Study: A new biometric database based on corneal topography

A couple of us had posted awhile ago about a Canadian scholar doing some work and presenting a paper on Corneal Topography as a biometric trait for ID mid this yr. I don't recall any of use getting full access to the paper unfortunately.

However, just managed to find the abstract from the paper, as I believe the presentation was in the last several weeks, so may find more on it as it gets published.

The other info in here is to remember where this started in regards to Jim Schwiegerling of Uni of Arizona and his first assessment of the tech - interestingly, not only was he engaged by SM as we know, but he was given a $453k grant between 2008 & 2012 by the National Science Foundation (NSF) Institute - Division of Information and Intelligent Systems (IIS) for the project.

Is just something to think about whilst we await our results as just underlines in my eyes (no pun intended) how valid the tech is, and Nassima offers quite a positive view in the abstract (nice last line in the abstract) - like to read the rest of the paper

Nice eve all.

http://grantome.com/grant/NSF/IIS-0813650

Jim Schwiegerling:

Preliminary studies indicate that corneal shape is a highly unique identifier that is well suited for biometry. In this application, the team will develop on this foundation to a more sophisticated implementation of the technology. This will illustrate its scalability and broad applications. Specifically, this application transitions the corneal geometry comparison algorithms from a simple correlation of shapes to a sophisticated description of unique and readily comparable corneal features. This new technique will allow a compact description of corneal shape, allowing scalable comparisons to large databases of individuals. Furthermore, this application proposes to develop an automated corneal topography measurement system, thus removing the requirement of a trained operator. The new device will be validated against a commercial operator-driven topographer. Finally, the possibility of extending the corneal biometry concept to long-range applications is explored. The culmination of these efforts will demonstrate a flexible biometric system that can be deployed in a variety of security environments. The impact of the proposed technology is an expansion of available technologies for authenticating individuals. All biometric measurement systems have conditions under which they cannot reliably measure an individual. Expanding the available technologies for performing biometry allows security personnel to tailor the technology to the specific environment and application.

Project Report

August 1, 2012 Jim Schwiegerling, PhD The term biometrics is used to describe the process of analyzing biological and behavioral traits that are unique to an individual in order to confirm or determine his or her identity. Many biometric modalities are currently being researched and implemented including, fingerprints, hand and facial geometry, iris recognition, vein structure recognition, gait and voice recognition. This project explored the possibility of using corneal topography measurements as a trait for biometric identification. Two new corneal topographers were developed and fabricated for this study. The first was designed to function as an operator-free device that will allow a user to approach the device and have his or her corneal topography measured. Human subject topography data were collected with this device and compared to measurements made with a commercially available topographer. The other topographer we fabricated allows for arbitrary pattern illumination through the use of LCD monitors. This topographer was built and tested to be used in future research studies. Topography data was collected from 59 subjects and mathematically fit to create a database for biometric identification. The data were analyzed to determine the biometric error rates associated with corneal topography measurements. Resultant matches between the devices showed between three to eight percent simultaneous false match and false non-match rates, were achieved. The study demonstrated that corneal topography is a viable means for identifying individuals and future work will involve refining the acquisition systems.

http://biometrics-biostatistics.conferenceseries.com/eposter-presentation.php#

Nassima Kihal

Abstract
Biometrics offers a natural and reliable solution to certain aspects of identity management with fully automated or semi-automated schemes to recognize individuals based on their inherent physical and/or behavioral characteristics including fingerprint, face, voice, iris, signature, hand and ear. Our goal is to use cornea as a new biometric trait for identity authentication by modeling it in 3D geometry. For this reason, we propose to start with the realization of a new corneal biometric database. The acquisition of the cornea is carried out by a Pentacam Topographer (Oculus). The Pentacam measurement process takes less than two seconds and minute eye movements are captured and corrected simultaneously. By measuring 25,000 true elevation points, precise representation, repeatability and analysis are guaranteed. These data points are then used to generate corneal maps used for diagnosis and treatment. Our database contains 344 corneal topographies captured from 43 different people of different ages using a both eyes. For each eye, we captured two sessions of corneal topography. The time interval between the two sessions was equal or greater than one month. In each session, 8 corneal topographies (4 left eyes and 4 right eyes) were captured. These multiple measurements are useful to assess repeatability, performance, collectability, etc. The corneal shape was recorded as a uniformly spaced (X-Y) grid (image) of raw elevations (Z). This elevation can be represented with an appropriate mathematical model such as a Zernike polynomial expansion. The model parameters (e.g., Zernike coefficients) can then be post-processed (e.g., with linear discriminant analysis LDA)) to become a ‘signature’ or ‘template’ of the cornea. Our proposed database was tested for person recognition in a preliminary study and gave an impressive equal error rate of less than 1%.