Biometric sytems.ppt

Signal and Image
Processing in Biometric
System Applications
Dr. R. Kalpana
Dean, Dpt. Of Biomedical
Engineering
Rajalakshmi Engineering College
Chennai
1

Topics








Introduction
Importance and significance
Different Biometrics
Different processing techniques
Matching and Decision
Governing bodies
Limitatons and Future scope

2-2

Introduction


It is an authentication system – replaces existing systems
like pins, signature etc.

2-3

Introduction cntd…

2-4

Importance and
Significance










More secured, cannot be stolen and changed
Unique, time invariant, acceptable, available
Systems has vendor specific algorithms and
hence proprietoryship
Much useful in public benefit schemes
Can be used in logical and physical networks
Enrollment and presentation – two stages of
operation
Template features
2-5

Biometric parameters












Voice
Infrared facial
thermography
Fingerprints
Face, Iris, Ear
EKG, EEG
Odor, Giat
keystroke dynamics, DNA
Signature,Retinal scan
Hand & finger geometry
Subcutaneous blood
vessel imaging

Various sensors
Ultra sound, thermal,
capacitive –
fingerprints
Camera, Mic, light
source

2-6

Accuracy





False match rate
False nonmatch rate
Failure to enrollment rate\
No single metric indicates how well a biometric system or
device performs: Analysis of all three metrics is necessary
to assess the performance of a specific technology.

2-7

Fingerprints
1. Preprocessing
2. Gabour filters
3. Enhancement, Binarisation and
thinning
4. Feature extraction
5. Template creation
6. Final presentation and comparision

2-8

Minutiae detection and post processing

Two main post-processing
types:
Structural postprocessing
Minutiae filtering in
the gray-scale domain
2-9

Iris recognition
Sensing – To
approaches
Daugmann and
Wildes et al.,
Light source
Localisation
Representation

2 - 10

Typical example

2 - 11

Iris localisation
Daugmann approach Gradient ascent
algorithm

Wildes et al., uses histogram based
and Hough transform

2 - 12

Matching – Iris and System performance




Correspondance
Shift, scaling, rotation, pupil dilation
Match goodness
Hamming distance – Daugmann
Normalisation correlation – Wields
 Decision







Bionomial distribution
Fisher’s linear discriminant function

System Performance
Performed by IriScan or Iridian
They generate FMR and FRR score
2 - 13

2 - 14