In this paper, we address some issues pertaining to adaptive multimodal biometric authentication. These issues include new user registration, sensor decay and small sample data size. A recursive formulation is introduced to track changes due to new user registration and possible sensor decay. The small sample size problem is handled using a feature scaling-space learning technique. Empirical experiments are conducted to observe the effects.
|Number of pages||12|
|Journal||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Publication status||Published - 2004 Dec 1|
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)