Dual Factor Biometric Authentication System Using GPCA and IFB

By:
Ruba Soundar Kathavarayan,
Ms Gul Sahira Banu Jahangeer,
Rega Paramasivam
To add a paper, Login.

In this paper we propose a novel dual factor biometric authentication system using Gabor-based Principal Component Analysis (GPCA) and Invariant Filter Bank (IFB). The first biometric factor is face which is authenticated as follows: Gabor wavelets first derive desirable facial features characterized by spatial frequency, spatial locality, and orientation selectivity to cope with the variations due to illumination and facial expression changes. PCA is applied on the Gabor features for expressing the data in such a way as to highlight their similarities and differences. Finally, Euclidian distance is found out to select one of the best matching training image. The second biometric factor is fingerprint, which is authenticated using a new filter bank-based fingerprint feature extraction and matching method without needing to detect minutiae. The method decomposes a fingerprint image into eight directional sub band outputs using a invariant filter bank (IFB) and then obtains directional energy distributions for each block from the decomposed sub band outputs. Only dominant directional energy components are employed as elements of the input feature vector, which serves to reduce noise and improve efficiency. For the rotational alignment, additional input feature vectors in which various rotations are considered. This proposed method is tested with ORL face database, which contains 400 images comprising 40 different subjects of 10 different variations and NIST fingerprint database with 40 images being matched with those 40 subjects in the face database.


Keywords: Gabor Wavelet, Principal Component Analysis, Spatial Frequency, Invariant Features, Biometrics, Directional Energy, Directional Filter Bank (DFB), Feature Extraction, Fingerprints, Matching, Verification
Stream: Technology in Community
Presentation Type: Paper Presentation in English
Paper: Dual Factor Biometric Authentication System Using GPCA and IFB


Ruba Soundar Kathavarayan

Lecturer, Computer Science and Engineering Department, Mepco Schlenk Engineering College
India

Having lot of research experience in Image Processing, Face Segmentation and Extraction in still and video images.

Ms Gul Sahira Banu Jahangeer

ME Student, Computer Science and Department, Mepco Schlenk Engineering College
India

Having lot of research experience in Image Processing, Face Segmentation and Extraction in still and video images.

Rega Paramasivam

ME Student, Computer Science and Engineering Department, Mepco Schlenk Engineering College
India

Having lot of experience in Signal and Image Processing.

Ref: T06P0279