Upload
sabina-saldanha
View
95
Download
3
Embed Size (px)
Citation preview
AIRPORT SECURITY SYSTEM USING IRIS RECOGNITION
PAGE INDEX
TOPIC PAGE NO.
1.Introduction …………………………………… …………. 2
2.Literature survey…………………………………………. 4
3.Project definition………………………………. ………….6
4. Block diagram………………………………………... ..… 8
5. Algorithm…………………………………………..9
6. Flowchart…………………………………………..10
7. Implementation steps…………………………......11
1
INTRODUCTION
In today’s information age it is not difficult to collect data about an individual and
use that information to exercise control over the individual. Individuals generally
do not want others to have personal information about them unless they decide to
reveal it. In this context, data security has become an inevitable feature.
Conventional methods of identification based on possession of ID cards or
exclusive knowledge like Social security number or a password are not altogether
reliable. ID cards can be almost lost, forged or misplaced: passwords can be
forgotten. Such that an unauthorized user may be able to break into an account
with little effort. So it is need to ensure denial of access to classified data by
unauthorized persons. Biometric technology has now become a viable alternative
to traditional identification systems because of its tremendous accuracy and speed.
Biometric system automatically verifies or recognizes the identity of a living
person based on physiological or behavioral characteristics. Since the persons to be
identified should be physically present at the point of identification, biometric
techniques gives high security for the sensitive information stored in mainframes
or to avoid fraudulent use of ATMs.
Biometric products are used for automated recognition of individuals based
on their behavioral and biological characteristics. Iris recognition biometric
products recognize individuals based on their iris images more specifically the
distinctive patterns in the irises created by various structures, such as crypts,
furrows, frills, ridges, ligaments, freckles, coronas, and collarettes. Other common
biometric products use fingerprint features, facial images, hand geometry,
2
characteristics of handwritten signatures, and voice recordings to recognize
individuals.
This project basically aims at designing an iris matching software system.
Firstly, image preprocessing is performed followed by extracting the iris portion of
the eye image which is called Localization. The extracted iris part is then
normalized using daugman’s rubbersheet model, and Iris Code is constructed.
Finally two Iris Codes are compared to find Hamming Distance, which is fractional
measure of the dissimilarity. Experimental image results show that unique codes
can be generated for every eye image and Hamming Distance between any two
different iris code has maximum value.
With the increasing demand of enhanced security in our daily lives, reliable
personal identification through biometrics is currently an active topic in the
literature of pattern recognition. Nowadays many automatic security systems based
on iris recognition have been deployed worldwide for border control, restricted
access Iris recognition is based on the most mathematically unique biometric - the
iris of the eye. The human iris is absolutely unique, even between twins or an
individual's right and left eyes. The iris itself is stable throughout a person's life
(approximately from the age of one). The physical characteristics of the iris do not
change with age. One key tool in this area is the use of biometrics. Humans have
always identified each other by recognizing faces, voices or some other physical
characteristic. Personal recognition or identification by a witness is also entrenched
in our law and commercial structures. Now the use of biometric technologies is
providing a means to positively identify or authenticate large numbers of people
without having to primarily rely on human to human identification.
3
LITERATURE SURVEY
J. Daugman. “How iris recognition works”. IEEE trans.Proceedings of 2002
International Conference on Image Processing, Vol. 1, 2002.
The Hough transform is a standard computer vision algorithm that can be
used to determine the parameters of simple geometric objects, such as lines and
circles, present in an image. Also there is Daugman’s integro-differential operator
for locating the circular iris and pupil regions, and the arcs of the upper and lower
eyelids. The integro-differential can be seen as a variation of the Hough transform,
since it too makes use of first derivatives of the image and performs a search to
find geometric parameters. Since it works with raw derivative information, it does
not suffer from the thresholding problems of the Hough transform. However, the
algorithm can fail where there is noise in the eye image, such as from reflections,
since it works only on a local scale. Hence we go ahead with Hough transform for
implementing Localization. The homogenous rubber sheet model devised by
Daugman remaps each point within the iris region to a pair of polar coordinates
(r,θ) where r is on the interval [0,1] and θ is angle [0,2π] .Image registration is
another technique, which geometrically warps a newly acquired image into
alignment with a selected database image . Virtual circles is same as Daugman’s
rubber sheet model, however scaling is at match time, and is relative to the
comparing iris region, rather than scaling to some constant dimensions. Also, it is
not mentioned by Boles, how rotational invariance is obtained. The extracted iris
portion can be brought to the standard format for comparison by Daugman’s
rubber sheet model in our implementation. Wavelet encoding can be used to
decompose the data in the iris region into components that appear at different
4
resolutions. The output of applying the wavelets is then encoded in order to
provide a compact and discriminating representation of the iris pattern
Daugman, J. (1993) "High confidence visual recognition of persons by a test of
statistical independence." IEEE Transactions on Pattern Analysis and Machine
Intelligence, vol. 15(11), pp. 1148-1161.
. Gabor filter is constructed by modulating a sine/cosine wave with a Gaussian
which provides localization in space, though with loss of localization in frequency.
A disadvantage of the Gabor filter is that the even symmetric filter will have a DC
component whenever the bandwidth is larger than one octave. However, zero DC
components can be obtained for any bandwidth by using a Gabor filter which is
Gaussian on a logarithmic scale, this is known as the Log-Gabor filter. Gabor filter
is used in our implementation.
Daugman J (2006) "Probing the uniqueness and randomness of IrisCodes: Results
from 200 billion iris pair comparisons." Proceedings of the IEEE, 94(11), pp 1927-
1935.
The Hamming distance gives a measure of how many bits are the same between
two bit patterns. Using the Hamming distance of two bit patterns, a decision can be
made as to whether the two patterns were generated from different irises or from
the same one. The weighted Euclidean distance (WED) can be used to compare
two templates, especially if the template is composed of integer values. The
weighting Euclidean distance gives a measure of how similar a collection of values
are between two templates. We will use hamming distance.
5
PROJECT DEFINITION
3.1 Aim
Design an airport security system, which uses Iris Recognition Technique using
Mat-Lab software
3.2 Objective
Security system is designed such that it can be used in public or private places for
visitor identification at Airports and offices etc. In order to overcome the security
problems biometric system such as Iris Recognition system is most accurate,
reliable and efficient way to recognize and distinguish the people. Its main aim is
to nab the terrorist/criminals and identify important personalities during security
checking at the airport, whose iris code is present in the database.
3.3 Scope
Future development of the software is quite possible, and in fact it is very helpful
in improving its efficiency and flexibility in use.
Matlab is a powerful tool when it comes to mathematical operations but it
uses lots of vectors and matrix and these vectors and matrices uses too much
memory, hard disk and slow down the processing unit of the computer. Thus
implementing our project by C#.net or Visual Basic.net to save memory and time.
Also program will work independent from the Matlab.
6
3.5 How it is different?
Fig 3.1 Iris accuracy graph
Accuracy is the parameter which differentiates Iris recognition technology from
rest of the Biometric techniques such as Face, Voice, Fingerprint, Retina,
Signature etc. Iris recognition is more accurate, stable and scalable. Hence iris
recognition is popularly used in many applications.
3.6 Relevance
It is relevantly used at the Airports as a part of security check routine in many
western countries like the UAE, London-Heathrow Airport. It is used during
immigration and at the borders where instead of identity cards Iris is used to
ascertain the identity of the person. It can replace passwords at banks, ATM,
Military.
7
BLOCK DIAGRAM
4.1 Block diagram
Fig 4.1 Block diagram of iris recognition
8
ALGORITHM
1. Start.
2. Capture the eye image.
3. Extract the iris portion by Localization.
4. Get the extracted portion to standard form by Normalization.
5. Generate iris code.
6. Compare the generated code with the code in database.
7. Display the result.
8. Stop.
9
FLOW CHART
Fig 4.4 Flowchart
10
IMPLEMENTATION STEPS
4.5.1 Acquisition
Fig 4.5.1 Image capture setup
The acquisition stage, captures the iris image in near infrared light ranging from
700-900nm. The typical distance from camera to user is about one meter, with
user cooperation (i.e. fixed position with user looking into camera).
The primary goal of this project is to design a fully automated image quality
block that is capable of discriminating between good and poor quality images.
Moreover, the quality metric should be able to predict performance. Finally, this
project should provide insight on which factors negatively impact performance
when using traditional iris recognition systems. So for improving the performance
iris recognition System, we use infrared light images.
11
There are 2 important requisites for this process
i) It is desirable to acquire images of the iris with sufficient resolution and
Sharpness to support recognition
ii) It is important to have good contrast in the interior iris pattern without
restoring to a level of illumination that annoys the Operator, that is adequate
intensity of source constrained by operators comfort with brightness..
Requirements for Iris camera:
For more recognition rate, we will require infrared light illuminated database. In
CASIA databases, all images are captured from proper distance and captured in
light illumination. The general purpose is a high confidence and real time
recognition of an individual’s identity by mathematical analysis of the random
patterns that are scanned from the iris of an eye from some distance. The
procedure of filming human irises must meet some general requirements to be
applicable in real scenarios:
The process of measurement should be fast, comfortable for the measured
person as well as robust against natural modifications of the eyes. Hence, used
reflections, scaling, deformation (e.g. pupil dialation) and possible camera and
light differences like shading or noise.
4.5.2 Localization
Localization represents the process of segmenting the pupil, sclera, and eyelid
regions. Pupil and iris detection/segmentation in a traditional system is carried out
by using an Hough transform that acts as a circular edge detector. Linear hough
transform is an algorithm used to detect the upper and lower eyelids. Also we
make use of Thresholding to detect the eyelashes. These algorithms are employed
in order to extract the iris portion.
i) Hough Transform
12
The Hough transform is a standard computer vision algorithm that can be used to
determine the parameters of simple geometric objects, such as lines and circles,
present in an image. The circular Hough transform can be employed to deduce the
radius and centre coordinates of the pupil and iris regions
It is decided to use circular Hough transform for detecting the iris and pupil
boundaries. This involves first employing canny edge detection to generate an
edge map. Gradients were biased in the vertical direction for the outer iris/sclera
boundary. Vertical and horizontal gradients were weighted equally for the inner
iris/pupil boundary. A modified version of Kovesi’s canny edge detection
MATLAB® function can be implemented, which allowed for weighting of the
gradients.
Eyelids were isolated by first fitting a line to the upper and lower eyelid
using the linear Hough transform. A second horizontal line is then drawn, which
intersects with the first line at the iris edge that is closest to the pupil. This
process and is done for both the top and bottom eyelids. The second horizontal
line allows maximum isolation of eyelid regions. Canny edge detection is used to
create an edge map, and only horizontal gradient information is taken. The linear
Hough transform is implemented using the MATLAB® Radon transform, which is
a form of the Hough transform. If the maximum in Hough space is lower than a
set threshold, then no line is fitted, since this corresponds to non-occluding
eyelids. Also, the lines are restricted to lie exterior to the pupil region, and interior
to the iris region.
For isolating eyelashes in the CASIA database a simple thresholding
technique was used, since analysis reveals that eyelashes are quite dark when
compared with the rest of the eye image.
13
4.5.3 Normalization
Normalization is carried out to represent the segmented iris region with regard to
invariance of size, position and orientation. Daugman’s transforms the coordinate
system from Cartesian coordinates to a doubly dimensionless nonconcentric polar
coordinate system. From Fig .we can illustrates the normalization process. Θ
represents the angle (between 0 and 360) and r represents radial resolution
(between 0 and 1).
Fig 4.5.3 Daugman’s Rubber Sheet Model
For normalization of iris regions a technique based on Daugman’s rubber
sheet model was employed. The centre of the pupil was considered as the
reference point, and radial vectors pass through the iris region. A number of data
points are selected along each radial line and this is defined as the radial
resolution. The number of radial lines going around the iris region is defined as
the angular resolution.
4.5.4 Feature Encoding
A Gabor filter is constructed by modulating a sine/cosine wave with a Gaussian.
Modulation of the sine with a Gaussian provides localization in space, though
14
with loss of localization in frequency. Decomposition of a signal is accomplished
using a quadrature pair of Gabor filters, with a real part specified by a cosine
modulated by a Gaussian, and an imaginary part specified by a sine modulated by
a Gaussian. The real and imaginary filters are also known as the even symmetric
and odd symmetric components respectively.
Feature encoding is implemented by convolving the normalized iris pattern
with 1D Gabor wavelets. The 2D normalized pattern is broken up into a number
of 1D signals, and then these 1D signals are convolved with 1D Gabor wavelets.
The rows of the 2D normalized pattern are taken as the 1D signal, each row
corresponds to a circular ring on the iris region. The angular direction is taken
rather than the radial one, which corresponds to columns of the normalized
pattern, since maximum independence occurs in the angular direction.
The output of filtering is then phase quantized to four levels using the
Daugman method with each filter producing two bits of data for each phasor. The
output of phase quantization is chosen to be a grey code, so that when going from
one quadrant to another, only 1 bit changes. This will minimize the number of
bits disagreeing, if say two intra-class patterns are slightly misaligned and thus
will provide more accurate recognition.
The encoding process produces a bitwise template containing a number of
bits of information, and a corresponding noise mask which corresponds to corrupt
areas within the iris pattern, and marks bits in the template as corrupt.
4.5.5 Matching
For matching, the Hamming distance was chosen as a metric for recognition,
since bit-wise comparisons were necessary. The Hamming distance algorithm
employed also incorporates noise masking, so that only significant bits are used
15
in calculating the Hamming distance between two iris templates. Now when
taking the Hamming distance, only those bits in the iris pattern that corresponds
to ‘0’ bits in noise masks of both iris patterns will be used in the calculation. The
Hamming distance will be calculated using only the bits generated from the true
iris region, and this modified Hamming distance formula is given as
where Xj and Yj are the two bit-wise templates to compare, Xnj and Ynj are
the corresponding noise masks for Xj and Yj, and N is the number of bits
represented by each template.
Although, in theory, two iris templates generated from the same iris will
have a Hamming distance of 0.0, in practice this will not occur. Normalization is
not perfect, and also there will be some noise that goes undetected, so some
variation will be present when comparing two intra-class iris templates.
In order to account for rotational inconsistencies, when the Hamming
distance of two templates is calculated, one template is shifted left and right bit-
wise and a number of Hamming distance values are calculated from successive
shift. This method is suggested by Daugman and corrects for misalignments in
the normalized iris pattern caused by rotational differences during imaging. From
the calculated Hamming distance values, only the lowest is taken, since this
corresponds to the best match between two templates.
The number of bits moved during each shift is given by two times the
number of filters used, since each filter will generate two bits of information from
one pixel of the normalized region. The actual number of shifts required to
normalize rotational inconsistencies will be determined by the maximum angle
16
difference between two images of the same eye, and one shift is defined as one
shift to the left, followed by one shift to the right.
4.5.6 Decision
On the basis of similarity of two different persons hamming distances, we can
take decision that enrolled user is genuine person or imposter and in other words,
we can give access to enrolled user or reject its authentication.
4.6 Platform
i) Digital Image Processing.
4.7 Language
Matlab
17