| Paper: | DISPS-P2.3 | ||
| Session: | Hardware and Software Implementations of DSP Systems | ||
| Time: | Thursday, May 18, 16:30 - 18:30 | ||
| Presentation: | Poster | ||
| Topic: | Design and Implementation of Signal Processing Systems: Algorithm Transformation and Mapping Techniques | ||
| Title: | Human Iris Identification via Low-Complexity Circular Periodicity Transform | ||
| Authors: | Lei Wang, University of Connecticut, United States | ||
| Abstract: | In this paper, we propose a low-complexity circular periodicity transform (CPT) for human iris identification. The proposed CPT enhances the standard periodicity transform by enabling the extraction of a wide spectrum of periodic patterns hidden in human iris. A computation-efficient architecture is developed for accurate iris identification with minimal hardware cost. The CPT-based iris identification system achieves up to 99.2% recognition rates with 21.8%-56.2% reduction in computation complexity as compared with the standard periodicity transform. The proposed CPT in combination with architecture optimizations address the challenges in single-chip implementation of biometric systems. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012