Iris recognition is an automated method of biometric identification that uses mathematical pattern-recognition techniques on video images of one or both of the irises of an individual's eyes, whose complex patterns are unique, stable, and can be seen from some distance. The discriminating powers of all biometric technologies depend on the amount of entropy they are able to encode and use in matching. Iris recognition is exceptional in this regard, enabling the avoidance of "collisions" (False Matches) even in cross-comparisons across massive populations. Its major limitation is that image acquisition from distances greater than a meter or two, or without cooperation, can be very difficult. However, the technology is in development and iris recognition can be accomplished from even up to 10 meters away or in a live camera feed.
Retinal scanning is a different, ocular-based biometric technology that uses the unique patterns on a person's retina blood vessels and is often confused with iris recognition. Iris recognition uses video camera technology with subtle near infrared illumination to acquire images of the detail-rich, intricate structures of the iris which are visible externally. Digital templates encoded from these patterns by mathematical and statistical algorithms allow the identification of an individual or someone pretending to be that individual. Databases of enrolled templates are searched by matcher engines at speeds measured in the millions of templates per second per (single-core) CPU, and with remarkably low false match rates.
At least 1.5 billion people around the world (including 1.29 billion citizens of India, in the UIDAI / Aadhaar programme as of December 2022) have been enrolled in iris recognition systems for national ID, e-government services, benefits distribution, security, and convenience purposes such as passport-free automated border-crossings. A key advantage of iris recognition, besides its speed of matching and its extreme resistance to false matches, is the stabilityP. Grother, J. Matey, E. Tabassi, G. Quinn, and M. Chumakov, IREX VI: temporal stability of iris recognition accuracy, NIST Interagency Report 7948, pp. 1–3, 2013 of the iris as an internal and protected, yet externally visible organ of the eye.
In 2023, Pakistan's National Database & Registration Authority (NADRA) has launched IRIS for citizen registration/ Civic Management during registration at its offices for the National ID Card. After its initial stage, the eye-recognition verification access will be available for LEAs, banking sectors, etc.
History Although John Daugman developed and in the 1990s patented the first actual algorithms to perform iris recognition, published the first papers about it and gave the first live demonstrations, the concept behind this invention has a much longer history and today it benefits from many other active scientific contributors. In a 1953 clinical textbook, F.H. AdlerAdler, F.H., Physiology of the Eye (Chapter VI, page 143), Mosby (1953) wrote: "In fact, the markings of the iris are so distinctive that it has been proposed to use photographs as a means of identification, instead of fingerprints." Adler referred to comments by the British ophthalmologist J.H. Doggart,Doggart, J.H., Ocular Signs in Slit-Lamp Microscopy, Kimpton (1949), page 27 who in 1949 had written that: "Just as every human being has different fingerprints, so does the minute architecture of the iris exhibit variations in every subject examined. [Its features] represent a series of variable factors whose conceivable permutations and combinations are almost infinite." Later in the 1980s, two American ophthalmologists, L. Flom and Aran Safir managed to patent Adler's and Doggart's conjecture that the iris could serve as a human identifier, but they had no actual algorithm or implementation to perform it and so their patent remained conjecture. The roots of this conjecture stretch back even further: in 1892 the Frenchman A. Bertillon had documented nuances in "Tableau de l'iris humain". Divination of all sorts of things based on iris patterns goes back to ancient Egypt, to Chaldea in Babylonia, and to ancient Greece, as documented in stone inscriptions, painted ceramic artefacts, and the writings of Hippocrates. (Iris divination persists today, as "iridology.")
The core theoretical idea in Daugman's algorithms is that the failure of a test of statistical independence can be a very strong basis for pattern recognition, if there is sufficiently high entropy (enough degrees-of-freedom of random variation) among samples from different classes. In 1994 he patented this basis for iris recognition and its underlying computer vision algorithms for image processing, feature extraction, and matching, and published them in a paper.Daugman, J., "High confidence visual recognition of persons by a test of statistical independence", IEEE Transactions on Pattern Analysis and Machine Intelligence, 15 (11), pp 1148-1161 (1993) These algorithms became widely licensed through a series of companies: IriScan (a start-up founded by Flom, Safir, and Daugman), Iridian, Sarnoff, Sensar, LG-Iris, Panasonic, Oki, BI2, IrisGuard, Unisys, Sagem, Enschede, Securimetrics and L-1, now owned by French company Morpho.
Most flagship deployments of these algorithms have been at airports, in lieu of passport presentation, and for security screening using watch-lists. In the early years of this century, major deployments began at Amsterdam's Schiphol Airport and at ten UK airport terminals allowing frequent travellers to present their iris instead of their passport, in a programme called IRIS: Iris Recognition Immigration System. Similar systems exist along the US / Canada border, and many others. In the United Arab Emirates, all 32 air, land, and seaports deploy these algorithms to screen all persons entering the UAE requiring a visa. Because a large watch-list compiled among GCC States is exhaustively searched each time, the number of iris cross-comparisons climbed to 62 trillion in 10 years. The Government of India has enrolled the iris codes (as well as fingerprints) of more than 1.2 billion citizens in the UIDAI (Unique Identification Authority of India) programme for national ID and fraud prevention in entitlements distribution. In a different type of application, iris is one of three biometric identification technologies internationally standardised since 2006 by ICAO for use in e-passports (the other two are fingerprint and face recognition).
Visible vs near infrared imaging Iris melanin, also known as chromophore, mainly consists of two distinct heterogeneous macromolecules, called eumelanin (brown–black) and pheomelanin (yellow–reddish), whose absorbance at longer wavelengths in the NIR spectrum is negligible. At shorter wavelengths within the VW spectrum, however, these chromophores are excited and can yield rich patterns. Hosseini, et al. provide a comparison between these two imaging modalities. An alternative feature extraction method to encode VW iris images was also introduced, which may offer an alternative approach for multi-modal biometric systems.
Visible wavelength iris image Near infrared (NIR) version NIR imaging extracts structure Visible light reveals rich pigmentation details of an Iris by exciting melanin, the main colouring component in the iris. Pigmentation of the iris is invisible at longer wavelengths in the NIR spectrum. Even "dark brown" eyes reveal rich iris texture in the NIR band, and most corneal specular reflections can be blocked.
Operating principle
First the system 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, 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 templates 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.
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) 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 (and the left and right eyes of the same individual) 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's IrisCode, 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)."Probing the uniqueness and randomness of IrisCodes: Results from 200 billion iris pair comparisons." Proceedings of the IEEE, vol. 94 (11), 2006, pp. 1927-1935. 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.
Iris recognition works with clear contact lenses, eyeglasses, and non-mirrored sunglasses. The early Sensar technology worked by first finding the face, then the eyes, and then took the Iris images. This was all done using infrared lighting. It is possible to identify someone uniquely in a dark room while they were wearing sunglasses.