Embed Size (px)
Citation preview
Iris Recognition System.
G.Vishnu Vadhan Reddy K.Vinod Kumar3rd ECE 3rd [email protected] [email protected]
VIGNANA BHARATHI INSTITUTE OF TECHNOLOGY
Abstract:
The pressures on today’s system administrators to have secure systems are ever increasing. One area where security can be improved is in authentication and identification. Biometrics provides a promising solution meeting all demands. Biometric identification utilizes physiological and behavioral characteristics to authenticate a person’s identity which include facial recognition, fingerprints, palm prints, hand geometry, retinal patterns and iris patterns and signature, voice pattern and key stroke dynamics. Many have suffered from high cost and unsatisfactory error rates.
The technology is accurate, easy to use, non-intrusive, and difficult to forge and, despite what many people may think is actually quite a fast system once initial enrollment has taken place With new technologies the eyes are more than “windows to your soul.” People are carrying with them a living key or password that will never be forgotten and will always be there. The technology is available now through work in computer vision, pattern recognition, and man-machine interface to create a reliable lock that a person's iris pattern will open. The randomness of iris patterns has a very high dimensionality making recognition decisions reliable with a high level of confidence.
This paper, at the outset, throws light on the technology involved in iris recognition, the algorithms followed by a statistical analysis highlighting its edge over other biometric identification systems and its applications in the present day world. This paper is intended for security practitioners who are knowledgeable, but not technically or scientifically oriented.
Introduction:
Iris recognition is a method of biometric authentication that uses pattern recognition techniques based on high-resolution images of the irides of an individual's eyes. Not to be confused with another less prevalent ocular-based technology, retina scanning, iris recognition uses camera technology, and subtle IR illumination to reduce specular reflection from the convex cornea to create images of the detail-rich, intricate structures of the iris. These unique structures converted into digital templates, provide mathematical representations of the iris that yield unambiguous positive identification of an individual.Iris recognition efficacy is rarely impeded by glasses or contact lenses. Iris technology has the smallest outlier (those who cannot use/enroll) group of all biometric technologies. The only biometric authentication technology designed for use in a one-to many search environment, a key advantage of iris recognition is its stability, or template
longevity as, barring trauma, a single enrollment can last a lifetime.
Breakthrough work to create the iris recognition algorithms required for image acquisition and one-to-many matching was pioneered by John G. Daugman, Ph.D, OBE (University of Cambridge Computer Laboratory), who holds key patents on the method. These were utilized to effectively debut commercialization of the technology in conjunction with an early version of the Iris Access system designed and manufactured by Korea's LG Electronics. Daugman's algorithms are the basis of almost all currently (as of 2006) commercially deployed iris-recognition systems. It has a so far unmatched practical false-accept rate of zero; that is there is no known pair of images of two different irises that the Daughman algorithm in its deployed configuration mistakenly identifies as the same. (In tests where the matching thresholds are – for better comparability – changed from their default settings to allow a false-accept rate in the region of 10−3 to 10−4, the Iris Code false-reject rates are comparable to the most accurate single-finger fingerprint matchers.).
An Overview of Biometrics:
Biometrics refers to the automatic identification of a person based on his/her physiological or behavioral characteristics. This method of identification offers several advantages over traditional methods involving ID cards (tokens) or PIN numbers (passwords) for various reasons: (i) the person to be identified is required to be physically present at the point-of-identification; (ii) identification based on
biometric techniques obviates the need to remember a password or carry a token. With the increased integration of computers and Internet into our everyday lives, it is necessary to protect sensitive and personal data. By replacing PINs (or using biometrics in addition to PINs), biometric techniques can potentially prevent unauthorized access to ATMs, cellular phones, laptops, and computer networks. Unlike biometric traits, PINs or passwords may be forgotten, and tokens like passports and driver's licenses may be forged, stolen, or lost. Thus, biometric systems are being deployed to enhance security and reduce financial fraud. Various biometric traits are being used for real-time recognition, the most popular being face, iris and fingerprint. However, there are biometric systems that are based on retinal scan, voice, signature and hand geometry.
A biometric system is essentially a pattern recognition system which recognizes a user by determining the authenticity of a specific physiological or behavioral characteristic possessed by the user. Several important issues must be considered in designing a practical biometric system. First, a user must be enrolled in the system so that his biometric template can be captured. This template is securely stored in a central database or a smart card issued to the user. The template is retrieved when an individual needs to be identified. Depending on the context, a biometric system can operate either in verification (authentication) or an identification mode.
Verification vs. Identification:
There are two different ways to recognize a person: verification and identification. Verification involves confirming or denying a person's claimed identity. On the other hand, in identification, the system has to recognize a person (Who am I ? ) from a
list of N users in the template database. Identification is a more challenging problem because it involves 1: N matching compared to 1:1 matching for verification.
Operating principle
An iris-recognition algorithm first has to identify the approximately concentric circular outer boundaries of the iris and the pupil in a photo of an eye. The set of pixels covering only the iris is then transformed into a bit pattern that preserves the information that is essential for a statistically meaningful comparison between two iris images. The mathematical methods used resemble those of modern lossy compression algorithms for photographic images. In the case of Daugman's algorithms, a Gabor wavelet transform is used in order to extract the spatial frequency range that contains a good best signal-to-noise ratio considering the focus quality of available cameras. The result are a set of complex numbers that carry local amplitude and phase information for the iris image. In Daugman's algorithms, all amplitude information is discarded, and the resulting 2048 bits that represent an iris consist only of the complex sign bits of the Gabor-domain representation of the iris image. Discarding the amplitude information ensures that the template remains largely unaffected by changes in illumination and virtually negligibly by iris color, which contributes significantly to the long-term stability of the biometric template. To authenticate via identification (one-to many template matching) or verification (one-to one template matching) a template created by imaging the iris, is compared to a stored value template in a database. If the Hamming Distance is below the decision threshold, a positive identification has effectively been made.A practical problem of iris recognition is that the iris is usually partially covered by eye lids and eye lashes. In order to reduce the false-reject risk in such cases, additional algorithms are needed to identify the locations of eye lids and eye lashes, and exclude the bits in the resulting code from the comparison operation.
How Iris works:
Reliable automatic recognition of persons has long been an attractive goal. As in all pattern recognition problems, the key issue is the relation between interclass and intra-class variability: objects can be reliably classified only if the variability among different instances of a given class is less than the variability between different classes. For example in face recognition, difficulties arise from the fact that the face is a changeable social organ displaying a variety of expressions, as well as being an active 3D object whose image varies with viewing angle, pose, illumination, accoutrements, and age. It has been shown that for facial images taken at
least one year apart; even the best current algorithms have error rates of 43%. Against this intra-class (same face) variability, inter-class variability is limited because different faces possess the same basic set of features, in the same canonical geometry.
For all of these reasons, iris patterns become interesting as an alternative approach to reliable visual recognition of persons when imaging can be done at distances of less than a meter, and especially when there is a need to search very large databases without incurring any false matches despite a huge number of possibilities.
Figure 1: Example of an iris pattern, imaged monochromatically at a distance of about 35 cm. The outline overlay shows results of the iris and pupil localization and eyelid detection steps. The bit stream in the top left is the result of demodulation withcomplex-valued 2D Gabor wavelets to encode the phase sequence of the iris pattern.
In addition, as an internal (yet externally visible) organ of the eye, the iris is well protected from the environment and stable over time. As a planar object its image is relatively insensitive to angle of illumination and changes in viewing angle cause only affine transformations; even the nonfat net pattern distortion caused by papillary dilation is readily reversible. Finally, the ease of localizing eyes in faces, and the distinctive annular shape of the iris, Facilitate reliable and precise isolation of this feature and the creation of a size-invariant representation.
The iris begins to form in the third month of gestation and the structures creating its pattern are largely complete by the eighth month, although pigment accretion can continue into the first
postnatal years. Its complex pattern can contain many distinctive features such as arching ligaments, furrows, ridges, crypts, rings, corona, freckles, and a zigzag collarets, some of which may be seen in Figure2.
Iris color is determined mainly by the density of melanin pigment in its anterior layer and stroma, with blue irises resulting from an absence of pigment: long wavelength light penetrates and is absorbed by the pigment epithelium, while shorter wavelengths are reflected and scattered by the stroma. All testing organizations have reported a false match rate of 0 in their tests, some of which involved millions of iris pairings. This paper explains how the algorithms work, and presents new data on the statistical properties and singularity of iris patterns based on 9.1 million comparisons.
Finding an Iris in an Image
To capture the rich details of iris patterns, an imaging system should resolve a minimum of 70 pixels in iris radius. In the field trials to date, a resolved iris radius of 100 to 140 pixels has been more typical. Monochrome
CCD cameras (480 x 640) have been used because NIR illumination in the 700nm -900nm band was required for imaging to be invisible to humans. Some imaging platforms deployed a wide angle camera for coarse localization of eyes in faces, to steer the optics of a narrow-angle pan/tilt camera that acquired higher resolution images of eyes. There exist many alternative methods for finding and tracking Facial features such as the eyes
Images passing a minimum focus criterion were then analyzed to find the iris, with precise localization of its boundaries using a coarse-to-fine strategy terminating in single-pixel precision estimates of the center Coordinates and radius of both the iris and the pupil. Although the results of the iris search greatly constrain the pupil search, concentricity of these boundaries cannot be assumed. Very often the pupil center is nasal, and inferior, to the iris center. Its radius can range from 0.1 to 0.8 of the iris radius. Thus, all three parameters defining the pupillary circle must be estimated separately from those of the iris. A very effective integrodifferential operator for determining these parameters is:
Where I(x; y) is an image such as Fig 1 containing an eye. The operator searches over the image domain (x; y) for the maximum in the blurred partial derivative with respect to increasing radius r, of the normalized contour integral of I(x; y) along a circular arc
of radius r and center coordinates (x0; y0). The result of all these localization operations is the isolation of iris tissue from other image regions, as illustrated in Fig 1 by the graphical overlay on the eye.
Phase-Quadrant Demodulation Code
Figure 3: The phase demodulation process used to encode iris patterns. Local regions of an iris are projected (Eqt 2) onto quadrature 2D Gabor wavelets, generating complex-valued coefficients whose real and imaginary parts specify the coordinates of a phasorin the complex plane. The angle of each phasor is quantized to one of the four quadrants, setting two bits of phase information. This process is repeated all across the iris with many wavelet sizes, frequencies, and orientations, to extract 2,048 bits.
Altogether 2,048 such phase bits (256 bytes) are computed for each iris, but in a major improvement over the earlier (Daugman 1993) algorithms, now an equal number of masking bits are also computed to signify whether any iris region is obscured by eyelids, contains any eyelash occlusions, specular reflections, boundary artifacts of hard contact lenses, or poor signal-to-noise ratio and thus should be ignored in the demodulation code as artifact. Figure 4: Illustration that even for poorly focused eye images, the bits of a demodulation phase sequence are still set, primarily by random CCD noise. This prevents poorly focused eye images from resembling each other in the pattern matching stage, in the way that (e.g.) poorly resolved face images look alike and can be confused with each other. Only phase information is used for recognizing irises because amplitude information is not very discriminating, and it depends upon extraneous factors such as imaging contrast, illumination, and camera gain.
The Test of Statistical Independence:
The key to iris recognition is the failure of a test of statistical independence, which involves so many degrees-of-freedom that this test is virtually guaranteed to be passed whenever the phase codes for two different eyes are compared, but to be uniquely failed when any eye's phase code is compared with another version of itself. The test of statistical independence is implemented by the simple Boolean Exclusive-OR operator (XOR) applied to the 2,048 bit phase vectors that encode any two iris patterns, masked (AND'ed) by both of their corresponding mask bit vectors to prevent non-iris artifacts from influencing iris comparisons. The XOR operator N detects disagreement between any corresponding pair of bits, while the AND operator T ensures that the compared bits are both deemed to have been uncorrupted by eyelashes, eyelids, specular reflections, or other noise.
Figure 4: Distribution of Hamming Distances from all 9.1 million possible comparisons between different pairs of irises in the database. The histogram forms a perfect binomial distribution solid curve. The data implies that it is extremely improbable for two different irises to disagree in less than about a third of their phase information. Informative searches are performed at a rate of about 100,000 irises per second.
Iris Scanning Products and Services:
Network Security: Authenticam with PrivateID
The iris recognition camera can be integrated into a variety of software applications to provide security for information and electronic commerce.
Authenticam incorporates high-resolution videoconference ability. It is estimated that the return on investment (ROI) can be realized in less than a year making this highly affordable.
This innovation will allow companies to control access to computer workstations, networks, and sensitive corporate data, as well as positively identifying system users.
Next-generation ATMs: IrisPass:
Built by Japan’s OKI Electric Industry, IrisPass is currently integrating iris scanning in ATM machines in Asia and the US. IrisPass will eliminate the need for PINs to make identification of account ownership.
This technology may also be used for in-bank teller stations for account verification.
Issues to Consider
Other biometrics recognition systems include fingerprinting, palm prints, hand geometry, nail bed identification, facial recognition, and retinal scan.
Behavioral characteristics used for identification are signature dynamics, keyboard dynamics, and voice recognition.
Iris and retina scans are the most accurate of all biometric techniques and, currently, the most costly.
The entire biometrics market is projected to reach $10 billion by 2008.
Speed:
Iris recognition systems can cycle through 1,500,000 matches per minute, which is 20 times greater processing speed than any other biometrics systems.
In real-life applications this translates into an identity decision being made in seconds. The enrollment process is also speedily accomplished, typically in three minutes or less.
Safety and Perceived Invasiveness:
Enrollment and use of an iris recognition system requires no contact, only cooperation of the user.
The devices capture images of the eye from a comfortable distance without bright lights or lasers.
The Iris Code is hashed and encrypted as a security measure to prevent theft. If a person feels their recognition patterns have been compromised, re-enrollment is possible an infinite number of times by using a permuted Iris Code.
Iris recognition because it looks at the exterior part of the eye,
unlike retinal scans that look at vascular patterns inside the eye, is not invasive.
Also, there is no possibility of gathering information such as medical conditions, a possibility with retinal scans.
Both irises and retinas are stable throughout a lifetime, except in the case of degenerative diseases that may affect the retina.
There is no need to remove glasses or contact lenses during identification. As long as they do not obscure the iris, recognition can be made through them. Iris recognition can be hampered by partially occluded or drooping eyelids.
Costs:
Iris scanning can increase profitability by minimizing both costs and vulnerabilities associated with password and password management.
Research indicates US businesses spend an average of $200 per person a year on password management.
The value proposition of implementing an iris recognition system is three-dimensional. Cost, accuracy, and ease of use are all important considerations.
Recent advances in camera technology is bringing down the cost of iris recognition. Camera prices have gone down while processing ability has gone up.
The size of the camera has decreased also. With a strategic partnership with Panasonic, the costs have gone down significantly.
Both iris scanning and retina scanning are at the upper end of the scale in cost compared to other biometric systems.
Ease of Use:
Many of the users who have already encountered iris-scanning technology consider it a convenience rather than an intrusion, speeding the process of identity verification.
Glasses or contact lens use does not affect it.
Most eye surgeries do not change the iris. In the few, such as iridotomy and iridectomy, both associated with glaucoma, re-enrollment may be necessary. These technologies can be integrated into existing business systems easing the installation requirements.
In this instance, retina scanning has an advantage over iris scanning in that retina scanning utilizes a very compact template. Retina scanning requires 96 bytes while iris scanning uses 512 bytes.
A greater number of templates can be stored in a standalone device if retina scanning is employed.
Advantages:
The iris of the eye has been described as the ideal part of the human body for biometric identification for several reasons: It is an internal organ that is well
protected against damage and wear by a highly transparent and
sensitive membrane (the cornea). This distinguishes it from fingerprints, which can be difficult to recognize after years of certain types of manual labor.
The iris is mostly flat and its geometric configuration is only controlled by two complementary muscles (the sphincter pupillae and dilator pupillae), which control the diameter of the pupil. This makes the iris shape far more predictable than, for instance, that of the face.
The iris has a fine texture that – like fingerprints – is determined randomly during embryonic gestation. Even genetically identical individuals have completely independent iris textures, whereas DNA (genetic "fingerprinting") is not unique for the about 1.5% of the human population who have a genetically identical monozygotic twin.
An iris scan is similar to taking a photograph and can be performed from about 10 cm to a few meters away. There is no need for the person to be identified to touch any equipment that has recently been touched by a stranger, thereby eliminating an objection that has been raised in some cultures against finger-print scanners, where a finger has to touch a surface, or retinal scanning, where the eye can be brought very close to a lens (like looking into a microscope lens).
The only currently commercially deployed iris recognition algorithm, John Daugman's IrisCode, has an unprecedented
false match rate (better than 10−11). Not a single false match has ever been reported for this algorithm, which has already been used to cross-compare more than 200 billion combinations of iris pairs as part of the immigration procedures in the United Arab Emirates.
Disadvantages:
Iris scanning is a relatively new technology and is incompatible with the very substantial investment that the law enforcement and immigration authorities of some countries have already made into finger-print recognition.
Iris recognition is very difficult to perform at a distance larger than a few meters and if the person to be identified is not cooperating by holding the head still and looking into the camera.
As with other photographic biometric technologies, iris recognition is susceptible to poor image quality, with associated failure to enroll rates.
As with other identification infrastructure (national residents databases, ID cards, etc.), civil rights activists have voiced concerns that iris-recognition technology might help governments to track individuals beyond their will.
Security considerations:
Like with most other biometric identification technology, a still not satisfactorily solved problem with iris recognition is the problem of "live tissue verification". The reliability of any biometric identification depends on ensuring that the signal acquired and compared has actually been recorded from a live body part of the person to be identified, and is not a manufactured template. Many commercially available iris recognition systems are easily fooled by presenting a high-quality photograph of a face instead of a real face, which makes such devices unsuitable for unsupervised applications, such as door access-control systems. The problem of live tissue verification is less of a concern in supervised applications (e.g., immigration control), where a human operator supervises the process of taking the picture.Methods that have been suggested to provide some defence against the use of fake eyes and irises include:
Changing ambient lighting during the identification (switching on a bright lamp), such that the pupillary reflex can be verified and the iris image be recorded at several different pupil diameters
Analysing the 2D spatial frequency spectrum of the iris image for the peaks caused by the printer dither patterns found on commercially available fake-iris contact lenses
Using spectral analysis instead of merely monochromatic cameras to distinguish iris tissue from other material
Observing the characteristic natural movement of an eyeball (measuring nystagmus, tracking eye while text is read, etc.)
Testing for retinal retroreflection (red-eye effect)
Testing for reflections from the eye's four optical surfaces (front and back of both cornea and lens) to verify their presence, position and shape
Using 3D imaging (e.g., stereo cameras) to verify the position and shape of the iris relative to other eye features .
C O N C L U S I O N
Highly accurate, positive personal recognition is feasible today using the iris of the human eye. This unique and complex organ, which has more dimensions (Measures) of variation than any other biometric feature currently in use, remains stable throughout a lifetime and is readily available for sampling in a nonintrusive way. And has the speed required minimizing user frustration when accessing company systems. The process uses simple and non-threatening video technology to take images of the iris, digitize the features, and create a 512-byte code, which is then compared against an entire database in less than two seconds. Recognitions can then be used to control access and entry, to provide recognition information to an existing entry control system or for any other purpose where positive identification is needed. Recent testing, under U.S. Government controlled conditions, in three real-world environments, and in a variety of operational applications have proven the practicality and feasibility of the extremely accurate iris recognition for any function requiring positive recognition.
