Multimodal Biometric Authentication System

Trinity Journal of Management, IT & Media (TJMITM)
Year: 2012 (Jan-Dec), Volume: (3), Issue. (1)
First page: (19) Last page: (22)
Online ISSN: A/F
Print ISSN: 2320-6470

Multimodal Biometric Authentication System
Rajesh Kumar Jain

1HoD, Comp. Sc. Deptt., Sirifort College of Computer Technology & Mgt, Rohini
Corresponding author email id:

16 -04-2012


Online Published:

How to cite the Article

This image has an empty alt attribute; its file name is images.png
View on Google Scholar


A wide variety of applications require reliable and secure  verification schemes to confirm the identity of an individual  requesting authorized accessing to a specified service.  Examples of such applications include secure access to  buildings, personal computer systems, laptops, cellular  phones and ATMs. In the absence of robust verification  schemes, these systems are subject to the tricks of an  impostor. Biometric authentication (BA) is a problem of  verifying an identity claim using a person’s behavioral and  physiological characteristics. Biometric identity authentication is based on a binary  classification problem: reject or accept identity claim. The  basis of some matching criteria. Several verification  systems have been developed based on different  biometrics characteristics (fingerprint, face, speech, iris  etc) which help to distinguish individuals from each other.  Each biometric has its own advantages and drawbacks  due to its discriminative power, complexity, robustness  involved. Research in last few years has shown that no  single biometric system can achieve 100% authentication  accuracy. This problem can be alleviated by combining two  or more biometric modalities [1], also known as the field of  multimodal biometric authentication. Auser authentication  scenario involving two modalities (Face and Speech) [2] is  figured out in Fig. 1.  performance and robustness of identity authentication  systems can be improved by combining two or more  different modalities (speech, face, fingerprint, etc.).


Binary classifiers, biometrics, classifier fusion, feature  extraction, support vector machine, ROC.