Iris Recognition: Mahendra Kumaresh.S

IRIS RECOGNITION
The future of identity
by,
Mahendra Kumaresh.S
Biometric Technology
•Verifies/Recognizes identity of a living person.
•Eg:- Iris recognition, face recognition, fingerprint,

voice recognition etc.
•Tremendous accuracy and speed.
•Provides high level of security for sensitive

information.
•Helps prevent fraudant use of ATM’s etc.

Biometric devices have 3 primary components
4. Mechanism to capture image of a living personal

characteristic.
6. Compression ,processing ,storage and comparison

of image with stored data.
8. Interfaces with application systems.

•A biometric system can be divided into 2 stages :
3. Enrolment Module.
5. Identification Module.
• During the enrolment stage:
• Biometric sensor scans the person’s physiognomy

to create a digital representation.
• A feature extractor processes the representation to

generate a template.
•During the identification stage:
•Biometric sensors capture the characteristics of

person to be identified.
•Converts it into the same format as the template.
•The resulting template is fed into a feature matcher.
•Feature matcher compares it against the stored

template to determine whether the two templates
match.
Cost and Accuracy
•Cost is directly related to accuracy and vice-versa.
IRIS
RETINA
COSTSIGNATURE
FINGER
FACE
VOICE
ACCURACY
The Human Iris
•The iris is a membrane in the eye, responsible for
controlling the amount of light reaching the retina .
•The pupil is a circular opening located in the center of

the iris of the eye.
•The white outer area is the sclera. the central

transparent part of which is the cornea.
•The unique pattern on the surface of the iris is
formed during the first year of life.
•Formation of the unique patterns of the iris is

random and not related to any genetic factors.
•Due to the epigenetic nature of iris patterns,

identical twins possess uncorrelated iris patterns.
•The two eyes of an individual contain completely

independent iris patterns.
IRIS RECOGNITION
•The process of identifying
a person by analyzing the
pattern of the iris.
•Provides an unmatched
identification technology.
•Uses highly accurate algorithms with very little

chance of occurrence of error.
•Used in large scale identification implementations.

History of Iris Recognition
• It was in the early nineties that Cambridge researcher,
John Daugman, implemented a working automated iris
recognition system
• The Daugman system is

patented and the rights are
now owned by the company
Iridian Technologies(1994).
• Zero failure rate
• Perfectly identifies a person given millions possibilities.

Why IRIS?
•Data rich physical structure about 400 identifying features
•Genetic independence
•Stability over time
•Inherent isolation and protection from external

environment.
•Impossibility of surgically modifying it.
•Its physiological response to light.

Design and Implementation
Block diagram of Iris Recognition

system
•The design and implementation of the system

is subdivided into 3:
3. Image acquisition
4. Iris localization
5. Patter Matching
Image Acquisition
•Capturing of high quality images of the iris.
•Wide angle cameras are used.
•LED based point light sources

are used for illumination.
•The operator positions in front

of the camera.
•At a distance of not more than 1m.

•The operator is provided with a live video feed.
•Allowing the operator

to see what the camera
is capturing.
•Once a series of images

are captured it is
automatically forwarded .
3 important requisites :
• Images of the iris with sufficient resolution and
sharpness to support recognition.
• Good contrast in the interior iris pattern.
• Images must be well framed .

Iris Localization
• Delimits the iris from the rest of the acquired image
• Daugman’salgorithm is used to narrow in from the

right and left of the iris to locate its outer edge.
• Iris recognition technology converts the visible

characteristics of the iris into a 512 byte Iris Code
• Future comparisons are based on the Iris Code

Iris Localization
Pattern Matching
•Evaluates the goodness of match between the newly

acquired iris pattern and the candidate’s data base entry.
•Hamming Distance - fraction of mismatched bits
•We use one-to-many search of database
•The entire process takes about 2 seconds.

Accuracy
•Accuracy higher than other forms of biometrics.
•Odds of returning 75% match- 1 in 10^16
•Equal error rate 1 in 12 million
•Odds of identical iris codes-1 in 10^52

Independence of bits across iris codes
•No systematic correlation between different irises.
•Probability that bits in different position are set to 1 .
•Distribution hovers over .5

•All bits are equally likely to be 0 or 1.
•Hence we can infer that is no systematic

Correlation between irises.
Advantages
•Small error rates (EER) or high 'accuracy‘.
•No failure to enroll (FTE) .
•Easy for someone to use, not intrusive .
•Fast verification, fast identification (1,00000 /sec).
•Robust against 'countermeasures' and fraud .
•Social acceptance .
Comparison with other biometric
technologies
A-Excellent D-Poor
Iris
Recognition:Issues
•The technology requires a certain amount of user
interaction.
•Unusual lighting situation may affect the ability of the

camera to acquire its subject.
•Backup must be in place if the unit becomes

inoperable.
•But these issues do not affect the effectiveness of the

technology
Applications
•ATM’s
•Tracking prisoner movement
•Computer Login
•Ticket less travel
•Driving license and other

personal certificates.
•Mobile phones.
•Tracking of missing or wanted persons
•Credit card authentication
•Anti –terrorism( eg :- screening at

airports etc)
•Secure financial transaction

Conclusion
•As iris technology grows less expensive, it could very
likely unseat a large portion of the biometric industry.
•Its technological superiority has already allowed it

to make significant inroads into identification and
security venues which had been dominated by
other biometrics
•Iris-based biometric technology has always been an

exceptionally accurate one, and it may soon grow much
more prominent.
THANK YOU

Iris Recognition: Mahendra Kumaresh.S

Uploaded by

Copyright:

Available Formats

Iris Recognition: Mahendra Kumaresh.S

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Iris Recognition: Mahendra Kumaresh.S

Uploaded by

Copyright:

Available Formats

IRIS RECOGNITION

The future of identity

•Eg:- Iris recognition, face recognition, fingerprint,

•Tremendous accuracy and speed.

•Provides high level of security for sensitive

•Helps prevent fraudant use of ATM’s etc.

4. Mechanism to capture image of a living personal

6. Compression ,processing ,storage and comparison

8. Interfaces with application systems.

• During the enrolment stage:

• Biometric sensor scans the person’s physiognomy

• A feature extractor processes the representation to

•Biometric sensors capture the characteristics of

•Converts it into the same format as the template.

•The resulting template is fed into a feature matcher.

•Feature matcher compares it against the stored

•Cost is directly related to accuracy and vice-versa.

•The pupil is a circular opening located in the center of

•The white outer area is the sclera. the central

•Formation of the unique patterns of the iris is

•Due to the epigenetic nature of iris patterns,

•The two eyes of an individual contain completely

•Uses highly accurate algorithms with very little

•Used in large scale identification implementations.

• The Daugman system is

• Zero failure rate

• Perfectly identifies a person given millions possibilities.

•Stability over time

•Inherent isolation and protection from external

•Impossibility of surgically modifying it.

•Its physiological response to light.

Block diagram of Iris Recognition

•The design and implementation of the system

•Wide angle cameras are used.

•LED based point light sources

•The operator positions in front

•At a distance of not more than 1m.

•Allowing the operator

•Once a series of images

• Good contrast in the interior iris pattern.

• Images must be well framed .

• Daugman’salgorithm is used to narrow in from the

• Iris recognition technology converts the visible

• Future comparisons are based on the Iris Code

•Evaluates the goodness of match between the newly

•Hamming Distance - fraction of mismatched bits

•We use one-to-many search of database

•The entire process takes about 2 seconds.

•Accuracy higher than other forms of biometrics.

•Odds of returning 75% match- 1 in 10^16

•Equal error rate 1 in 12 million

•Odds of identical iris codes-1 in 10^52

•Probability that bits in different position are set to 1 .

•Distribution hovers over .5

•Hence we can infer that is no systematic

•No failure to enroll (FTE) .

•Easy for someone to use, not intrusive .

•Fast verification, fast identification (1,00000 /sec).

•Robust against 'countermeasures' and fraud .