A New Approach to Heart Sounds Biometric Recognition Based on Gram-PNN

Date issued

2013

Journal Title

Journal ISSN

Volume Title

Publisher

University of West Bohemia

Abstract

In this paper we introduce a new approach to heart sounds biometric recognition based on Gram polynomials and probabilistic neural networks (PNN). The usage of heart sounds as physiological biometric traits was first introduced in [1], in which the authors proposed and started exploring this idea. Heart sound recognition is based on the analysis of PCG (PhonoCardioGram) sequences. The proposed system presents good performance obtaining an error rate of 13.70 % over a database of 50 people, containing multiple heart sequences per person, each lasting from 20 to 70 seconds.

Description

Subject(s)

biometrie, rozpoznávání srdečních ozev, pravděpodobnostní neuronové sítě, fonokardiogram

Citation

ISTET 2013: International Symposiumon Theoretical Electrical Engineering: 24th – 26th June 2013: Pilsen, Czech Republic, p. III-5-III-6.
OPEN License Selector