Sequential decision fusion for controlled detection errors


Autoria(s): Nallagatla, Vishnu P.; Chandran, Vinod
Contribuinte(s)

Nallagatla, Vishnu Priya

Chandran, Vinod

Data(s)

01/07/2010

Resumo

Information fusion in biometrics has received considerable attention. The architecture proposed here is based on the sequential integration of multi-instance and multi-sample fusion schemes. This method is analytically shown to improve the performance and allow a controlled trade-off between false alarms and false rejects when the classifier decisions are statistically independent. Equations developed for detection error rates are experimentally evaluated by considering the proposed architecture for text dependent speaker verification using HMM based digit dependent speaker models. The tuning of parameters, n classifiers and m attempts/samples, is investigated and the resultant detection error trade-off performance is evaluated on individual digits. Results show that performance improvement can be achieved even for weaker classifiers (FRR-19.6%, FAR-16.7%). The architectures investigated apply to speaker verification from spoken digit strings such as credit card numbers in telephone or VOIP or internet based applications.

Formato

application/pdf

Identificador

http://eprints.qut.edu.au/32325/

Publicador

IEEE

Relação

http://eprints.qut.edu.au/32325/1/c32325.pdf

http://www.fusion2010.org/

Nallagatla, Vishnu P. & Chandran, Vinod (2010) Sequential decision fusion for controlled detection errors. In Nallagatla, Vishnu Priya & Chandran, Vinod (Eds.) Proceedings of the 13th International Conference on Information Fusion, IEEE, Edinburgh International Conference Centre, Edinburgh.

Direitos

Copyright 2010 International Society of Information Fusion and the Authors

Fonte

Faculty of Built Environment and Engineering; School of Engineering Systems

Palavras-Chave #080109 Pattern Recognition and Data Mining #090609 Signal Processing #Biometrics #Multi-modal Biometrics #Multi-instance Fusion, #Sequential Decision Fusion #Detection Error Trade-off #Multi-sample Fusion
Tipo

Conference Paper