6. IRIS
RECOGNITION
Iris patterns are extremely
complex.
Patterns are individual
Patterns are formed by six
months after birth, stable
after a year. They remain
the same for life.
Imitation is almost
impossible.
Patterns are easy to
capture and encode
8. IRIS SCANNERS
• High resolution cameras capture image from up to 3 feet
away (usually 10 to 12 inches)
• Converts picture of the distinctive fibers, furrows, flecks,
crypts, rifts, pits and coronas of the iris into a bar-code
like identifier
• Template around 256 Bytes in size
• Human iris is distinct with 250 differentiating features
• The recognition of irises by their IrisCodes is based upon
the failure of a test of statistical independence.
– Any given IrisCode is statistically guaranteed to pass
a test of independence against any IrisCode computed
from a different eye; but it will uniquely fail this same
test against the eye from which it was computed.
11. •
FINGER PRINT (continued )
Fingerprint matching techniques can be placed into two categories: minutiaebased and correlation based.
– Minutiae-based techniques first find minutiae points and then map their
relative placement on the finger. However, there are some difficulties
when using this approach.
• It is difficult to extract the minutiae points accurately when the
fingerprint is of low quality.
• Also this method does not take into account the global pattern of
ridges and furrows.
– The correlation-based method is able to overcome some of the difficulties
of the minutiae-based approach. However, it has some of its own
shortcomings.
• Correlation-based techniques (i.e. pattern matching) require the
precise location of a registration point and are affected by image
translation and rotation.
• Larger templates (often 2 – 3 times larger than minutiae-based)
12. FACE RECOGNITION
•Typical Eigenfaces
•Utilizes two dimensional,
•global grayscale images
•representing distinctive
•characteristics of
•a facial image
•Variations of eigenface are
•frequently used as the basis of other
face recognition methods.
13. FACIAL (continued)
•
•
Eigenface: "one's own face," a technology patented at MIT that uses 2D
global grayscale images representing distinctive characteristics of a facial
image. Most faces can be reconstructed by combining features of 100-125
eigenfaces. During enrollment, the user's eigenface is mapped to a series of
numbers (coefficients). Upon a 1:1 match, a "live" template is matched
against the enrolled template to obtain a coefficient variation. This
variation either accepts or rejects the user.
Local Feature Analysis (LFA): also a 2D technology, though more capable
of accommodating changes in appearance or facial aspect (e.g., smiling,
frowning). LFA uses dozens of features from different regions of the face;
incorporates the location of these features. Relative distances and angles of
the "building blocks" of the face are measured. LFA can accommodate 25degree angles in the horizontal plane and 15 degrees in the vertical plane.
LFA is a derivative of the eigenface method and was developed by
Visionics, Corp.
15. FACIAL (continued)
• Varying light (i.e. outdoors) can affect accuracy
• Some systems can compensate for minor changes such as
puffiness and water retention
• Smiling, frowning, etc can affect accuracy
• Some systems can be confused by glasses, beards, etc
• Human faces vary dramatically over long term (aging) and
short term (facial hair growth, different hair styles, plastic
surgery)
• Expected high rate of acceptance as people are already used
to being photographed or monitored
• Best method for identification systems (e.g. airports)
16. VOICE RECOGNITION
• The software
remembers the way
you say each word.
• Voice recognition
possible even though
everyone speaks with
varying accents and
inflection.
• Telephony : the
primary growth area
17. VOICE VERIFICATION
•A complete signal has an
overall pattern, as well as a
much finer structure, called the
frame. This frame is the
essence of voice verification
technology.
•It is these well-formed, regular
patterns that are unique to
every individual. These
patterns are created from the
size and shape of the physical
structure of a person's vocal
tract. Since no two vocal tracts
are exactly the same, no two
signal patterns can be the
same.
18. VOICE VERIFICATION
•These unique features
consist of cadence,
pitch, tone, harmonics,
and shape of vocal tract.
•The image at right
shows how
characteristics of voice
actually involve much
more of the body than
just the mouth.
19. HAND
GEOMETRY
• 32,000-pixel CCD digital
camera .
• The hand-scan device can
process the 3-D images in
less than 5 seconds & the
verification usually takes
less than 1 second.
• U.S INPASS PROGRAM
21. HAND/FINGER
GEOMETRY READERS
• The first modern biometric device was a hand
geometry reader that measured finger length
• These devices use a 3D or stereo camera to map
images of the hands and/or fingers to measure
size, shape and translucency
• Actual sensor devices are quite large in size
• Templates are typically small (approx 10 Bytes)
• High acceptance rate among users
22. SIGNATURE
RECOGNITION
How the signature was
made. i.e. changes in
speed, pressure and
timing that occur during
the act of signing
An expert forger may
be able to duplicate what
a signature looks like,
but it is virtually
impossible to duplicate
the timing changes in X,
Y and Z (pressure)
23. SIGNATURE ANALYSIS
(continued)
•Built-in sensors register the dynamics of the act of writing. These dynamics
include the 3D-forces that are applied, the speed of writing, and the angles in
various directions.
•This signing pattern is unique for each individual, and thus allows for strong
authentication. It also protects against fraud since it is practically impossible to
duplicate "how" someone signs.
24. • A multimodal
biometric system uses
the integration of
biometric systems in
order to meet stringent
performance
requirements.
•Much more vital to
fraudulent technologies
26. CONCLUSION
Once the exclusive preserve
of sci-fi books and movies,
biometrics now has to be
considered as one of the many
challenges of modern day
management.