analysis and Machine
Intelligence, vol. 15, no. 11, pp. 1148 1161,
1993. 3. J. Daugman, United States Patent No.
5,291,560 (issued on March 1994). Biometric
Personal Identification System Based on Iris
Analysis, Washington DC U.S. Government Printing
Office, 1994. 4. J. Daugman, The Importance of
Being Random Statistical Principles of Iris
Recognition, Pattern Recognition, vol. 36, no.
2, pp 279-291. 5. R. P. Wildes, Iris
Recognition An Emerging Biometric Technology,
Proc. of the IEEE, vol. 85, no. 9, 1997, pp.
1348-1363. 6. Y. Zhu, T. Tan, and Y. Wang,
Biometric Personal Identification Based on Iris
Patterns, ACTA AUTOMATICA SINICA , No.1, 2002.