Iris
Recognition
Iris based personal identification (PI) uses the unique visible characteristics of the human iris (the tinted annular portion of the eye bounded by the black pupil and the white sclera) as it's biometric. Most commercially available iris PI systems are based on research and patents held by Dr. John Daugman of the University of Cambridge, Cambridge, U.K [2]. An iris PI system requires no intimate contact between the user and the image capture device. Typically, a conventional CCD camera is used to capture an image of an eye. Algorithms then isolate and transform the iris portion of the images into templates that offer exceptional matching performance for both FAR and FRR. So much so that iris based PI is one of the few biometric systems with proven "user identification" mode capability for large (national, international, and even planetary) sized template databases [2] [4] [5].
The human iris is composed of elastic connective tissue called the trabecular meshwork. The trabecular meshwork is completely developed by the eighth month of gestation. It consists of a host of visible features namely rings, furrows, freckles, and several others that require a medical degree and/or dictionary for explanation and comprehension. The color of the iris often changes during the first year of life, however, clinical evidence indicates that the trabecular pattern is stable throughout ones lifespan. The iris is immune to the environment except for the pupil's response to light. A remarkable fact about the iris (and one of the reasons that the iris image makes an excellent biometric) is that each possesses a highly detailed and unique visible texture. These textures are unique even when considering genetics. That is to say that, not only do identical twins have unique irises, the two irises of any individual are each unique and have uncorrelated textures [1] [2] [3] [5].
Iris pattern based PI is one of the few methodologies that has been proven to work well in user "identification mode". It exhibits a high degree of accuracy and therefore can be used in applications where high security is paramount. Although user interaction is required for an adequate image capture, the technology requires no physical contact and is basically non-intrusive. Once educated and acclimated, users have regularly accepted the technology for PI applications. If a PI system requires user "identification mode" over large template databases, this technology may be one of only two options (the other is retina scanning) for the PI system developer.
[1] A. Jain, R. Bolle, S. Pankanti, editors, "BIOMETRICS Personal Identification in Networked Society," Kluwer Academic Press, Boston, 1999.[2] International Biometric Group, http://www.biometricgroup.com. Last accessed: 30 July 2001.
[3] J. Daugman, "High Confidence Visual Recognition of Persons By a Test of Statistical Independence," IEEE Transactions on Pattern Analysis and Machine Intellegence, pages 1148-1161, 1993. http://www.labs.bt.com/library/papers/PAMIpaper/PAMIpaper.html. Last accessed: 30 July 2001.
[4] J. Daugman, "How Iris Recognition Works," http://www.cl.cam.ac.uk/~jgd1000. Last accessed: 31 July 2001.
[5] J. Daugman, "Wavelet demodulation codes, statistical independence, and pattern recognition," Institute of Mathematics and its Applications, Proc. 2nd IMA-IP, pp 244 -