Biometrics

spectacularscarecrowΤεχνίτη Νοημοσύνη και Ρομποτική

17 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

55 εμφανίσεις

Biometrics

James Nelson

Overview


Abstract


Introduction


Expository


Fingerprint Scanning


Facial Recognition


Hand Geometry


Vein Pattern Recognition


Iris Scanning


Retina Scanning


Gait Recognition


Voice Recognition


Experiment


Accomplishments,
Redirections & False Starts


How Someone Might
Continue


Conclusion

Abstract


Biometrics is extremely effective in security
applications and is used in multiple instances that
include the government, military and business. It is
also a field that is rapidly growing and expanding.
This project will look at various biometric methods
including iris scanning, retina scanning, hand
geometry, facial geometry, fingerprint scanning and
others to see how they are done, and their general
effectiveness. My hypothesis is that biometrics is the
future of security and that it is coming into the
mainstream where even more people will be able to
take advantage of the technology.

Introduction


Focus on security in technology


Up to this point


“What you know”


Passwords, PINs


“What you have”


Smart Card, Remote


Biometrics is different


“What you are”


Utilizes a physical or behavioral characteristic of a user’s body
for identification

Introduction


With the potential accuracy of these methods,
biometrics should be used in all situations, right?


Not necessarily


Following are descriptions, advantages,
disadvantages and real
-
life situations for biometrics


A biometric device, the SanDisk Cruzer Profile, will
have its everyday use detailed

Project Goals


Provide an overview of biometric technology


Demonstrate the benefits of biometrics


Make readers aware of the potential pitfalls that can
be caused by biometrics


Evaluate the potential benefit of a biometric device for
Juniata faculty and staff

Fingerprint Scanning


Well known


Based on two assumptions:


Fingerprint details are
permanent


Fingerprints of an individual
are unique


“A fingerprint is made up of a
pattern of ridges and furrows
as well as characteristics
that occur at Minutiae points
(ridge bifurcation or a ridge
ending)” (Beal).

Fingerprint Scanning


Scan the fingerprint of the subject with a built
-
in
sensor


It may take a picture of the fingerprint, or it may take
points of the fingerprint and store a “template” rather
than the whole fingerprint


Often linked with other authentication methods such
as PINs or Smart Cards

Fingerprint Scanning


Advantages


Template uses little space


Easy registration process


Little interaction needed


Disadvantages


User hesitancy


Dirty, cut fingerprints


Easier to fool

Fingerprint Scanning


Real
-
World Application


Child Care Facilities and Schools


Access control


Allows staff to do tasks other than door monitoring


Can be used for time accounting

Facial Recognition


Two
-
Dimensional


System is designed to find close matches to particular facial
features such as eyes, nose, mouth, cheekbones, chin and
forehead. These features are then compared to a database
of static images (Reynolds).


Three
-
Dimensional


“Technology captures facial images by using a number of
digital cameras positioned around the subject's face or by
using a structured light grid that captures facial
-
structure
data” (Greenemeier).

Facial Recognition


Advantages


Able to pick out a face in the crowd


When match is made, alert the proper people


Disadvantages


Performance decreases linearly as the enrollment database
size goes up


Two
-
dimensional easily fooled


“Big Brother”


Facial Recognition


Real
-
World Application


Tampa Bay Police Department


Airport and Border Security

Hand Geometry


Users align their hand according to guide marks on a
hand geometry reader


The reader then captures a three
-
dimensional image
of the user’s hand, including the fingers and knuckles


This data is then stored in a template in the system

Hand Geometry


Advantages


Robustly built


Easy to use


20 byte template


Disadvantages


Scanners must be large enough to hold a spread out human
hand


Small amount of data points collected by the scanner can
lead to higher false positives and false negatives than some
other biometric technologies


Hygiene perception


Hand Geometry


Real
-
World Application


Scott Air Force Base


Access control system


100% uptime


2.7% false negative


0.0% false positive


As good as or better at access control than human security
guards


Allowed guards to tend to other duties


Vein Pattern Recognition


Scanner has a CCD camera built
-
in, along with LEDs that
project near
-
infrared waves


“The high absorbance rate of near
-
infrared wave length of
hemoglobin in the blood vessel enables the finger vein pattern to
be acquired by CCD camera” (Hitachi).

Vein Pattern Recognition


Advantages


Different even for identical twins


Not easily observed


Stable, robust


Disadvantages


Hand deformations


Hygiene perceptions

Vein Pattern Recognition


Real
-
World Application


Lora SE210 Laptop


Japanese ATMs


Future vehicle door handles

Iris Scanning


The iris, or the colored part of the human eye, has
unique characteristics that make it a viable source for
identification


Many biometric methods are said to have 13 to 60
distinct characteristics, the iris is believed to have 266
unique spots


It is also believed that every iris is unique and stays
the same over time and across environments

Iris Scanning


A user’s iris is photographed with a black and white, high
-
resolution camera


Once the image is collected, analysis and processing of a
connecting tissue in the iris, called the trabecular meshwork


Differentiating characteristics of the iris include ligaments, furrows,
ridges, crypts, rings, corona, and freckles


Create an “optical ‘fingerprint’” (GlobalSecurity.org)


The optical fingerprint is put into a digital form so it can be stored
within a database

Iris Scanning


Advantages


Highly protected, internal organ


Can be imaged from a reasonable distance


High degree of randomness


Disadvantages


Relatively small target


Moving target


Eyelashes, eyelids and lenses can also obscure the iris

Iris Scanning


Real
-
World Application


Freehold Borough School District in New Jersey

Retina Scanning


Sometimes mistaken for iris scanning, but has
nothing to do with the iris


The retina is the patterns of blood vessels on the thin
nerve at the back of the eyeball that processes light
entering the pupil


The idea that retinal patterns could be used as
unique identifiers was brought up in research during
the 1930’s


Even with this knowledge, the first retinal scanner
was not commercially produced until 1984


Eyedentification 7.5

Retina Scanning


During a retinal scan, a user brings their eye very
close to the lens of the camera being used for the
scanning


Must look directly into the lens


Enrollment can also take up to a minute or more


Must be scanned every time for authentication

Retina Scanning


Advantages


Highly accurate


Patterns remain stable


If killed for their retina patterns, a deceased person’s retina
deteriorates too quickly to be of any use to an attacker


No known way to duplicate a retina


Disadvantages


Diseases such as glaucoma, diabetes, high blood pressure,
and Autoimmune Deficiency Syndrome can affect retina
patterns


Demanding enrollment process

Retina Scanning


Real
-
World Application


Nuclear Power Plants


Missile Silos


Other high security installations


Gait Recognition


First method


The first method uses a radar system, very similar to the kind that is
used by police officers trying to catch speeders on highways.


This looks at the “gait cycle” that is formed by a person’s
movements over time (Cristol).


As a person’s upper and lower leg, knee, and foot move while
walking, the radar is keeping track of the continuous cycle of
movement.


This creates the “gait cycle.”


Second method


Measure static characteristics such as a person’s leg length


Can be viewed from many angles

Gait Recognition


Advantages


In an ideal state, it would be able to identify subjects up to
500 feet away, under any conditions


Can be coupled with other biometrics


Non
-
intrusive


Disadvantages


Public perception of bringing together facial recognition and
gait recognition technology, creating a “24/7 human
identification system” (Koerner)


Not fully reliable

Gait Recognition


Real
-
World Application


None found


Attribute this to the unreliable nature of the technology


Most information pointed to a potential commercial
availability time of 2008

Voice Recognition


May be confused with speech recognition, but they
are not the same


Speech recognition technology is used to convert
spoken words into text


Voice recognition takes the unique sound of a voice
and matches it to a particular person, authenticating
that person for entry into a system (Briggs)

Voice Recognition


Users are usually made to speak a short pass phrase


“It also could be a set of phrases spoken to a challenge or
prompt, such as ‘What is the name of your favorite pet?’”
(Briggs)


If it's responding correctly, the technology will attempt to match
your utterance's timbre and pattern of speech against the
pattern stored in the system in digital form


This comparison is used to create a mathematical score using
algorithms to tell if there is a high confidence or low confidence
in the authenticity of the user’s voice

Voice Recognition


Advantages


Hands
-
free


Ability to be mobile


Disadvantages


Need relatively quiet environment


Distance between device and user

Voice Recognition


Real
-
World Application


University of California Davis Medical Center

Experiment


SanDisk Cruzer Profile 512 MB USB Flash Drive


Information about the device can be found at SanDisk.com


Purchased from NewEgg.com


If the Cruzer was successful during testing, it could be
implemented as a secure way to transfer more sensitive files for
faculty and staff across campus, or to and from their homes

Experiment


First Plugin


Enrollment Tutorial


Fingerprint Enrollment


Two drives appear in My Computer


System tray icon appears


User can now use the drive at any computer


Profile program is contained on USB drive


Window will pop up asking to open the drive through
the Profile program


Swipe finger and access is obtained

Experiment


CruzerPass


Login storage for web sites


Only works with IE


CruzerLogin


For logging in to Windows machines


Program must be installed on computer

Experiment


Trying to “break” the scanner


Dirty fingerprints


Wet fingerprints


Daily wear and tear


Recommendation


Could be an excellent product for a number of faculty and
staff on campus

Accomplishments, Redirections and

False Starts


Accomplished my project goals


Redirections


New expository format


Explanation


Pros


Cons


Real
-
World Application


Second
-
half schedule


False starts


Removed DNA Scanning and societal impact sections

How Someone Might Continue


DNA Scanning


Fingerprint reader and timekeeping software

Conclusion


Biometric methods are the future


Security is about striking balances


All security methods will have issues


Education on biometric methods is key


Be ready to see biometrics, as it may usher in a new
era for technology