Submitted By-
Miss. Almas Kanjiyani

---

[attachment=12556]
ABSTRACT
A method based for rapid visual recognition of personal identity is described on the failure of statistical test of independence. The most unique feature visible in a person’s face is the detailed texture of each eye’s iris. Iris based identity recognition is one of the most important parts of biometrics.
The problem of the personal identification has become a great matter in today’s world. Biometrics, which means biological features based identity recognition, has provided a convenient and reliable solution to this problem. This recognition technology is relatively new with many significant advantages, such as speed, accuracy, hardware, simplicity and applicability.
An iris has a mesh like texture to it, with numerous overlays and patterns. Basically, iris recognition system comprises of four main modules: Image Acquisition, Preprocessing, Feature Extraction and Pattern Matching. Firstly an image containing the users eye is captured by the system. Then the image is preprocessed. Thirdly, features representing the iris patterns are extracted. Finally, decision is made by means of matching.
To show that, although iris recognition is still in its final research and development stage throughout the world, it has many possible applications, some of which are listed below:
1.Secure access to bank case machine accounts.
2.Ticketless air travel.
3. Driving license, and other personal certificates.
4.Internet security, control of access to privileged information
 INTRODUCTION
With communications among people constantly increasing nowadays, how to recognize people’s identity has become an essential problem. The traditional methods such as keys, certificates, passwords, etc, can hardly meet the requirements of identity recognition in the modern society. Biometrics, which means biological features based identity recognition, has provided a convenient and reliable solution to this problem.
Iris based identity recognition is one of the most important parts of biometrics due to its various advantages, such as preciseness and no need of direct contact with the testees. According to some comparative research, the error rate of iris recognition is the lowest one among all the biometrics approaches till date.
Many users are skeptical of the use of such a technology due to wearers of eyeglasses, contact lenses, or sunglasses. However, the recognition system is able to perform right through glasses or lenses since they do not interfere with the process. There is no need for a customer to take off their glasses in order to be identified quickly and accurately.
Iris recognition technology was designed to be less intrusive than retina scans, which often require infrared rays or bight light to get an accurate reading. Scientists also say a person’s retina can change with age, while an iris remains intact. And no two blueprints are mathematically alike, even between identical twins and triplets. During the course of examining large number of files, anatomists and ophthalmologists have noted that the detained pattern of an iris, even the left and the right iris of a single person, seem to be highly distinctive.
In addition recent medical advances such as refractive surgery, cataract surgery and cornea transplants do not change iris’ characteristics. In fact, it is impossible to modify the iris without risking blindness. And even a blind person can participate. As long as a sightless eye has an iris, that eye can be identified by iris recognition.
 WHAT IS AN IRIS ???
An iris has a mesh-like texture to it, with numerous overlays and patterns. The iris is located behind the cornea of the eye, but in front of the lens. Its only physiological purpose is to control the amount of light that enters the eye through the pupil, but its construction from elastic connective tissue gives it a complex, fibrillous pattern.
 WHY IRIS ???
Research shows the iris is one of the most unique data rich physical structures on the human body. An iris has 256 independent measurable characteristics, or degrees of freedom, nearly six times as many as a finger print. Thus, the probability of two irises producing the same code is approximately 1 in 1078. , With the population of the earth being approximately 1010 people.
Thus, the performance of iris recognition is at a much higher level of scientific certainty and has many greater capabilities then any other form of Human recognition, including finger prints, Facial or voice recognition, and retinal recognition. This recognition technology is relatively new with many significant advantages, such as speed, accuracy, hardware, simplicity, and applicability.
Accurately identifying individuals is a major concern for governmental agencies, police department, medical institutions, Banking and legal institutions, and corporation, to name just a few. The importance lies in the necessity for the control of fraud, efficiency in administration, and benefits to users of various systems.
Iris has stable and distinctive features for personal identification. That is because every iris has fine and unique patterns and does not change over time since two or three years after the birth, so it might be called as a kind of optical finger print.
 THEORY AND IMPLEMENTATION
The iris identification program may be divided into four main functional blocks:
Firstly, an image containing the user’s eye is captured by the system. Then, the images preprocessed to normalize the scale and illumination of the iris and localize the iris pattern are extracted. Finally decision is made by the means of matching.
 IRIS IMAGE ACQUISITION:
An image surrounding human eye region is obtained at a distance from a CCD camera without any physical contact to the device. Figure shows the device configuration for acquiring human eye images. To acquire more clear images through a CCD camera and minimize the effect of the reflected lights caused by the surrounding illumination, we arrange two halogen lamps as the surrounding lights, as the figure illustrates. The size of the image acquired under this circumstance is 320 x 240.
 PREPROCESSING:
The acquired image always contains not only the “useful” parts (IRIS) but also some “relevant” parts (e.g. eyelid, pupil). Under some conditions, the brightness is not uniformly distributed. In addition, different eye-to-camera distance may result in different image sizes of the same eye. For the purpose of analysis, the original image needs to be processed. The processing is composed of two steps:
1. Iris Localization.
2. Edge Detection.
• Iris Localization:
In this stage, we should determine an iris part of the image by localizing the position of the image derived from inside the limbus (outer boundary) and outside the pupil (inner boundary), and finally convert the iris part into a suitable representation. Because there is some obvious difference in the intensity around each boundary, an edge detection method is easily applied to acquire the edge information.
• Edge Detection:
It is used to find complex object boundaries by marking potential edge points corresponding to places in an image where rapid change in brightness occurs. After edge points have been marked, they can be merged to for lines. Edge detection operators are based on idea that edge information in an image is found by looking at the relationship of a pixel with its neighbors. In other words, edge is defined by discontinuity in gray values. An edge separates two distinct objects.
FEATURE EXTRACTION:
The approach used for the feature extraction is to extract the relevant pixel values from the iris image using the Fast Fourier Transform (FFT). Before seeing something about FFT, let us see what Fourier Transform is:
In many signal-processing applications, the distinguishing features of signals are mostly interpreted in the frequency domain. The main analytic tool for the frequency domain properties of discrete time signals and the frequency domain behavior of discrete time system is the Fourier Transform. We extract these features by extracting a buffer of 256 pixel values. Once this buffer is extracted, we find the FFT (Fast Fourier Transform) of the extracted buffer. The modules value of FFT is stored as an array in the database and this is used for matching a test image with one’s available in database. Also the phase is calculated and is stored in other array.
IRIS IDENTIFICATION USING PATTERN MATCHING:
The pattern matching process may be decomposed into four parts:
1. Bringing the newly acquired iris pattern into spatial alignment with a candidate database entry.
2. Choosing a representation of the aligned iris pattern that makes their distinctive pattern apparent.
3. Evaluating the goodness of a match between the newly acquired and database representation.
4. Deciding if the newly acquired data and the database entry were derived from the same iris based on the goodness of the match.
The comparison between a new “test” iris code and database of existing codes is performed in the following manner:
The exclusive-OR function is taken for each difference between the two codes. Bit #1 from the reference iriscode code record, bit #2 from the presented iriscode record is compared to bit #2 from the reference iriscode record, and so on. If two bits are alike, the system assigns a value of zero to that pair comparison. If the two bits are different, the system assigns a value of one to that pair comparison. After all pairs are compared, the total number of bit-pair divides the number of disagreeing bit-pairs. This value is termed as “Hamming Distance”. A Hamming Distance of .10 means that two iriscode records differed by 10%.
At Hamming Distance (.342), the probability of a False Reject is approximately the same as the probability of a False Accept. When two iris code differ I more than 34.2% of their bits, they are considered to be different, if fewer than 34.2% of their bits difference they are considered to be from identical irises.
 Using Euclidian Distance:
Euclidian Distance =  (Ai -Bi) 2
Where, Ai = Absolute FFT element of test image.
Bi = Absolute FFT elements of image from database.
The minimum Euclidian distance corresponds to the image in the database, which matches most closely, with the image. A very high threshold level for Euclidian distance is set so as to accept the image in database as the correctly matched image with high authencity. The Euclidian distance above which the image is declared as rejected is said to be 0.005, whereas the typical Euclidian distance for other images are of the order of 103 and 104
SOME CURRENT AND FUTURE APPLICATIONS OF IRIS RECOGNITION:
 Secure accesses to bank cash machine accounts:
The banks of United, Diebold and Sensar have applied it. After enrolling once (a “30 ”second process), the customers need only approach the ATM, follow the instruction to look at the camera, and be recognized within 2-4 seconds. The ultimate aim is to provide safe and secure transactions.
 Ticket less, document-free air travel:
Passengers and airline employees will store digital images of their irises on a database. After the image of your iris is on the file, a video camera will be able to instantly verify your identity and clear you to board the aircraft.
 Computer login: the iris an living password.
 National border controls: the iris as a living passport.
 Premises access control (homes, office, laboratory).
 Credit card authentication.
 Secure financial transactions.
 Internet security.
ADVANTAGES OF THE IRIS IDENTIFICATION:
 Highly protected internal organ of the eye.
 Externally visible patterns imaged from a distance.
 Iris patterns possesses a high degree of randomness
 Variability: 266 degrees-of-freedom.
 Limited genetic penetrance of iris patterns.
 Patterns apparently stable throughout life.
 Encoding and decision-making are tractable.
 Image analysis and encoding time: 1 second.
 Search speed: 100,000 Iris codes per seconds.
DISADVANTAGES OF THE IRIS FOR IDENTIFICATION:
 Small targets (1 cm) to acquire from a distance of 1m.
 Moving target…within another…on yet another.
 Obscured by eyelashes, reflections.
 Partially occluded by eyelids.