Speech and Iris based Multimodal Biometric System

  Sakshi Sahore,  Tanvi Sood

Biometrics is the science and technology used for human identification and verification for various applications. Biometric systems can be developed using a single biological attribute is called a unimodal biometric system and the one developed using more than one biological attribute is called multimodal biometric system. While multimodal biometric system suffer limitations in the system performance due to noisy data, interclass variations, non-universality and spoof attacks, multimodal systems are proved to be more accurate and reliable due to multiple sources of information. This paper discusses a multimodal biometric system using speech and iris. The paper presents the fusion of speech and iris modalities at score level. The features of speech are extracted using MFCC while that of iris are extracted using SIFT. GA is used to optimize the features if speech and iris and scores are calculated and fused. Hamming distance is used for matching. The system provides favourable FAR and FRR providing better accuracy.

Keywords- biometrics, unimodal, multimodal, MFCC, SIFT, GA, FAR, FRR

Unique Identification Number - IJEDR1503086
Page Number(s) - 1-7
Pubished in - Volume 3 | Issue 3 | 28 August 2015
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Sakshi Sahore,  Tanvi Sood,   "Speech and Iris based Multimodal Biometric System", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.3, Issue 3, pp.1-7, 28 August 2015, Available at :http://www.ijedr.org/papers/IJEDR1503086.pdf