Retinal Recognition

Retinal Recognition

JITENDRA KUMAR ROUT

Introduction

Retinal Recognition of a person is done by acquiring an
internal body image, the retina of a person .
Unlike other biometric technologies retinal recognition is
not widely deployed in commercial applications.
 While considered invasive and expensive, retinal
recognition is still the most reliable and stable means of
biometric identification.
Although the advantages of retinal recognition currently
outweigh the disadvantages, its widespread use is held
back by public acceptance.

Retina

The retina is a thin layer of cells at the back of the
eyeball of vertebrates.
It is the part of the eye which converts light into nervous
signals
 The retina consists of multiple layers of sensory tissue
and millions of photoreceptors(cells) whose function is to
transform light rays into neural impulses.
 These impulses subsequently travel to the brain via the
optic nerve, where they are converted to images.

Retina

Two distinct types of photoreceptors exist within the
retina: the rods and the cones. While the cones (6-
million per eye) help us to see different colors, the rods
(125 million per eye) facilitate night and peripheral
vision.
It is the unique structure of the blood vessel pattern in
the retina that forms the foundation for retinal recognition
and has been used for biometric identification.

Structure of rods and cons

Rods sense
To brain brightness

Cones sense
color

The retina, in
the back of our
eye, has cells
that are
sensitive to
light. They
connect
directly to your
brain.

RETINA vs. IRIS

When talking about the eye, especially in relation to
biometrics, the iris and the retina are often confused.
While they may both be categorized as ‘eye biometrics’, their
respective functions are completely different.
The iris is the colored band of tissue that surrounds the pupil of
the eye.
The primary purpose of the iris is to dilate and constrict the
size of the pupil. In this sense, the iris is analogous with the
aperture of a camera.
The retina is a thin layer of cells at the back of the eyeball of
vertebrates. It is said that the retina “is to the eye as film is to a
camera.”

Side view of the eye

 The iris is located in the front of the eye, while the retina is located at the back.
Because of its position within the eye, the retina is not exposed to the external
environment. As a biometric, it is therefore very stable.

Front view of the blood vessel pattern within the
retina

 The red lines represent the actual blood vessels; the yellow
section indicates the position of the optic disc (the place where
the optic nerve joins the retina). It is from here that information is
transmitted to and received from the brain. The circle in the
diagram indicates the area that is typically captured by a retinal
scanning device. It contains a unique pattern of blood vessels.

Retina as Human Descriptor
There are two famous studies that have confirmed the
uniqueness of the blood vessel pattern found in the retina.
The first was published by Dr Carleton Simon and Dr
Isodore Goldstein in 1935, and describes how every retina
contains a unique blood vessel pattern.
In a later paper, they even suggest using photographs of
these patterns as a means of identification.
The second study was conducted in the 1950s by Dr Paul
Tower. He discovered that - even among identical twins -
the blood vessel patterns of the retina are unique and
different

Background

The first company to become involved in the research,
development and manufacture of retinal scanning devices was
EyeDentify Inc.
 The company was established in 1976 and its first retina
capturing devices were known as ‘fundus cameras’. While
intended for use by ophthalmologists, modified versions of the
camera were used to obtain retina images. The device had
several shortcomings, however.
First, the equipment was considered very expensive and
difficult to operate.
Second, the light used to illuminate the retina was considered
too bright and too discomforting for the user.

Background

Further research and development yielded the first true
prototype scanning device, which was unveiled in 1981.
The device used infrared light to illuminate the blood vessel
pattern of the retina. The retina is essentially transparent to this
wavelength of light.
The advantage of infrared light is that the blood vessel pattern
in the retina can ‘absorb’ such light much faster than other
parts of the eye tissue. The reflected light is subsequently
captured by the scanning device for processing. In addition to a
scanner, several algorithms were developed for the
extraction of unique features.
Further research and development gave birth to the first true
retinal scanning device to reach the market: the
EyeDentification System 7.5.

Background

The three basic fuctions of EyeDentification System 7.5 are
1. Enrollment - where a person's reference eye signature is
built and a PIN number and text (such as the person's name) is
associated with it.
2. Verification - a person previously enrolled claims an identity
by entering a PIN number. The System scans the subject’s eye
and compares it with the reference eye signature associated
with the entered PIN. If a match occurs, access is allowed.
3. Recognition – The system scans the subject’s eye and
“looks-up” the correct, if any, reference eye signature. If a
match occurs, access is allowed.

Background

 The last retinal scanner to be manufactured by EyeDentify
was ICAM 2001, a device capable of storing up to 3000
templates and 3,300 transactions. The product was eventually
withdrawn from the market on account of its price as well as
user concerns.
Only a single company is currently in the process of creating a
retinal scanning device: Retinal Technologies, LLC.
It is believed that the company is working on a prototype
device that will be much easier to implement in commercial
applications. It will also be much more user friendly.

Retinal recognition

Retinal recognition system [Icam 2001 by Eyedentify]

Technology

The overall retinal scanning process may be broken down into
three sub-processes:
1. Image/signal acquisition and processing :-this sub-process
involves capturing an image of the retina and converting it to a
digital format.
2. Matching:- a computer system is used to verify and identify
the user (as is the case with the other biometric technologies ).
3. Representation:- the unique features of the retina are
presented as a template.
The process for enrolling and verifying/identifying a retinal
scan is the same as the process applied to other biometric
technologies (acquisition and processing of images; unique
feature extraction; template creation).

System operation

The image acquisition and processing phase is the most
complicated. The speed and easy with which this sub-process
may be completed largely depends on user cooperation.
To obtain a scan, the user must position his/her eye very close
to the lens. To safeguard the quality of the captured image, the
user must also remain perfectly still at this point.
Moreover, glasses must be removed to avoid signal
interference (after all, lenses are designed to reflect).
On looking into the scanner, the user sees a green light against
a white background. Once the scanner is activated, the green
light moves in a complete circle (360 degrees). The blood
vessel pattern of the retina is captured during this process.

System operation
 Generally speaking, three to five images are captured at this stage.
Depending on the level of user cooperation, the capturing phase can
take as long as one minute. This is a very long time compared to
other biometric techniques.
 The next stage involves data extraction. One very considerable
advantage of retinal recognition becomes evident at this stage. As
genetic factors do not dictate the pattern of the blood vessels, the
retina contains a diversity of unique features. This allows up to 400
unique data points to be obtained from the retina. For other
biometrics, such as fingerprints, only 30-40 data points (the
minutiae) are available.
 During the third and final stage of the process, the unique retina
pattern is converted to an enrolment template. At only 96 bytes, the
retina template is considered one of the smallest biometric
templates.

Causes of problems (errors)

As is the case with other biometric technologies, the
performance of the retinal scanning device may be affected
by a number of variables, which could prevent an accurate
scan from being captured.
Poor quality scans may be attributable to:
1. Lack of cooperation on the part of the user -the user must
remain very still throughout the entire process, especially
when the image is being acquired. Any movement can
seriously affect lens alignment.

Causes of problems (errors)

2. The distance between the eye and the lens is incorrect and/or
fluctuates - for a high quality scan to be captured, the user must
place his or her eye in very close proximity to the lens. In this
sense, iris scanning technology is much more user friendly; a
quality scan can be captured at a distance of up to three feet from
the lens.
3. A dirty lens on the retinal scanning device. This will obviously
interfere with the scanning process.
4. Other types of light interference from an external source.
5. The size of the user’s pupil. A small pupil reduces the amount
of light that travels to (and from) the retina. This problem is
exacerbated if the pupil constricts as a result of bright lighting
conditions, which can result in a higher False Reject rate.

Biometric performance standards
 All biometric technologies are rated against a set of performance
standards. As far as retinal recognition is concerned, there are two
performance standards: the False Reject Rate, and the Ability To Verify
Rate. Both are described below.
 False Reject Rate (also known as Type 1 Errors)
 Describes the probability of a legitimate user being denied
authorization by the retinal scanning system.
 Retinal recognition is most affected by the False Reject Rate. This is
because the factors described above have a tangible impact on the
quality of the retinal scan, causing a legitimate user to be rejected.
 Ability to Verify Rate
 Describes the probability of an entire user group being verified on a
given day. For retinal recognition, the relevant percentage has been as
low as 85%. This is primarily attributable to user-related concerns and
the need to place one’s eye in very close proximity to the scanner lens.

The strengths and weaknesses of retinal
recognition

 Just like all other biometric technologies, retinal recognition has its
own unique strengths and weaknesses. The strengths may be
summed up as follows:
 1. The blood vessel pattern of the retina rarely changes during a
person’s life (unless he or she is afflicted by an eye disease such as
glaucoma, cataracts, etc).
 2. The size of the actual template is only 96 bytes, which is very
small by any standards. In turn, verification and identification
processing times are much shorter than they are for larger files.
 3. The rich, unique structure of the blood vessel pattern of the retina
allows up to 400 data points to be created.
 4. As the retina is located inside the eye, it is not exposed to
(threats posed by) the external environment. For other biometrics,
such as fingerprints, hand geometry, etc., the opposite holds true.

Weaknesses

The most relevant weaknesses of retinal recognition are:
1. The public perceives retinal scanning to be a health threat;
some people believe that a retinal scan damages the eye.
2. User unease about the need to position the eye in such close
proximity of the scanner lens.
3. User motivation: of all biometric technologies, successful
retinal scanning demands the highest level of user motivation
and patience.
4. Retinal scanning technology cannot accommodate people
wearing glasses (which must be removed prior to scanning).
5. At this stage, retinal scanning devices are very expensive to
procure and implement.

Retinal recognition applications

 As retinal recognition systems are user invasive as well as expensive to
install and maintain, retinal recognition has not been as widely
deployed as other biometric technologies .
 To date, retinal recognition has primarily been used in combination
with access control systems at high security facilities. This includes
military installations, nuclear facilities, and laboratories.
 Retinal recognition was first introduced in Granite City and East Alton
(southern Illinois) towards mid-1996. Once the fingerprint recognition
program had been initiated, the authorities drew up a comparison
between fingerprint and retinal recognition, concluding that “retinal
scanning is not client or staff friendly and requires considerable time to
secure biometric records”.
 Based on these factors, retinal scanning technology is not yet ready for
state-wide adaptation to the Illinois welfare department…”. As a result,
the use of retinal recognition systems was stopped.

Limitations
 Perceived Health Threat- While the low light level is harmless to
the eye, there is a widely held perception that RR can hurt the
retina.
 Outdoors vs. Indoors- A small pupil reduces the amount of light that
travels to (and from) the retina. Further, outdoor environments are
less conducive to reliable RR performance than indoor
environments because of ambient light conditions.
 Ergonomics- The need to bring the RI device to an eye or the eye to
the device makes the RR more difficult to use in some applications
than other biometric identification technologies. For instance, it is
quite easy for a subject, regardless of his height to reach a hand to a
fingerprint or hand geometry.
 Severe Astigmatism- Because eyeglasses must be removed in order
to use RR systems reliably, people with severe astigmatism may
have trouble aligning the dots in the camera's align/fixate target.
 High Sensor Cost- The camera requirement of RR puts a lower limit
on the cost of the system.

Conclusion

In view of the rich and unique blood vessel patterns in the
retina, there is no doubt that retinal recognition is the
‘ultimate’ biometric. Its high cost and user-related
drawbacks have prevented it from making a commercial
impact. However, as technology continues to advance, it
seems likely that retinal recognition will one day be
widely accepted and used.

References

1 “Retina Identification”, Robert “Buzz” Hill. Article is
from the book: “Biometrics: Personal Identification in
Networked Society”, by Anil Jain, Ruud Bolle, and
Sharath Pankati
2 “Retinal Identification System: Performance Analysis”
Scientific whitepaper provided by Retinal Technologies,
LLC
3Ravi Das. “Retinal Recognition:Biometric technology in
practice” Keesing Journal of Documents & Identity,
issue 22, 2007
4 www.dss.state.ct.us/digital/illions.htm

Retinal Recognition

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Retinal Recognition (20)

More from Piyush Mittal (20)

Recently uploaded (20)

Retinal Recognition

Editor's Notes