Sie sind auf Seite 1von 21

IRIS DETECTION

GROUP MEMBERS

AMIT PRASAD

(12100112058)

DANISH AHMAD ANSARI

(12100112069)

RAHUL MUKHERJEE

(12100112093)

RAHUL KUMAR OJHA

(12100112091)

RIYA GUPTA

(12100112094)

ABSTRACT

Iris Recognition is one of the important biometric


recognition systems that identify people based on their
eyes and iris.

Iris based recognition is one of the most proven technique


because it is stable, unique, flexible, reliable and even
two identical twins also has different iris patterns.

The iris is the only internal organ readily visible from the
outside.

INTRODUCTION

Iris detection is the process of recognizing the iris pattern


by analysing the image of an eye.

This iris is the area of the eye where the pigmented or


coloured circle, usually brown or blue, rings the dark pupil
of the eye.

The amount of information that can be measured in a


single iris is much greater than any other biometrics such
as fingerprints, face, voice etc.

THE HUMAN IRIS

The iris is a thin circular diaphragm, which lies between the cornea and the
lens of the human eye.

The function of the iris is to control the amount of light entering through the
pupil.

METHOD TO SEGMENT IRIS


IMPLEMENTATION OF DAUGHMANS ALGORITHM

DAUGHMANS ALGORITHM

Named after the professor John Daugman, is an


integrodifferential operator that searches over an image
of the eye for the circular pupil and the limbic borders of
the iris.

It is a circular edge detector that searches for the


parameters of a circular border.

DAUGMANS OPERATOR

The task is to find the maximum intensity difference and


the radius of the iris and the pupil and Daugmans
equation is employed for this task.

The centre point of Daugmans theory of border


recognition is the integrodifferential equation as follows:

FLOW
CHART

RGB TO GRAY

COMPLEMENTING THE IMAGE

After complementing the image

FILLING OF HOLES

Filling the holes

After removing holes

THRESHOLDING

Thresholding is a method of selection of pixels in an


image.

The concept of thresholding below is used before


implementing Daugmans equation.

Thresholding below marks some pixels as object pixels


if those pixels have an intensity value below a certain
threshold value.

SELECT BY LOCAL MINIMA

The threshold image is further scanned, pixel by pixel to determine


whether the pixel is a local minimum in that particular pixels
immediate 3-by-3 neighbourhood.

This means that every pixel intensity is compared to the intensities of


the pixels in its immediate eight neighbourhood pixels.

The pixel with the lowest intensity value amongst these nine pixels is
used for further calculations. The rest of the pixels are discarded.

COLLECTION OF ALL POINTS

The threshold image is further scanned, pixel by pixel amongst the possible
centre.

Pixels from previous step to determine whether the pixel is a local minimum
in that particular pixels immediate 3-by-3 neighbourhood.

This means that every pixels intensity is compared to the minimum of


intensities of the pixels in its immediate eight neighbourhood pixels.

Only those pixels having value equal to local minima is considered further rest
are discarded.

APPLY DAUGMANS OPERATOR TO FIND


PUPIL CENTER AND RADIUS

During the iris detection process, for every assumed


center pixel the sum of values of circumferential pixels is
noted at different levels of radii.

Circular regions of pixels that do not completely lie inside


the eye image are neglected.

Amongst all such intensity difference one with maximum


difference is selected and that is pupil.

DO THE SIMILAR FOR IRIS BUT WITH SAME RADIUS &


IN RANGE

After getting the pupil radius and its centre co-ordinates


,iris border can also be found out by applying the
following the following concept of Daugman:
0.1*Iris Border Radius < Pupil Border Radius < 0.8*Iris Border Radius

SEGMENT THESE CIRCLES

After

Considering formula of a circle having centre at co-ordinate


(h,k) and having radius r which is as follows:

The following portion is kept and

for the rest of the portion, it will be assigned 255 (white).

iris detection, i.e., detecting the iris region as concentric


circles, we will consider only the points inside the detected iris
region

REFERENCES

J. Daugman. High confidence visual recognition of persons by a test of


statistical independence. IEEE Transactions on Pattern Analysis and Machine
Intelligence, Vol. 15, No. 11, 1993

]DaugmanJ."How iris recognition works." IEEE Trans. CSVT,vol. 14, no. 1. 2004

John Daugman.Recognizing persons by their iris patterns. Cambridge


University Cambridge, UK. 2001

Recognition of Human Iris Patterns for Biometric Identification by Libor


Masek

ACKNOWLEDGEMENT
We are really indebted to MR. SUDIPTA BHATTACHARYA,
Assistant Professor of CSE Department, Bengal Institute of
Technology whose role as a project guide was invaluable for
the project work. I am extremely thankful for the keen
interest he took in advising and suggesting us. I am also
thankful to Nilanjan Banerjee, Head of the Department,
Computer Science, Bengal Institute of Technology for making
facilities available and giving his opinion.

THANK YOU

Das könnte Ihnen auch gefallen