Publication Type : Conference Paper
Publisher : IEEE International Conference on Pattern Recognition, Informatics and Medical Engineering
Source : IEEE International Conference on Pattern Recognition, Informatics and Medical Engineering(PRIME 2012), Salem, Tamilnadu, p.387-392 (2012)
ISBN : 9781467310376
Accession Number : 12770798
Keywords : Algorithms, biomedical engineering, Biometrics, Error rate, Feature extraction, GMM, Information science, Likelihood ratio tests, Multi-modal biometrics, Score-level fusion, Template, Unimodal
Campus : Coimbatore
School : School of Engineering
Department : Electronics and Communication
Verified : Yes
Year : 2012
Abstract : Biometrics consists of methods for uniquely recognizing humans based upon one or more intrinsic physical or behavioral traits. This paper describes the feature extraction techniques for three modalities viz. fingerprint, iris and face. The extracted information from each modality is stored as a template. The information are fused at the match score level using a density based score level fusion, GMM followed by the Likelihood ratio test. GMM parameters are estimated from training data using the iterative Expectation-Maximization (EM) algorithm. © 2012 IEEE.
Cite this Research Publication : S. A. Vivek, J. Aravinth and S. Valarmathy, "Feature extraction for multimodal biometric and study of fusion using Gaussian mixture model," International Conference on Pattern Recognition, Informatics and Medical Engineering (PRIME-2012), Salem, India, 2012, pp. 387-392, doi: 10.1109/ICPRIME.2012.6208377.