Professional Documents
Culture Documents
Motivation
Multimodal
Palm
Biometrics
Provides:
Easy enrolment
Unique parameters
Hard to falsify
Rihards Fuksis
White LEDs
IR LEDs
Rihards Fuksis
CMF ( x, y ) e
G ( x, y, l )
Filtering result
Vector set
Rihards Fuksis
Vector set A
Vector set B
Rihards Fuksis
Vector comparison
v p ( A) vq (B )
Rihards Fuksis
Magnitud
es:
v p ( A) vq ( B )
Vector comparison
v p ( A) vq (B )
Magnitud
es:
r
r
v p ( A) vq ( B )
Rihards Fuksis
Vector comparison
v p ( A) vq (B )
Magnitud
es:
r
r
v p ( A) vq ( B )
2
2
||
Rihards Fuksis
Vector comparison
v p ( A) vq (B )
Magnitud
es:
r
r
v p ( A) vq ( B )
Distance:
exp
s p ,q
d||2
d 2
2 exp 2
||
r
r
r
r
v p ( A) vq ( B) cos
(v p ( A); vq ( B)) exp
Dot product
Rihards Fuksis
d||2
2 exp
||
d 2
2
r
v p ( A) vq ( B ) cos
(v p ( A); vq ( B)) exp
Similarity of two
vector sets:
d||2
2 exp
||
d 2
2
s ( A, B ) s p ,q
p
s ( A, B)
Rihards Fuksis
s ( A, B)
s ( A, A) s( B, B)
International Conference on Hand-based Biometrics
1
1
0
0
0 0 1 1
1 1 1 0
Encrypted
data
1
0
0
0
1
0
Raw biometric
data
Rihards Fuksis
Biohash
du
Pixels
CMF
Vectors
(u,v)
Palm image
Inner
product
Inner
product
...
1st vector
u1
v1
R-th vector
uR
vR
Inner
product
...
Random
number
matrix
du1 dv1
...
Token
dv
Vector Set
duR dvR
...
Thresholding
1
...
Rihards Fuksis
Biocode
Biocode consists of
4R bits
Biohash Advancements(I)
Rihards Fuksis
Biohash Advancements(I)
76
27
187
49
83
163
87
44
52
85
146
83
41
57
87
51
Rihards Fuksis
Biohash Advancements(I)
76
27
187
49
83
163
87
44
52
85
146
83
41
57
87
51
Rihards Fuksis
Biohash Advancements(I)
76
27
187
49
83
163
87
44
52
85
146
83
41
57
87
51
Data vector
u1
v1
vR
Rihards Fuksis
duR dvR + 0
1 ... 0
Biohash Advancements(I)
76
27
187
49
83
163
87
44
52
85
146
83
41
57
87
51
Data vector
u1
v1
vR
duR dvR + 0
1 ... 0
Biohash Advancements(II)
..
1 1 0 .
1 0 1
Random
number
matrix
..
.
..
1 1 0 .
4 3 2
..
1 1 1 .
Data
vector
Dot
product
Rihards Fuksis
Thresholding
International Conference on Hand-based Biometrics
Biohash Advancements(II)
Bit #1
Bit #2
0.61
1 0.33 0 0.04
...
0.47
0 0.12 1 0.15
...
0.59
1 0.47 0 0.18
...
0.46
1 0.39 1 0.14
...
2.13
3 1.19 2 0.29
...
Distance to
the threshold
Bit #1
Bit #2
Bit #3
Rihards Fuksis
..
.
Bit #3
Biohash Advancements(II)
Distance to
the threshold
Bits
Sort bits into groups
Distance to
the threshold
4
Rihards Fuksis
2
International Conference on Hand-based Biometrics
Biohash Advancements(II)
Distance to
the threshold
2
Sort bits in every group
in ascending order
Distance to
the threshold
4
Rihards Fuksis
2
International Conference on Hand-based Biometrics
Biohash Advancements(II)
Distance to
the threshold
Rihards Fuksis
Biocode comparison
= 4 mistakes
Similarity:
l Dh
Sb
l
Rihards Fuksis
Database evaluation
Two databases; 500 images from 50 persons
5 images in IR and 5 in visible light spectrum
Raw biometric data comparison
results[EER]
Palm
Palm
Fused
Veins
Prints
data
EER [%]
0.32
2.79
0.1
Biohash test results
Palm [EER]Palmprint Fused
Veins
s
data
Mean
14.043
12.073
6.190
StDev
1.152
Proposed
[EER]
Palm
Veins
1.102
Biohash test
Palmprint
s
0.803
results
Fused
data
Mean
1.073
0.471
StDev
0.304
0.231
Rihards Fuksis
Conclusions
Complex 2D Matched Filtering approach
speeds up the feature extraction procedure.
Biohashing with proposed advancements
can be used as a method for securing the
biometric data with similar or better
precision as raw biometric data comparison
gives
Future work: Tests on larger databases and
evaluation of other biometric encryption
methods
Rihards Fuksis
Thank you!
Questions?
This presentation was supported by ERAF
funding under the agreement
No.2010/0309/2DP/2.1.1.2.0/10/APIA/VIA/01
2
Rihards Fuksis