Face Recognition based on Logarithmic Fusion of SVD and KT
Ramachandra A C1, Raja K B2, Venugopal K R3, L M Patnaik4

1Ramachandra A C, Department of Electronics and Communication, Alpha College of Engineering, Bangalore, India.
2K B Raja, Department of Electronics and Communication, University Visveswaraya College of Engineering, Bangalore, Karnataka, India.
3Venugopal K R, Principal, University Visveswaraya College of Engineering, Bangalore, India.
4L M Patnaik, Honorary professor, Indian Institute of Science, Bangalore, India.

Manuscript received on July 01, 2012. | Revised Manuscript received on July 04, 2012. | Manuscript published on July 05, 2012. | PP: 508-116 | Volume-2, Issue-3, July 2012. | Retrieval Number: C0832062312/2012©BEIESP
Open Access | Ethics and Policies | Cite 
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: The identification of a person based on biometric is accurate and robust compared to traditional methods of identifying a person using PIN, ID cards etc., In this paper Face Recognition based on Logarithmic Fusion of SVD and KT (FRLSK) is proposed. The Singular Value Decomposition (SVD) is applied on face images to derive Co-efficients. The Co-efficient Matrix of SVD are resized to 64×64 to form features. The test image SVD features are compared with SVD feature of database images using Euclidian distance, Equal Error Rate (EER) and Total Success Rate are computed (TSR). The Kekre Transform (KT) is applied on Resized (64×64) face images to form features. The test image KT Features are compared with KT features of Database images using Euclidian distance to compute EER and TSR. The EER and TSR values obtained by SVD techniques are fused with the value of EER and TSR obtained from KT using logarithmic transforms to get better value of EER and TSR. It is observed that the value of EER and TSR are better in the case of proposed algorithm compared to existing algorithm.

Keywords: Biometrics, SVD, KT, Total Success Rate.