While retina recognition is recognized as a highly accurate and difficult to forge biometric, it has not seen widespread acceptance. In addition to user acceptance of what is at times considered an invasive technique, this limited acceptance was caused in part by the relatively high cost of signal acquisition. To alleviate the latter concern, this paper therefore describes a retina recognition algorithm based on Hill's algorithm together with a description of its implementation and experimental evaluation. The algorithm is designed to operate with signals acquired in the visual spectrum, which can be obtained using a number of off-the-shelf camera systems and provides acceptable performance compared to approaches based on near-infrared retinal images.
|Title of host publication
|Proc. 2008 International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIHMSP '08)
|IEEE Computer Society Press
|Published - 15 Aug 2008