Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | DISPS-P3.12 |
| Session: | High Performance DSP Architectures and Systems |
| Time: | Friday, May 21, 13:00 - 15:00 |
| Presentation: |
Poster |
| Topic: |
Design and Implementation of Signal Processing Systems: Hardware, Software, and Algorithm Tradeoffs and Integration |
| Title: |
A REALTIME, MEMORY EFFICIENT FINGERPRINT VERIFICATION SYSTEM |
| Authors: |
Shenglin Yang; University of California, Los Angeles | | |
| | Ingrid Verbauwhede; University of California, Los Angeles | | |
| Abstract: |
Creating a biometric verification system in an energy and area constrained embedded environment is a challenging problem. Our paper describes a secure and efficient embedded fingerprint verification system part of the “ThumbPod” embedded device, in which a complete real-time fingerprint recognition module, including both minutiae extraction and matching, works on a 50MHz LEON-2 processor. As the result of SW/HW speed and memory optimizations, we achieve 65% reduction on execution time and 67% reduction on memory size against a traditional implementation reference. |
| |
| Back | |