Biometrics and Sensors

1
Biometrics and Sensors

• Venu Govindaraju
• CUBS, University at Buffalo
• govind_at_buffalo.edu

2
Organization

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Signature Verification
• Hand Geometry
• Multimodal biometrics
• Securing Biometric Data
• Sensors and Devices

3
Research at UB

• Multimodal Identification
• Biometrics
• Fingerprint
• Signature
• Hand Geometry
• Sensors
• Materials and Light Sources
• Analog VLSI and Optical Detectors
• Packaging and Reliability Engineering

4
Applications And Scope of Biometrics
5
Scope of Research In Biometrics
6
Biometrics

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Signature Verification
• Hand Geometry
• Multimodal biometrics
• Securing Biometric Data
• Sensors and Devices

7
Conventional Security Measures

• Token Based
• Smart cards
• Swipe cards
• Knowledge Based
• Username/password
• PIN
• Disadvantages of Conventional Measures
• Tokens can be lost or misused
• Passwords can be forgotten
• Multiple tokens and passwords difficult to manage

8
Biometrics

• Definition
• Biometrics is the science of verifying and
  establishing the identity of an individual
  through physiological features or behavioral
  traits
• Examples
• Physical Biometrics
• Fingerprint, Hand Geometry,Iris,Face
• Behavioral Biometrics
• Handwriting, Signature, Speech, Gait
• Chemical Biometrics
• DNA, blood-glucose

9
Fingerprint Verification

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Signature Verification
• Hand Geometry
• Multimodal biometrics
• Securing Biometric Data
• Sensors and Devices

10
Fingerprint Verification
Fingerprints can be classified based on the ridge
flow pattern
Fingerprints can be distinguished based on the
ridge characteristics
11
Fingerprint Image Enhancement

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

High contrast print
Typical dry print
Low contrast print
Typical Wet Print
12
Traditional Approach

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

Local Orientation ?(x,y) Gradient Method
Enhancement Frequency/Spatial
Local Ridge Spacing F(x,y) Projection Based Method
13
Fourier Analysis Approach

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

Energy Map E(x,y)
FFT Analysis
Orientation Map O(x,y)
FFT Enhancement
Ridge Spacing Map F(x,y)
14
Fourier Analysis Applied to fingerprints
Fingerprint ridges can be modeled as an oriented
wave
15
Fourier Analysis Energy Map

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

Original Image
Energy Map
Thresholded Map
16
Fourier Analysis Frequency Map

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

Original Image
Local Ridge Frequency Map
17
Fourier Analysis-Orientation Map

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

Local Ridge Orientation Map
Original Image
18
FFT Based Enhancement

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

Original Image
Enhanced Image
19
Common Feature Extraction Methods

• Thinning-based Method
• Thinning produces artifacts
• Shifting of Minutiae coordinates

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

• Direct Gray-Scale Extraction Method
• Difficult to determine location and orientation
• Binarized Image is noisy.

20
Chaincoded Ridge Following Method

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

21
Minutiae Detection

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

• Several points in each turn are detected as
  potential minutiae candidate
• One of each group is selected as detected
  minutiae.
• Minutiae Orientation is detected by considering
  the angle subtended by two extreme points on the
  ridge at the middle point.

22
Pruning Detected Minutiae

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

• Ending minutiae in the boundary of fingerprint
  images need to be removed with help of FFT Energy
  Map
• Closest minutiae with similar orientation need to
  be removed

23
Secondary Features

• Pure localized feature
• Derived from minutiae representation
• Orientation invariant
• Denote as (r0, r1, d0, d1, ?)
• r0, r1 lengths of MN0 and MN1
• d0, d1 relative minutiae orientation w.r.t. M
• ? angle of N0MN1

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

24
Dynamic Tolerance Areas

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

• Tolerance Area is dynamically decided w.r.t. the
  length of the leg.
• Longer leg Tolerates more distortion in length
  than the angle.
• Shorter leg tolerates less distortion in length
  than the angle.

Dynamic Windows
Dynamic tolerance
25
Feature Matching

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

• For each triangle, generate a list of candidate
  matching triangles
• To recover the rotation between the prints. Find
  the most probable orientation difference
• Apply the results of the pruning and match the
  rest of the points based on the reference points
  established.

26
Validation

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

OD0.7865

• For each triangle, generate a list of candidate
  matching triangles
• To recover the rotation between the prints. Find
  the most probable orientation difference
• Apply the results of the pruning and match the
  rest of the points based on the reference points
  established.

27
Minutia Matching

• Preprocessing
• Enhancement
• Feature Extraction
• Matching

• For each triangle, generate a list of candidate
  matching triangles
• To recover the rotation between the prints. Find
  the most probable orientation difference
• Apply the results of the pruning and match the
  rest of the points based on the reference points
  established

28
Data Sets
29
Preliminary Results

• Min Total Error 1.16
• ERR 1.0
• FRR at 0 FAR 5.0

Threshold
FAR
FRR
State of the art

• Min Total Error 0.19
• FRR at 0 FAR 0.38

30
Signature Verification

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Signature Verification
• Hand Geometry
• Multimodal biometrics
• Securing Biometric Data
• Sensors and Devices

31
Signature Verification
Online Signature verification
Off line Signature Verification
32
Preprocessing

• Preprocessing
• Make signature invariant to scale, translation
  and rotation.

• Preprocessing
• Template generation
• Matching

(-170)- (-125)
(-3.0)- (4.0)
mean-std norm.
Resampling
-1.5-3.5
0-160
Smoothing
33
Template Generation- Challenges

• Preprocessing
• Template generation
• Matching

• Extracting features.
• Usually we can not expect more than 6 genuine
  signatures for training for each subject. This is
  unlike handwriting recognition
• Decide the consistent features.
• There are over 100 features for signature, such
  as Width, Height, Duration, Orientation, X
  positions, Y positions, Speed, Curvature,
  Pressure, so on.

34
Matching Similarity Measure

• Simple Regression Model

• Preprocessing
• Template generation
• Matching

Y (y1 , y2 , , yn) X (x1 , x2 , , xn)
Similarity by R2 91
Similarity by R2 31
35
Traditional Regression approach

• Advantages Invariant to scale and translation.
  Similarity (Goodness-of-fit) makes sense.
• Disadvantages One-one alignment, brittle.

• Preprocessing
• Template generation
• Matching

One-One alignment
Dynamic alignment
36
Dynamic Regression approach(1)

• Preprocessing
• Template generation
• Matching

( y2 is matched x2, x3, so we extend it to be two
points in Y sequence.)

• Similarity R2

Where (x1i, y1i, v1i) are points in the
sequence And a, b, c are the weights, e.g., 0.5,
0.5, 0.25

• DTW warping path in a n-by-m matrix is the path
  which has min cumulative cost.
• The unmarked area is the constrain that path is
  allowed to go.

37
Offline Signature Verification

• Shapes can be described using structural or
  statistical features
• We use an analytical approach that uses the
  attributes of structures.

38
Attributes of structural features
Statistical analysis of the feature attributes
39
Hidden Markov Models and SFSA

• The occurrence of the structural features can be
  modeled as a HMM
• The HMM can be converted to a SFSA by assigning
  observation and probability to the transitions
  instead of to the states

40
Hand Geometry

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Signature Verification
• Hand Geometry
• Multimodal biometrics
• Securing Biometric Data
• Sensors and Devices

41
Hand Geometry

• Used where Robustness, Low cost are the concerns.
• Comparatively less accurate.
• Combination with other Biometric techniques,
  increases accuracy.
• Sufficient for verification where finger print
  use may infringe on privacy.

42
Feature Extraction

• A snapshot of the top and side views of the
  users right hand gives the contours outlining
  the hand.
• Features necessary to identify the hand are
  extracted from these contours. Using simple image
  processing techniques, the contours of the set
  of two images of the hand are obtained.
• Hand-verification is done by correlating these
  features.
• Research New features and algorithms for better
  discrimination between two hands.

43
Multimodal Biometrics

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Signature Verification
• Hand Geometry
• Multimodal biometrics
• Securing Biometric Data
• Sensors and Devices

44
Combination of biometric matchers
Combination of the matching results of different
biometric features provides higher accuracy.
45
Sequential combination of matchers
Fingerprint matching
Combination algorithm 1
No
Desired confidence achieved?
Yes
Signature matching
Combination algorithm 2
Yes
No
Desired confidence achieved?
Hand geometry matching
Combination algorithm 3
46
Securing Biometric Data

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Signature Verification
• Hand Geometry
• Multimodal biometrics
• Securing Biometric Data
• Sensors and Devices

47
Securing password information
It is impossible to learn the original password
given stored hash value of it.
48
Securing fingerprint information
Wish to use similar functions for fingerprint
data
49
Obstacles in finding hash functions
Fingerprint space
Hash space
f1
h(f1)
h
f2
h(f2)

• Since match algorithm will work with the values
  of hash functions,
• similar fingerprints should have similar hash
  values
• rotation and translation of original image
  should not have big impact on hash values
• partial fingerprints should be matched

50
Sensors and Devices

• Biometrics and Sensor research at UB
• Biometrics
• Fingerprint Verification
• Securing Biometric Data
• Signature Verification
• Hand Geometry
• Sensors and Devices

51
Sensors and Biometrics
52
Sensors
Detector System

• CMOS
• CCDs
• Photodiodes
• Image Processing

• Biosurfaces - Biofouling
• Bioinspired Pattern Recognition
• Biomimetics Artificial Vision, Smell.
• Bioinspired Super Correlator

Analyte
Sensing Layer

• Tissues
• Cells
• Proteins
• DNA and RNA
• Organic and Inorganic Dyes
• Molecular Imprinting

• Biosurfaces Biofouling
• Immobilization and Stabilization
• Transduction mechanism
• Multi-Analyte detection
• Photonic Bandgap (PBG) Resonators
• Evanescent Wave Devices (PBG)

Stimulator and Support System

• Light Sources (OLEDs, LEDs, Lasers)
• Signal Generators
• Driver Circuits
• Power Supply

• Biosurfaces Biofouling
• Nano-LEDs
• Bioinspired Photovoltaics, Biofuel Cells
• Environmental Testing
• Low Power Light Sources

c) Device
b) Enabling Technologies
a) Fundamental Knowledge
53
Sensor Components
Stimulator and Support System
Sensing Layer
Detector System
Blocking Filter
Output Device
54
CMOS Integrated Sensor System
55
Sensor System Components
60 mm
1.2 mm thick
56
PIXIES Protein Imprinted Xerogels with Integrated
Emission Sites

• The sensors selectively respond to Ovalbumin
• Orders of magnitude greater than other components
• Each site can individually respond to different
  analytes

57
Summary

• A unique collaborative initiative that enables
  state-of-the-art Biometric Science and
  Technology
• Creating a multi-disciplinary environment
  attracting faculty and students from engineering
  and sciences
• Preparing and educating future Biometric
  Scientists and Engineers
• Targeting all the aspects of Biometrics from
  authentication to materials and including them
  into a packaged device

58
Acknowledgements
Websites

• www.cubs.buffalo.edu
• www.photonics.buffalo.edu
• www.cedar.buffalo.edu
• www.packaging.buffalo.edu

• Financial support of
• National Science Foundation (NSF)
• Office of Naval Research (ONR)
• Calspan UB Research Center (CUBRC)
• University at Buffalo Center for Advanced
  Technology (UBCAT)

59
Thank You