IRIS Scanning System - Engineering Seminar

IRIS scanning system
                                  Iris recognition is an automated method of biometric identification that uses mathematical pattern-recognition techniques on video images of the irides of an individual's eyes, whose complex random patterns are unique and can be seen from some distance.Not to be confused with another, less prevalent, ocular-based technology, retina scanning, iris recognition uses camera technology with subtle infrared illumination to acquire images of the detail-rich, intricate structures of the iris. Digital templates encoded from these patterns by mathematical and statistical algorithms allow unambiguous positive identification of an individual.
Many millions of persons in several countries around the world have been enrolled in iris recognition systems, for convenience purposes such as passport-free automated border-crossings, and some national ID systems based on this technology are being deployed. A key advantage of iris recognition, besides its speed of matching and its extreme resistance to False Matches, is the stability of the iris as an internal, protected, yet externally visible organ of the eye.
The core algorithms that underlie iris recognition were developed in the 1990's by Professor John Daugman, Ph.D, OBE (University of Cambridge Computer Laboratory). These were licensed to many developers of commercial iris cameras and systems including LG Electronics, Oki, Panasonic, Sagem, IrisGuard, and Sarnoff Labs. As of 2008, Daugman's algorithms are the basis of all commercially deployed iris recognition systems, although many alternative approaches have been studied and compared in the academic literature in hundreds of publications. Iris recognition remains a very active research topic in computing, engineering, statistics, and applied mathematics.
Like face recognition, voice biometrics provide a way to authenticate identity without the subject's knowledge. It is easier to fake (using a tape recording); it is not possible to fool an analyst by imitating another person's voice. Iris scanning is a method of biometric identification; pattern recognition is used to determine the identity of the subject. 
Iris scans create high-resolution images of the irides of the eye; IR illumination is used to reduce specular reflection from the cornea. The iris itself is a "good subject" for biometric identification, because it is an internal organ that is well protected, it is mostly flat and it has a fine texture that is unique even for identical twins. 
Iris scans can be done regardless of whether the subject is wearing contact lenses or glasses. However, it is necessary for the system to take eye lids and eye lashes into account; both can obscure the necessary parts of the eye and cause false information to be added into automated systems. 
Iris scans are extremely accurate. 
Iris recognition is a biometric identification technology that uses high-resolution images of the irides of the eye. The iris of the eye is well suited for authentication purposes. It is an internal organ protected from most damage and wear, it is practically flat and uniform under most conditions and it has a texture that is unique even to genetically identical twins. 
Iris recognition is accomplished by applying proprietary algorithms for image acquisition and subsequent one-to-many matching that were developed initially by John G. Daugman, Ph.D., OBE. 
Iris recognition algorithms produce remarkable results. Daugman's algorithms have produced accuracy rates in authentication that are better than those of any other method. IrisCode, a commercial system derived from Daugman's work, has been used in the United Arab Emirates as a part of their immigration process. After more than 200 billion comparisons, there has never been a false match.
Earlier today, we revealed the future of security and advertising -- and how it will all fall under the watchful beam of an iris scanner. The company behind the technology, Global Rainmakers Inc., has big plans for the system, which is launching in the city of Leon, Mexico. To help wrap our heads around the project, we spoke with Jeff Carter, chief business development officer of GRI, to find out how it will change our lives. 
What is an iris?
The colored part of the eye is called the iris. It is an internal organ that is part of the eye; protected by the eyelids It is the only internal organ of the body that is externally visible. The iris is depicted in the below drawing.
The iris has distinctive patterns that allow for very accurate person identification, far more accurate than fingerprints or facial recognition.   

Biometric dates back to ancient Egyptians who measured people to identify them. Biometric devices have three primary components.
1. Automated mechanism that scans and captures a digital or analog image of a living personal characteristic
2. Compression, processing, storage and comparison of image with a stored data.
3. Interfaces with application systems.
A biometric system can be divided into two stages: the enrolment module and the identification module. The enrolment module is responsible for training the system to identity a given person. During an enrolment stage, a biometric sensor scans the person’s physiognomy to create a digital representation.
 A feature extractor processes the representation to generate a more compact and expressive representation called a template. For an iris image these include the various visible characteristics of the iris such as contraction, Furrows, pits, rings etc. The template for each user is stored in a biometric system database. The identification module is responsible for recognizing the person. During the identification stage, the biometric sensor captures the characteristics of the person to be identified and converts it into the same digital format as the template. The resulting template is fed to the feature matcher, which compares it against the stored template to determine whether the two templates match. The identification can be in the form of verification, authenticating a claimed identity or recognition, determining the identity of a person from a database of known persons.
 In a verification system, when the captured characteristic and the stored template of the claimed identity are the same, the system concludes that the claimed identity is correct. In a recognition system, when the captured characteristic and one of the stored templates are the same, the system identifies the person with matching template.

  Biometrics encompasses both physiological and behavioral characteristics. A physiological characteristic is a relatively stable physical feature such as finger print, iris pattern, retina pattern or a Facial feature. A behavioral trait in identification is a person’s signature, keyboard typing pattern or a speech pattern. The degree of interpersonal variation is smaller in a physical characteristic than in a behavioral one. For example, the person’s iris pattern is same always but the signature is influenced by physiological characteristics.
Disadvantages:-Even though conventional methods of identification are indeed inadequate, the biometric technology is not as pervasive and wide spread as many of us expect it to be. One of the primary reasons is performance. Issues affecting performance include accuracy, cost, integrity etc.
Even if a legitimate biometric characteristic is presented to a biometric system, correct authentication cannot be guaranteed. This could be because of sensor noise, limitations of processing methods, and the variability in both biometric characteristic as well as its presentation.
Cost is tied to accuracy; many applications like logging on to a pc are sensitive to additional cost of including biometric technology.

Iris identification technology is a tremendously accurate biometric. Iris recognition leverages the unique features of the human iris to provide an unmatched identification technology. So accurate are the algorithms used in iris recognition that the entire planet could be enrolled in an iris database with only a small chance of false acceptance or false rejection. The technology addresses the FTE (Failure to Enroll) problems which lessen the effectiveness of other biometrics. Only the iris recognition technology can be used effectively and efficiently in large scale identification implementations. The tremendous accuracy of iris recognition allows it, in many ways, to stand apart from other biometric technologies.

The word IRIS dates from classical times (a rainbow). The iris is a Protective internal organ of the eye. It is easily visible from yards away as a colored disk, behind the clear protective window of the cornea, surrounded by the white tissue of the eve. It is the only internal organ of the body normally visible externally. It is a thin diaphragm stretching across the anterior portion of the eye and supported by lens. This support gives it the shape of a truncated cone in three dimensions. At its base the eye is attached to the eye’s ciliary body. At the opposite end it opens into a pupil. The cornea and the aqueous humor in front of the iris protect it from scratches and dirt, the iris is installed in its own casing. It is a multi layered structure. It has a pigmented layer, which forms a coloring that surrounds the pupil of the eye. One feature of this pupil is that it dilates or contracts in accordance with variation in light intensity. The human iris begins to form during the third month of gestation. The structures creating its distinctive pattern are completed by the eighth month of gestation hut pigmentation continues in the first years after birth. The layers of the iris have both ectodermic and embryological origin, consisting of: a darkly pigmented epithelium, pupillary dilator and sphincter muscles, heavily vascularized stroma and an anterior layer chromataphores with a genetically determined density of melanin pigment granules. The combined effect is a visible pattern displaying various distinct features such as arching ligaments, crypts, ridges and zigzag collaratte. Iris color is determined mainly by the density of the stroma and its melanin content, with blue irises resulting from an absence of pigment: long wavelengths are penetrates and is absorbed by the pigment epithelium, while shorter wavelengths are reflected and scattered by the stroma. The heritability and ethnographic diversity of iris color have long been studied. But until the present research, little attention had been paid to the achromatic pattern complexity and textural variability of the iris among individuals. A permanent visible characteristic of an iris is the trabecular mesh work, a tissue which gives the appearance of dividing the iris in a radial fashion. Other visible characteristics include the collagenous tissue of the stroma, ciliary processes, contraction furrows, crypts, rings, a corona and pupillary frill coloration and sometimes freckle. The striated anterior layer covering the trabecular mesh work creates the predominant texture with visible light.                                                                                                                                                                     

Iris is the focus of a relatively new means of biometric identification. The iris is called the living password because of its unique, random features. It is always with you and can not be stolen or faked. The iris of each eye is absolutely unique. The probability that any two irises could be alike is one in 10 to 78th power — the entire human population of the earth is roughly 5.8 billion. So no two irises are alike in their details, even among identical twins. Even the left and right irises of a single person seem to be highly distinct. Every iris has a highly detailed and unique texture that remains stable over decades of life. Because of the texture, physiological nature and random generation of an iris artificial duplication is virtually impossible. The properties of the iris that enhance its suitability for use in high confidence identification system are those following. 
1. Extremely data rich physical structure about 400 identifying features
 2. Genetic independence no two eyes are the same.
3. Stability over time.
4. Its inherent isolation and protection from the external environment.
5. The impossibility of surgically modifying it without unacceptable risk to vision.
6. Its physiological response to light, which provides one of several natural tests against artifice.
7. The ease of registering its image at some distance forms a subject without physical contact. unobtrusively and perhaps inconspicuously
8. It intrinsic polar geometry which imparts a natural co-ordinate system and an origin of co-ordinates
9. The high levels of randomness in it pattern inter subject variability spanning 244 degrees of freedom - and an entropy of 32 bits square million of iris tissue.

The idea of using patterns for personal identification was originally proposed in 1936 by ophthalmologist Frank Burch. By the 1980’s the idea had appeared in James Bond films, but it still remained science fiction and conjecture. In 1987, two other ophthalmologists Aram Safir and Leonard Flom patented this idea and in 1987 they asked John Daugman to try to create actual algorithms for this iris recognition. These algorithms which Daugman patented in 1994 are the basis for all current iris recognition systems and products. Daugman algorithms are owned by Iridian technologies and the process is licensed to several other Companies who serve as System integrators and developers of special platforms exploiting iris recognition in recent years several
products have been developed for acquiring its images over a range of distances and in a variety of applications. One active imaging system developed in 1996 by licensee Sensar deployed special cameras in bank ATM to capture IRIS images at a distance of up to 1 meter. This active imaging system was installed in cash machines both by NCR Corps and by Diebold Corp in successful public trials in several countries during I997 to 1999. a new and smaller imaging device is the low cost “Panasonic Authenticam” digital camera for handheld, desktop, ecommerce and other information security applications. Ticket less air travel, check-in and security procedures based on iris recognition kiosks in airports have been developed by eye ticket. Companies in several, countries are now using Daughman’s algorithms in a variety of products.


The iris recognition process begins with video-based image acquisition a process which deals with the capturing of a high quality image of the iris while remaining non-invasive to the human operator. There are 3 important requisites for this process
a) It is desirable to acquire images of the iris with sufficient resolution and sharpness to support recognition
b) 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.
c) These images must be well framed without unduly constraining the operator. The widely used recognition system is the daugmen system which captures images with the iris diameter typically between 100 and 200 pixels from a distance of 15, 46 cm using a 330 mm lens.

Image acquisition is performed as follows. It uses LED based point light sources in conjunction with a wide angle camera no more than 3 feet from the subject’s eye. By carefully positioning the light source below the operator, reflection of point source can be avoided in the imaged iris. The system makes use of light, which is visible to human eye. Infrared illumination can also be
employed. This System requires the operator to self position his eye in front of the camera. It provides the operator with a live video feed back via beam splitter This allows the operator to see what the camera is capturing and to adjust his position. Once a series of images of sufficient quality is acquired, it is
automatically forwarded for subsequent processing.

Image acquisition of iris can be expected to yield an image containing only the iris. Rather image acquisition will capture the iris as part of a larger image that also contains data derived from the surrounding eye region. Prior to performing iris pattern matching it is important to localize that portion of the image that corresponds to iris. Iris localization is a process that delimits the iris
from the rest of the acquired image. After the camera situates the eye, the Daugman’s algorithm narrows in from the right and left of the iris to locate its outer edge. This horizontal approach accounts for obstruction caused by the eyelids. It simultaneously locates the inner edge of the iris, excluding the lower 900 because of inherent moisture and lighting issues. Conversion of an iris image into a numeric code that can be easily manipulated is essential to its use. This process developed by John Daugman. Permits efficient comparison of irises. Upon the location of the iris, an iris code is computed based on the information from a set of Gabor wavelets. The Gabor
wavelet is a powerful tool to make iris recognition practical. These wavelets are specialized filter banks that extract in formation from a signal at a variety of locations and scales. The filters are members of a family of functions developed by Dennis Gabor, that optimizes the resolution in both spatial and frequency domains. The 2-D Gabor wavelets filter and map segments of iris into hundreds
of vectors. The wavelets of various sizes assign values drawn from the orientation and spatial frequency of select areas, bluntly referred to as the “what” of the sub-image, along with the position of these areas, bluntly referred to as the “where”. The “what” and where are used to form the Iris Code. Not all of iris is used: a portion of the top, as well as 450 of the bottom is unused to account for eyelids and camera—light reflections. The iris Code is calculated using 8 circular bands that have been adjusted to the iris and pupil boundaries. Iris recognition technology converts the visible characteristics of the iris into a 512 byte Iris Code, a template stored for future verification attempts. 5l2 bytes is a fairly compact size for a biometric template, but the quantity of  information derived from the iris is massive. From the iris 11 mm diameters, Dr. Daugman’s algorithms provide 3.4 bits of data per square mm. This density of information is such that each iris can be said to have 266 unique “spots”, as opposed to 13- 60 for traditional biometric technology. This 266 measurement is cited in all iris recognition literature, after allowing for the algorithms for relative functions and for characteristics inherent to most human eyes. Dr. Daugman concludes that 173 “independent binary degrees of freedom can be extracted
from his algorithm-and exceptionally large number fur a biometric, for future identification, the database will not be comparing images of iris, but rather hexadecimal representations of data returned by wavelet filtering and mapping The Iris Code for an iris is generated within one second. Iris Code record is immediately encrypted and cannot be reverse engineered  
When a live iris is presented for comparison, the iris pattern is processed and encoded into 512 byte Iris Code. The Iris Code derived from this process is compared with previously generated Iris Code. This process is called pattern matching. Pattern matching evaluates the goodness of match between the newly acquired iris pattern and the candidate’s data base entry. Based on this goodness of match final decision is taken whether acquired data does or doesn’t come from
the same iris as does the database entry. Pattern matching is performed as follows. Using integer XOR logic in a single clock cycle, a long vector of each to iris code can be XORed to generate a
new integer. Each of whose bits represent mismatch between the vectors being compared. The total number of 1s represents the total number of mismatches between the two binary codes. The difference between the two recodes is expressed as a fraction of mismatched bits termed as hamming distance. For two identical Iris Codes, the hamming distance is Zero. For perfectly unmatched Iris Codes, the hamming distance is 1. Thus iris patterns are compared. The entire
process i.e. recognition process takes about 2 seconds. A key differentiator for
iris recognition is its ability to perform identification using a one to many search of a database, with no limitation on the number of iris code records contained there in.

An “Iris Code” is constructed by demodulation of the iris pattern. This process uses complex-valued 2D Gabor wavelets to extract the structure of the iris as a sequence of phasors, whose phase angles are quantized to set the bits in the first code. This process is performed in a doubly—dimensionless polar co-ordinate system that is invariant to the size of the iris, and also invariant to the dilation diameter of the pupil within the iris. The demodulating wavelets are parameterized with four degrees-offreedom: Size, orientation and two positional co-ordinates. They span several octaves in size, in order to extract iris structure at many different scales of
analysis. Because the information extracted from the iris is inherently described in terms of phase, it is insensitive to contrast, camera gain and illumination level. The phase description is very compact, requiring only 256 bytes to represent each iris pattern. These 2D wavelets are optimal encoders under the inherent Heisenberg—Weyl uncertainty relation for extraction of information in conjoint spatial-spectral representations. The recognition of irises by their recodes is based upon the failure of a test of statistical independence. Any given Iris Code is statistically guaranteed to pass a test of independence against any Iris Code computed from a different eye;
but it will uniquely fail the same test against the eye from which it was composed. Thus the key to iris recognition is the failure of’ a test of statistical independence.

The Iris Code constructed from these Complex measurements provides such a tremendous wealth of data that iris recognition offers level of accuracy orders of magnitude higher than biometrics. Some statistical representations of the accuracy follow:
· The odds of two different irises returning a 75% match (i.e. having Hamming Distance of 0.25): 1 in 10 16.
· Equal Error Rate (the point at which the likelihood of a false accept and false reject are the same): 1 in 12 million.
· The odds of two different irises returning identical Iris Codes: 1 in 1052 Other numerical derivations demonstrate the unique robustness of these
algorithms. A person’s right and left eyes have a statistically insignificant increase in similarity: 0.0048 on a 0.5 mean. This serves to demonstrate the
hypothesis that iris shape and characteristic are phenotype - not entirely; determined by genetic structure. The algorithm can also account for the iris: even if 2/3 of the iris were completely obscured, accurate measure of the remaining third would result in an equal error rate of 1 in 100000. Iris recognition can also accounts for those ongoing changes to the eye and iris which are defining aspects of living tissue. The pupil’s expansion and contraction, a constant process separate from its response to light, skews and stretches the iris. The algorithms account for such alteration after having located at the boundaries of the iris. Dr. Daugman draws the analogy to a ‘homogenous rubber sheet” which, despite its distortion retains certain consistent qualities. Regardless of the size of the iris at any given time, the algorithm draws on the same amount or data, and its resultant Iris Code is stored as a 512 byte template. A question asked of all biometrics is there is then ability to determine fraudulent samples. Iris recognition can account for this in several ways the detection of pupillary changes, reflections from the cornea detection of contact lenses atop the cornea and use of infrared illumination to determine the state of the sample eye tissue.

The performance of any biometric identification scheme is characterized by its “Decision Environment’. This is a graph superimposing the two fundamental histograms of similarity that the test generates: one when comparing biometric measurements from the SAME person (different times, environments, or conditions), and the other when comparing measurements from DIFFERENT persons. When the biometric template of a presenting person is compared to a
previously enrolled database of templates to determine the Persons identity, a criterion threshold (which may be adaptive) is applied to each similarly score Because this determines whether any two templates are deemed to be “same” or “different”, the two fundamental distributions should ideally be well separated as any overlap between them causes decision errors. One metric for “decidability”, or decision—making power, is d. This is defined as the separation between the means of two distributions, divided by the square root of their average variance. One advantage of using d’ for comparing, the decision-making power of biometrics is the fact that it does not depend on any choice about the decision threshold used. Which of course may vary from liberal to conservative when selecting the trade-off between the False Accept Rate (FAR) and False Reject Rate (FRR)? The d’ metric is a measure of the inherent degree to which any decrease in one error rate must be paid for by an increase in the other error rate, when decision thresholds are varied. It reflects the intrinsic separability of the two distributions. Decidability metrics such as d’ can he applied regardless of what measure of similarity a biometric uses. In the particular case of iris recognition, the similarly measure is a Hamming distance: the fraction of bits in two iris Codes that disagree. The distribution on the left in the graph shows the result when  different images of the same eye are compared: typically about 10% of the bits may differ. But when Iris Codes from different eyes are compared: With rations to look for and retain the best match. The distribution on the right is the result. The fraction of disagreeing bits is very tightly packed around 45%. Because of the narrowness of this right-hand distribution, which belongs to the family of binomial extreme-value distributions, it is possible to make identification decisions with astronomical levels of confidence. For example, the odds of two different irises agreeing just by chance in more than 75 of their Iris Code bits, is only one in 10-to-the- 14th power. These extremely low probabilities of getting a False Match enable the iris recognition algorithms to search through extremely large databases, even of a national or planetary scale, without confusing one Iris Code for another despite so many error opportunities 

Although the striking visual similarity of identical twins reveals the genetic penetrance of overall facial appearance, a comparison of genetically dentical irises reveals just the opposite for iris patterns: the iris texture is an epigenetic phenotypic feature, not a genotypic feature. A convenient source of genetically identical irises is the right and left pair from any given person. Such pairs have the same genetic relationship as the four irises of two identical twins, or indeed in the probable future, the 2N irises of N human clones. Eye color of course has high genetic penetrance, as does the overall statistical quality of the iris texture, but the textural details are uncorrelated and independent even in genetically identical pairs. So performance is not limited by the birth rate of identical twins or the existence of genetic relationships.

It is important to establish and to measure the amount of independent variation both within an iris and between different irises. There are correlations within an iris because local structure is self-predicting; for example, furrows tend to propagate themselves radially. Such self-correlations limit the number of
degrees of freedom within irises. But even more important is the question of whether systematic correlations exist between different irises.  

This probability distribution suggests that they do not. It plots the probability that bits in different positions within the Iris Code are set to 1, for a   randomly sampled population of different Iris Codes. The fact that this distribution hovers near 0.5 indicates that all Iris Code bits are equally likely to be 0 or 1. This property makes Iris Codes “maximum entropy” codes in a bitwise sense. The fact that this distribution is uniform indicates that different irises do not systematically share any common structure. For example, if most irises had a furrow or crypt in the 12-o’clock position, then the plot shown here would not be flat. The recognition of persons by their Iris Codes is based Upon the failure of a test of statistical independence. The plot shown here illustrates why any given Iris Code is “statistically guaranteed” to pass a test of independence
against any Iris Code computed from a different eye. 

The Histogram given below shows the outcomes of 2,307,025  comparisons between different pairs of irises. For each pair comparison, the percentage of their Iris Code bits that disagreed was computed and tallied as a fraction. Because of the zero-mean property of the wavelet demodulators, the computed coding bits are equally likely to be 1 or 0. Thus when any corresponding bits of two different Iris Codes are compared, each of the four combinations (00), (01), (10), (11) has equal probability. In two of these cases the bits agree, and in the other two they disagree. Therefore one would expect on average 50% of the bits between two different Iris Codes to agree by chance. The above histogram presenting comparisons between 2.3 million different pairings of irises shows a mean fraction of 0.499 of their Iris Code bits agreeing by chance.

A critical feature of this coding approach is the achievement of commensurability among iris codes, by mapping all irises into a representation having universal format and constant length, regardless of the apparent amount of iris detail. In the absence of commensurability among the codes, one would be faced with the inevitable problem of comparing long codes with short codes, showing partial agreement and partial disagreement in their lists of features. It is not obvious mathematically how one would make objective decisions and compute confidence levels on a rigorous basis in such a situation. This difficulty has hampered efforts to automate reliably the recognition of fingerprints. Commensurability facilitates and objectifies the code comparison process, as well as the computation of confidence.

· Highly protected, internal organ of the eye
· Externally visible; patterns imaged from a distance
· Iris patterns possess a high degree of randomness
o Variability: 244 degrees-of-freedom
o Entropy: 3.2 bits per square-millimeter
o Uniqueness: set by combinatorial complexity
· Changing pupil size confirms natural physiology
· Pre-natal morphogenesis (7th month of gestation)
· Limited genetic penetrance of iris patterns
· Patterns apparently stable throughout life
· Encoding and decision-making are tractable
· Image analysis and encoding time: 1 second
· Decidability index (d-prime): d' = 7.3 to 11.4
· Search speed: 100,000 Iris Codes per second

· Small target (1 cm) to acquire from a distance (1m)
· Located behind a curved, wet, reflecting surface
· Obscured by eyelashes, lenses, reflections
· Partially occluded by eyelids, often drooping
· Deforms non-elastically as pupil changes size
· Illumination should not be visible or bright
Visible Wavelength (VW) vs Near Infrared (NIR) Imaging
The majority of iris recognition cameras use Near Infrared (NIR) imaging by emitting 750nm wavelength low-power light. This is done because dark-brown eyes, possessed by the majority of the human population, reveal rich structure in the NIR but much less in the visible band (400 - 700nm), and also because NIR light is invisible and unintrusive. A further important reason is that by allowing only this selected narrow band of illuminating light back into the camera via its filters, most of the ambient corneal reflections from a bright environment are blocked from contaminating the iris patterns.
The melanin, also known as chromophore, mainly consists of two distinct heterogeneous macromolecules, called eumelanin (brown–black) and pheomelanin (yellow–reddish).NIR imaging is not sensitive to these chromophores, and as a result they do not appear in the captured images. In contrast, visible wavelength (VW) imaging keeps the related chromophore information and, compared to NIR, provides rich sources of information mainly coded as shape patterns in iris. Hosseini, et al. provide a comparison between these two imaging modalities and fused the results to boost the recognition rate. An alternative feature extraction method to encode VW iris images was also introduced, which is highly robust to reflectivity terms in iris. Such fusion results are seemed to be alternative approach for multi-modal biometric systems which intend to reach high accuracies of recognition in large databanks.

Operating principle
An IriScan model 2100 iris scanner
An iris-recognition algorithm first has to localize the inner and outer boundaries of the iris (pupil and limbus) in an image of an eye. Further subroutines detect and exclude eyelids, eyelashes, and specular reflections that often occlude parts of the iris. The set of pixels containing only the iris, normalized by a rubber-sheet model to compensate for pupil dilation or constriction, is then analyzed to extract a bit pattern encoding the information needed to compare two iris images. In the case of Daugman's algorithms, a Gabor wavelet transform is used. The result is a set of complex numbers that carry local amplitude and phase information about the iris pattern. In Daugman's algorithms, most amplitude information is discarded, and the 2048 bits representing an iris pattern consist of phase information (complex sign bits of the Gabor wavelet projections). Discarding the amplitude information ensures that the template remains largely unaffected by changes in illumination or camera gain (contrast), and contributes to the long-term usability of the biometric template. For identification (one-to-many template matching) or verification (one-to-one template matching), a template created by imaging an iris is compared to stored template(s) in a database. If the Hamming distance is below the decision threshold, a positive identification has effectively been made because of the statistical extreme improbability that two different persons could agree by chance ("collide") in so many bits, given the high entropy of iris templates.


  • 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) that 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. Like the fingerprint, it is very hard (if not impossible) to prove that the iris is unique. However, there are so many factors that go into the formation of these textures (the iris and fingerprint) that the chance of false matches for either is extremely low. Even genetically identical individuals have completely independent iris textures.
  • 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 being 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 fingerprint scanners, where a finger has to touch a surface, or retinal scanning, where the eye must be brought very close to an eyepiece (like looking into a microscope).
  • The commercially deployed iris-recognition algorithm, John Daugman'sIrisCode, has an unprecedented false match rate (better than 10−11 if a Hamming distance threshold of 0.26 is used, meaning that up to 26% of the bits in two IrisCodes are allowed to disagree due to imaging noise, reflections, etc., while still declaring them to be a match).
  • While there are some medical and surgical procedures that can affect the colour and overall shape of the iris, the fine texture remains remarkably stable over many decades. Some iris identifications have succeeded over a period of about 30 years.
  • Shortcomings.
  • Many commercial Iris scanners can be easily fooled by a high quality image of an iris or face in place of the real thing.
  • The scanners are often tough to adjust and can become bothersome for multiple people of different heights to use in succession.
  • The accuracy of scanners can be affected by changes in lighting
  •  Iris scanners are significantly more expensive than some other forms of biometrics, password or prox card security systems
  • 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 fingerprint 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. However, several academic institutions and biometric vendors are developing products that claim to be able to identify subjects at distances of up to 10 meters ("standoff iris" or "iris at a distance" as well as "iris on the move" for persons walking at speeds up to 1 meter/sec).
  • 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
  • As 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
  •  Analysing the temporal frequency spectrum of the image for the peaks caused by computer displays
  • 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

Why did GRI choose iris scans ?  
Jeff Carter: Well, one of the big problems in corporate America is reference data--that is, all the data that is about us. We don't have any way to link it all together. It's one of the reasons why your bank account doesn't reconcile until 48 hours later because there's all this data behind it that they have to execute manually. 
When you look at the ways to link the data together, biometrics is the obvious choice. With a fingerprint, for instance, there's about 100 recognizable data points. For a really great fingerprint, you may get about 15 points--and that's if it's perfect. Of that, you only need 7 or 8 points to convict. So essentially, you only need 7 or 8 points across a huge population of people. It's one of the reasons fingerprints is causing so many problems.
With iris, you have over 2,000 points. Those 2,000 points appear when you're born. When you're in your mother's womb, your iris tears in a unique fashion. That tear stays constant until the day you die. If you die, and your body loses blood pressure, the eye flattens. So while a lot of what you see in Minority Report is very real today, the part about pulling out eyeballs is not real. 
With those 2,000 points, you can create a unique 16,000 bit stream of numbers that represents every human on the planet. That provides a reference point that can connect everything you do in all aspects of life, for the first time ever.
What are some of the innovations of GRI's iris technology?
Iris has been around for a long time. The technology that really everyone uses except us was developed by John Daugman. His understudy was Dr. Keith Hanna, who is one of our founders and chief technology officer. He also invented the yellow line in NFL football. Those two individuals are the top iris specialists in the world. Daugman really focused on matching. So, once you had an image of an iris, how you would match it across a billion participants to make sure you had the right one. 
Here's what is different with our technology. Hanna said, Anyone can match--it's simply a numbers game. He focused on acquiring the iris. In motion.From a distance. Even with the technology in airports several years ago, you had to hold still for about 30 seconds so it could find you. If you moved, it would blur. Most all of our competitors have that same issue. Ours is different. You can move. 
So we've even worked with three-letter agencies on technology that can capture 30-plus feet away. In certain spaces, eventually, you'll be able to have maybe one sensor the size of a dime, in the ceiling, and it would acquire all of our irises in motion, at a distance, hundreds--probably thousands as computer power continues to increase--at a time.
Do you believe this technology will be more important for security forces or advertisers?
I just use advertising as an example because it's something we all have experience with. But it's really all aspects of life. I liken it to what happened when we went from radio to TV. It's just a different world.
But what's important for advertisers is that this technology will determine your geo-location based on the iris acquisition and your spatial location. So: Where are you in that space? And, based on how you are looking and moving, and your acceleration, what is your intent?
In a retail environment, determining intent will be very important. Are you coming into the store? Are you leaving? Do you have packages? Are you looking at a sign? A sale? Matching that intent based on a lot of preferences that are all opt-in. 
The best security technology available that would give the TSA an alternative controversial body scanners is already in use worldwide -- just not here in the U.S. 
And it won't be here any time soon, either.
Thanks to privacy concerns and infrastructure issues, iris scanners aren't planned for the U.S., a DHS spokesman told Airports and security checkpoints could use the machines, which take an instant picture of the eyeball from a few feet away and compare it against an internal database, in the hunt for terror suspects or illegal immigrants. They're not. 
But nothing has stopped the United Arab Emirates, India and Jordan who already use the technology at airports and border crossings, and a major U.S. company will soon announce another major deployment elsewhere in the world.
“In UAE, we've scanned more than 40 million people from all nationalities and caught 600,000 trying to come back over the years by changing their name,” ImadMalhas, the founder of manufacturer IrisGuard, told
India has already enrolled about 600 million people in an initial phase, said Joe O’Carroll, the vice president at the company, which has deployed its scanners in Jordan and the United Arab Emirates (UAE).
Jeff Carter, the chief data officer at Hoyes Group, told that iris scanning is the best identification method available. He says a fingerprint only has about 100 points to identify, and even a perfect capture uses only 15 points. False IDs occur in about 1 out of every 10,000 captures.
Facial recognition systems, sometimes used to scan for terror suspects at public events, are even worse: they falsely identify one out of 100 captures. Iris scanning uses 2,048 points of the eye and a false identification occurs only once for every 100 million scans, Carter said. 
So why aren't we using this wonder technology in the U.S.? It's not for lack of trying.
Last October, the Department of Homeland Security (DHS) set up a trial in McAllen, Texas, at a border patrol station using technology developed by Hoyos Group. Chris Ortman, a DHS spokesman, confirmed the test -- and that the U.S. wouldn't be moving forward with it any time soon. 
"It was a preliminary test of how the technology performs," Ortman told "We have no specific plans for acquiring or deploying this type of technology at this point."
The “Minority Report” problem
So the ideal border security technology in in use elsewhere, but won't be made available to guard our shores? Is a Steven Spielberg movie partly to blame?
In the sci-fi flick Minority Report, which came out nine months after 9/11, Tom Cruise plays a detective who is scanned as he walks through a shopping mall. To circumvent the biometric readers prevalent in the movie's futuristic world, Cruise replaces his eyeballs.
That's an extreme measure, for sure. But many people,viewing iris scanning as a Big Government program meant to spy on citizens, have a similar response to it.
“Iris scanning seems pretty invasive to me,” says Lisa Malmstrom, a marketing consultant in Minnesota. “I'd be inclined to put my hand on a scanner for fingerprint identification much sooner than I would my eye.”
She's not alone. The American Civil Liberties Union agrees, stating publicly that retina scans are an invasion of privacy because there is no way to control who is scanned in public and when.
Shane MacDougall, a principal at Tactical Intelligence, also called iris scanning arguably the best identification method for use at border crossings, but there are several challenges that will make it difficult to deploy in the U.S. at major airports and borders.
“We would need to deploy [the terminals] across the country in large numbers, reconfigure the software, train people on how to use them, and most importantly build a retina scan database,” he says. Building that database may be the biggest challenge of all.
MacDougall says iris scanning could also have some competition in the next few years: real-time DNA scanning. Just put a drop of saliva in a reader to pass through a checkpoint -- something privacy advocates are sure to howl about as well. 
Meanwhile, the prevalence of retina scans will only increase in foreign countries. But will you be scanned as you pass through a U.S. airport anytime soon? Don't count on it.

Iris-based identification and verification technology has gained acceptance in a number of different areas. Application of iris recognition technology can he limited only by imagination. The important applications are those following:

  • ATM’s and iris recognition: in U.S many banks incorporated iris recognition technology into ATM’s for the purpose of controlling access to one’s bank accounts. After enrolling once (a “30second” process), the customer need only approach the ATM, follow the instruction to look at the camera, and be recognized within 2-4 seconds. The benefits of such a system are that the customer who chooses to use bank’s ATM with iris recognition will have a quicker, more secure transaction.
  • Tracking Prisoner Movement: The exceptionally high levels of accuracy provided by iris recognition technology broadens its applicability in high risk, high-security installations. Iris scan has implemented their devices with great success in prisons in Pennsylvania and Florida. By this any prison transfer or release is authorized through biometric identification. Such devices greatly ease logistical and staffing problems. Applications of this type are well suited to iris recognition technology. First, being fairly large, iris recognition physical security devices are easily

integrated into the mountable, sturdy apparatuses needed or access control, The technology’s phenomenal accuracy can be relied upon to prevent unauthorized release or transfer and to identify repeat offenders re-entering prison under a different identity.

  • Computer login: The iris as a living password.
  • National Border Controls: The iris as a living password.
  • Telephone call charging without cash, cards or PIN numbers.
  • Ticket less air travel.
  • Premises access control (home, office, laboratory etc.).
  • Driving licenses and other personal certificates.
  • Entitlements and benefits authentication.
  • Forensics, birth certificates, tracking missing or wanted person
  • Credit-card authentication.
  • Automobile ignition and unlocking; anti-theft devices.
  • Anti-terrorism (e.g.:— suspect Screening at airports)
  • Secure financial transaction (e-commerce, banking).
  • Internet security, control of access to privileged information.
  • “Biometric—key Cryptography “for encrypting/decrypting messages

Every biometric technology has its own challenges. When reviewing test results, it is essential to consider the environment and protocols of the test. Much industry testing is performed in laboratory settings on images acquired in ideal conditions. Performance in a real world application may result in very different performance as there is a learning curve for would-be user of the system and not every candidate will enroll properly or quickly the first time. There are some issues which affect the functionality and applicability of iris recognition technology in particular.
The technology requires a certain amount of user interaction the enroller must hold still in a certain spot, even if only momentarily. It would be very difficult to enroll or identify a non-cooperative subject. The eye has to have a certain degree of lighting to allow the camera to capture the iris; any unusual lighting situation may affect the ability of the camera to acquire its subject. Lastly, as with any biometric, a backup plan must be in place if the unit becomes inoperable. Network crashes, power failure, hardware and software problems are but a few of the possible ways in which a biometric system would become unusable. Since iris technology is designed to be an identification technology, the fallback procedures may not be as fully developed as in a recognition schematic. Though these issues do not reduce the exceptional effectiveness of iris recognition technology, they must be kept in mind, should a company decide
to implement on iris-based solution.

The technical performance capability of the iris recognition process far surpasses that of any biometric technology now available. Iridian process is defined for rapid exhaustive search for very large databases: distinctive capability required for authentication today. The extremely low probabilities of getting a false match enable the iris recognition algorithms to search through extremely large databases, even of a national or planetary scale. As iris  echnology grows less expensive, it could very likely unseat a large portion of the biometric industry, e-commerce included; its technological superiority has already allowed it to make significant inroads into identification and security venues which had been dominated by other biometrics. Iris-based biometric technology has always been an exceptionally accurate one, and it may soon grow much more prominent.

No comments:

Post a Comment

leave your opinion