3. Biometrics Overview
Definition:
Automated methods of identifying or verifying the identity of a
person based on a physiological or behavioral characteristic
Better than traditional method.
Physical:
Fingerprint
Facial Recognition
Hand Geometry
Iris Scan
Retinal Scan
DNA
Behavioral:
Speaker Recognition
Signature Recognition
4. NEED
Traditional methods have some flaws
To secure access to building, computer system,
laptop, phone, ATM, bank account, & many more.
For security & reduce financial fraud.
To prevent fraudulent acts like faking ID badges or
licenses.
Types of biometric:
1. single model biometric system.
2. multi model biometric system.
5. WORKING PRINCIPAL
1.Enrollment
Adding biometric information to data file.
2.Input Device
Sensors, cameras etc
3. Comparison with the database.
4. Decision
7. Biometrics Overview
Ways to recognize a person:
Verification
Matching against a single record.
It is 1:1 matching
Identification
Matching against all record in the database.
It is 1:N matching.
8. Error and Accuracy
2 types of errors:
False rejection: When an actual client gets identified as
imposter.
False acceptance: When an actual imposter gets identified as
client.
Accuracy:
FRR =number of false rejections/number of client accesses
FAR = number of false acceptance/number of client accesses
TER=(no of FA+no of FR)/total no accesses.
9. PHYSICAL MODES
o Fingerprint
o Facial Recognition
o Hand Geometry
o Iris Scan
o Retinal Scan
o DNA
10. Fingerprint
Typical System: fingerprint scanner maps the
series of whorls, ridges, furrows and minutiae on
the surface of the finger
Optical sensor and Solid State sensor are used.
12. FACIAL RECOGNITION
Typical system: digital video camera input of a
person’s face images- measures facial structure;
compares against database
2 Methods:
Face geometry
Facial thermograph: uses infrared heat scans to identify
facial characteristics.
13. HAND GEOMETRY
Analyses and computes features like length and
width of finger, ratio of palm or fingers, width and
thickness of the palm.
Basic procedure involves taking images of top and
side views of hand using camera.
14. Iris Recognition
Iris analysis
Typical system: scanner analyzes features like rings,
furrows, and freckles existing in the colored tissue
surrounding the pupil.
Sophisticated camera uses infrared imager to
capture high resolution photograph.
Advantage:
o Iris is well protected area of eye with minimum possibility of
injury.
15. Retina Scanning
Analysis of layer of blood vessels at the back of the
eye
Typical system: low-intensity infrared light source
and an optical coupler to scan a unique vascular
pattern of retina and its reflection is recorded.
Most accurate and reliable method.
16. DNA (Deoxyribonucleic acid)
A DNA sample is used to produce either a DNA
fingerprint or a DNA profile
Comparison of alleles of DNA sequences found in
nuclear genetic material.
Used in medical field and forensic investigation
purpose.
18. Speaker Recognition
Voice recognition and analysis
Typical system: uses the pitch, pattern, tone,
frequency, rhythm of speech for identification
purposes; only biometric that allows users to
authenticate remotely.
3 types:
Fixed text method
Text dependent method
Text independent method
19. Signature Recognition
Handwriting analysis
Typical System: user will sign on a tablet, the device will
record pressure, speed, sequence and shape of characters
and compare to a database.
Verification is based on features like no. of interior
contours and no. of vertical slope components.
Rarely used method.
20. Comparison Chart Accuracy/Reliability
Biometric Accuracy Reliability Errors
Fingerprint Very High High Dirt, dryness
Facial Recognition High Medium Hair, glasses, age
Hand Geometry High Medium Hand injury
Iris Scan Very High High Poor lighting
Retinal Scan Very High High Glasses
DNA Very High High none
Speaker Recognition Medium Low Noise, colds
Signature Recognition Medium Low Changing signatures
Accuracy: How well can the specific biometric is able to tell individual apart
Reliability: how dependable the specific biometric is for recognition purposes
21. APPLICATIONS
BIOMETRIC TECH APPLICATIONS
FINGERPRINTS Law enforcement, entry devices for offices
and colleges, enterprise security; medical
and financial
FACIAL RECOGNITION Automated bank tellers- user verification
purposes
HAND GEOMETRY Time and attendance systems, physical
access
IRIS SCAN Law Enforcement, Employee Security
Check, banking
22. RETINA SCAN High-end security, military, power plans
DNA Medical applications, Paternity Tests,
Criminal identification and forensics
VOICE RECOGNITION Call Centers, Law enforcement – house
arrest authentication
SIGNATURE Access to documents, Banking services
23. What Biometrics are Replacing
Traditional verification methods:
Passwords
PIN numbers
Tokens
Photo ID cards
Smart cards
Magnetic strip cards
Physical keys
24. Current Biometric Applications
Computer logins
ATMs
Grocery stores
Airport kiosks
Driver’s licenses
Internet banking
National ID cards
Welfare disbursement
International border crossing
Forensics – criminal identification
Annual amusement park pass holders
Speaker verification for television home shopping
25. BENEFITS
Increased security measures
Reduce costs of password and ID card maintenance
Ability to track employees and link activities
No passwords to forget
Elimination of problems such as lost/stolen
identification cards and PIN numbers
Faster login time
Improved security of sensitive information
27. CONCLUSION
Multi biometric systems have a huge potential of
growth.
Access procedures are made simpler, faster and
more secure.
Especially governments, law enforcement agencies,
military and industrial companies, already make
partial use of this technology.
In the future biometric devices will surely become
more involved in many civil areas.
Only Biometrics can verify you as YOU. Tokens – can be lost, stolen, duplicated (some), forgotten. Passwords – can be forgotten, shared, observed, broken.
Link to biometrics used in military: http://www.biometrics.dod.mil/References/BiometricUseCases.aspx