Complementary Feature Level Data Fusion for Biometric Authentication Using Neural Networks release_yapykgtatngphcdhkyiw2hh52e

Abstract

Data fusion as a formal research area is referred to as multi‐sensor data fusion. The premise is that combined data from multiple sources can provide more meaningful, accurate and reliable information than that provided by data from a single source. There are many application areas in military and security as well as civilian domains. Multi‐sensor data fusion as applied to biometric authentication is termed multi‐modal biometrics. Though based on similar premises, and having many similarities to formal data fusion, multi‐modal biometrics has some differences in relation to data fusion levels. The objective of the current study was to apply feature level fusion of fingerprint feature and keystroke dynamics data for authentication purposes, utilizing Artificial Neural Networks (ANNs) as a classifier. Data fusion was performed adopting the complementary paradigm, which utilized all processed data from both sources. Experimental results returned a false acceptance rate (FAR) of 0.0 and a worst case false rejection rate (FRR) of 0.0004. This shows a worst case performance that is at least as good as most other research in the field. The experimental results also demonstrated that data fusion gave a better outcome than either fingerprint or keystroke dynamics alone.
In text/plain format

Archived Files and Locations