International Journal of Engineering and Technical Research (IJETR)

ISSN: 2321-0869, Volume-2, Issue-9, September 2014

Personal Authentication Based on IRIS Recognition

Mayank M Badgandi, K Srinivas Rao
recognition where Statistical Estimation Theory can play a
Abstract (Iris recognition is the process of recognizing a big role.
person by analyzing the apparent pattern of his or her
iris. There is a strong scientific demand for the proliferation of Security and the authentication of individuals is
systems, concepts and algorithms for iris recognition and
necessary for many di_erent areas of our lives, with most
identification. This is mostly because of the comparatively short
time that iris recognition systems have been around. In
people having to authenticate their identity on a daily basis;
comparison to face, fingerprint and other biometric traits there examples include ATMs, secure access to buildings, and
is still a great need for substantial mathematical and international travel. Biometric identi_cation provides a valid
computer-vision research and insight into iris recognition. One alternative to traditional authentication mechanisms such as
evidence for this is the total lack of publicly available adequate ID cards and passwords, whilst overcoming many of the
datasets of iris images. The program converts a photo of an eye shortfalls of these methods; it is possible to identify an
to an 'unrolled' depiction of the subject's iris and matches the individual based on \who they are" rather than \what they
eye to the agent's memory. If a match is found, it outputs a best possess" or \what they remember". Iris recognition is a
match. The current functionality matches that proposed in the
particular type of biometric system that can be used to reliably
original requirements.
indentify a person by analysing the patterns found in the iris.
The iris is so reliable as a form of identi_cation because of the
Index Terms Human Iris Authentication, Statistical uniqueness of its pattern. Although there is a genetic inuence,
Correlation Coefficient, False Acceptance Rate, False Rejection particularly on the iris' colour, the iris develops through
Rate, Biometrics, image quality,Machine learning, iris folding of the tissue membrane and then degeneration (to
recognition. create the pupil opening) which results in a random and
unique iris. In comparison to other visual recognition
I. INTRODUCTION techniques, the iris has a great advantage in that there is huge
The human iris is so unique that no two irises are alike, even variability of the pattern between individuals, meaning that
among identical twins, in the entire human population. large databases can be searched without _nding any false
Recently, human iris biometric based identification has matches.
attracted the attention of research and development
community. Iris recognition has many advantages over the
other forms of biometric identification. Iris recognition is the II. IRIS RECOGNITION TECHNOLOGIES
most accurate form of identification known to man. It is also
capable of making a match from a database of over 1 million A. The Iris
records in less than a second. The iris image remains stable The iris has many features that can be used to distinguish
from the age of about 10 months up until death. Noninvasive, one iris from another. One of the primary visible
users wearing gloves, protective wear, glasses, safety goggles characteristic is the trabecular meshwork, a tissue which gives
and even contact lenses can operate iris recognition systems. the appearance of dividing the iris in a radial fashion4 that is
This means that not only is iris recognition the most accurate permanently formed by the eighth month of gestation. During
biometric technology, it is also the safest. the development of the iris, there is no genetic influence on it,
a process known as chaotic morphogenesis that occurs
All images taken for this work are from MMU1 Iris during the seventh month of gestation, which means that even
database. MMU1 iris database contributes a total number of identical twins have differing irises. The iris has in excess of
450 iris images which were taken using LG IrisAccess2200. 266 degrees of freedom i.e. the number of variations in the
This camera is semi automated and it operates at the range of iris that allow one iris to be distinguished from another. The
725 cm. They come from Asia, Middle East, Africa and fact that the iris is protected behind the eyelid, cornea and
Europe. Each of them contributes 5 iris images for each eye. aqueous humour means that, unlike other biometrics such as
These images are 100 100 24bit Bitmap (.bmp), each fingerprints, the likelihood of damage and/or abrasion is
occupying 32,768 bytes on a hard drive. After generating Iris minimal. The iris is also not subject to the effects of aging
pattern images take 10 different 12 8 iris pattern images of which means it remains in a stable form from about the age of
each authorized individuals of an organization to make a one until death. The use of glasses or contact lenses (coloured
database against his/her identification number. Now, By or clear) has little effect on the representation of the iris and
considering Biological characteristics of IRIS Pattern we use hence does not interfere with the recognition technology.
Statistical Correlation Coefficient for this IRIS Pattern B. Iris Recognition Process
Manuscript received September 20, 2014. The process of capturing an iris into a biometric template is
Mayank M Badgandi, Student Mtech, Dept.of CS&E, MRCET made up of 3 steps:
(JNTU) Hyderabad (AP), India 1. Capturing the image
K Srinivas Rao, Assistant Professor Dept.of CS&E, MRCET (JNTU) 2. Defining the location of the iris and optimising the image
Hyderabad (AP), India
.
3. Storing and comparing the image.

238 www.erpublication.org
 Personal Authentication Based on IRIS Recognition

a) Capturing the Image c) Storing and Comparing the Image

The image of the iris can be captured using a standard Once the image has been captured, an algorithm uses 2-D
camera using both visible and infrared light and may be either Gabor wavelets to filter and map segments of the iris into
a manual or automated procedure. The camera can be hundreds of vectors (known here as phasors)9. The 2-D
positioned between three and a half inches and one meter to Gabor phasor is Simply the what and where of the image.
capture the image. In the manual procedure, the user needs to Even after applying the algorithms to the iris image there are
adjust the camera to get the iris in focus and needs to be within still 173 degrees of freedom to identify the iris. These
six to twelve inches of the camera. This process is much more algorithms also take into account the changes that can occur
manually intensive and requires proper user training to be with an iris, for example the pupils expansion and
successful. The automatic procedure uses a set of cameras that contraction in response to light will stretch and skew the iris.
locate the face and iris automatically thus making this process This information is used to produce what is known as the
much more user friendly. IrisCode, which is a 512-byte record. This record is then
stored in a database for future comparison. When a
b) Defining the Location of the Iris and Optimising comparison is required the same process is followed but
the Image instead of storing the record it is compared to all the
Once the camera has located the eye, the iris recognition IrisCode records stored in the database. The comparison
system then identifies the image that has the best focus and also doesnt actually compare the image of the iris but rather
clarity of the iris. The image is then analysed to identify the compares the hexadecimal value produced after the
outer boundary of the iris where it meets the white sclera of algorithms have been applied.
the eye, the pupillary boundary and the centre of the pupil.
This results in the precise location of the circular The following table shows the probabilities of false accept
iris. and false reject with iris recognition technology:
Table 1 Hamming Distances & Error Probabilities10
Hamming False accept False reject
distance probability probability
.28 1 in 10E3D 1 in 11,400
.29 1 in 10 1 in 22,700
.30 1 in 6.2 billion 1 in 46,000
.31 1 in 665 million 1 in 95,000
.32 1 in 81million 1 in 201,000
.33 1 in 11 million 1 in 433,000
.34 1 in 1.7 million 1 in 950,000
.342 1 in 1.2 million 1 in 1.2 million

Fig 1: Circular Iris Location .35 295,000 1 in 2.12million

.36 57,000 1 in 4.84 million
The iris recognition system then identifies the areas of the
iris image that are suitable for feature extraction and analysis. .37 12,3000 1 in 11.3 million
This involves removing areas that are covered by the eyelids,
any deep shadows and reflective areas. The following
diagram shows the optimisation of the image. C. Iris process flow diagram
8 Stages of Iris Detection:
Stage 1: Scan Eye: The eye scanning will be simulated in this
system as we have no method of taking real-time images of
subjects. Therefore, all eye images are to be jpeg image files
at least 1000 x 1000 pixels in dimension. The eye is scanned
by manually selecting the file and instructing the agent to scan
it. The agent begins scanning an eye by turning the jpeg file
into an image object in full color (24 bit RGB).
Stage 2: Eye Grayscaling Algorithm: The agent next
converts the full-color image to an 8-bit representation. This
reduces space complexity, making further computations faster
without losing reliability.
Stage 3: Median Filter: Applies a median filter to the
grayscaled image to reduce the amount ofnoise and artifacts
before the pupil center detection.
Stage 4: Pupil Center Detection: To perform the fifth step,
Fig 2: Optimising the Image the agent must now find the center of the pupil to orient the
coordinate system at the center of the eye.

239 www.erpublication.org
 International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-2, Issue-9, September 2014
Stage 5: Canny Edge Detection: Step 5 also requires that Subtracts the mean filtered image from the
the edges of the iris and pupil be marked so an edge detection unwrapped image of the eye being matched to.
process needs to be completed by the agent. We decided to
have the agent use the 'Canny Edge Detection algorithm. Computes a percentage difference between the eye
being matched and the identity.

Selects the identity that has the highest percentage

match and displays the result to the user in a dialog
box.
To design a iris recognition system based on an empirical
analysis of the iris image and it is split in several steps using
local image properties.

III. PROPOSED SYSTEM

Biometric-based personal verification and identification
methods have gained much interest with an increasing
emphasis on security.Iris recognition is a fast, accurate and
secure biometric technique that can operate in both
verification and identification modes.The iris texture pattern
has no links with the genetic structure of an individual and
since it is generated by chaotic processes .Externally visible;
patterns imaged from a distance Iris patterns possess a high
degree of randomness.
ADVANTAGES: In proposed system we are using three new
methods to improve performance of a single biometric
matcher.
Uniqueness: set by combinatorial complexity.
Variability: 244 degrees-of-freedom.
Entropy: 3.2 bits per square-millimeter.
Highly protected, internal organ of the eye.

Buy SmartDraw !- purchased copies print this

document without a watermark .
Visit www.smartdraw.com or call 1-800-768-3729.
IV. EXPERIMENTAL RESULTS
In this section, we analyze the performance of the proposed
iris verification algorithm. So, to evaluate the performance of
Stage 6: Pupil And Iris Radius Detection Algorithm: In our algorithm, different experiments on the Institute of
this stage the agent attempts to find the radius of both the Automation, Chinese Academy of Science, CASIA databases
pupil and iris using the center found in stage 3 and the edges have been carried out. As mentioned in , automatic detection
found in stage 4. The radii will be used by the agent in both of the pupils is easier because CASIA database images have a
step 6 and 7. black circlein the pupil and some of these black circles cover
the part of iris and also the specular reflection. First, we report
Stage 7: Iris Localization: This process is mostly for visual
the experimental results obtained running our algorithm using
purposes and refers to removing erroneous information from
several wavelets categories. Thereafter, we present some
the original image outside of the iris radius, whereby leaving
result to study the effect of the wavelet packet combination.
only the image within the bounds of the iris radius intact.
Finally, we compare our performance with the algorithms
Stage 8: Iris Unrolling Algorithm: The agent must now use reported in which use the fixed threshold. For the first
the radii found in stage 5 as well as the pupil center found in experiment, we have uses the first wavelet packet feature that
stage 3 to perform the unrolling of the iris. Traversing the eye has the maximal energy value to generate the iris signature (to
in a polar coordinate, circular fashion, the agent maps encode 64 bits of the wavelet packets). This experiment was
information from the original image to the output motivated by an iris code computation using only two
(unwrapped) image. appropriate energies according to the equation to compute the
adapted threshold and one iris code according to the maximal
Stage 9: Iris Matching: Scans the eye to be matched using energy. The energy corresponding to sub image 0 has not
the previous 8 steps. been used because this sub image contains offset information.

Iterates over each of the identies in the memory and

performs the following steps. V. CONCLUSION AND FUTURE SCOPE
This paper presents a literature review on the various
Performs a mean filer on the unrolled image stored iris recognition techniques. This paper emphasizes on
in memory. biometric identification scheme that utilizes iris image as
verification source because this organs characteristics do not

240 www.erpublication.org
 Personal Authentication Based on IRIS Recognition

vary so much with ageing effect. Most of the time for iris
localization and segmentation wavelets are effective solution
and for coding purpose Gabor filters are used. Canny edge
detectors performance is much better compare to other edge
detection techniques. Finally, for template matching hamming
distance method perform well.
The primary focus of this work is a personal
authentication system based on human iris verification using
wavelet packets decomposition. The proposed technique uses
only appropriate packets with dominant energies to encode
iris texture according the adapted thresholds. The usefulness
of this approach was confirmed in the experiments conducted
here, which reveals that the verification results with an
EER=0.3% has been obtained for packets combination, which
means that our system is appropriate for very high security
environments.

REFERENCES
[1] Ales Muron and Jaroslav Pospisil, The human iris structure and its
usages, ActaUniv. Palacki. Olomuc. Fac.Rerum Nat. Phys., vol. 39,
2000, pp. 8795.
[2 C. Sanchez-Avilaa, R. Sanchez-Reillob, Two different approaches for
iris recognition using Gabor filters and multiscale zero-crossing
representation, Pattern Recognition, vol. 38, 2005, pp. 231 240.
[3] H. Proenc and L. A. Alexandre, Toward Noncooperative Iris
Recognition: A Classification Approach Using Multiple Signatures,
IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 29, no.
4, Nov 2007, pp
[4] J.G. Daugman, Demodulation by Complex-Valued Wavelets for
Stochastic Pattern Recognition, Intl J.Wavelets, Multiresolution and
Information Processing, vol. 1, no. 1, 2003, pp. 1-17.
[5] Kwanghyuk Bae, Seungin Noh, and Jaihei Kim, Iris feature extraction
using independent component analysis, Proc. 4th Intl. Conf. on Audio
and Video Based Biometric Person Authentication LNCS, vol. 2688,
2003, pp.
[6] R. Wildes, J. Asmuth, G. Green, S. Hsu, R. Kolczynski,J. Matey, and S.
McBride, A Machine-Vision System for Iris Recognition, Machine
Vision and Applications, vol. 9, 1996, pp. 1-8.

241 www.erpublication.org