Open AccessTranslation Based Face Recognition Using Fusion of LL and SV Coefficients
Author(s) -
B. Sujatha,
B.V. Venukumar,
Chetan Tippanna Madiwalar,
N.C. Abidali Munna,
K. Suresh Babu,
K B Raja,
K R Venugopal
Publication year - 2016
Publication title -
procedia computer science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.334
H-Index - 76
ISSN - 1877-0509
DOI - 10.1016/j.procs.2016.06.077
Subject(s) - computer science , artificial intelligence , biometrics , euclidean distance , face (sociological concept) , facial recognition system , pattern recognition (psychology) , translation (biology) , support vector machine , gaussian filter , sample (material) , gaussian , filter (signal processing) , computer vision , image (mathematics) , biochemistry , chemistry , messenger rna , gene , social science , physics , chromatography , quantum mechanics , sociology
The face is a physiological trait used to identify a person effectively for various biometric applications. In this paper we propose Translation based Face Recognition using Fusion of LL and SV coefficients. The novel concept of translating many sample images of a single person into one sample per person is introduced. The face database images are preprocessed using Gaussian filter and DWT to generate LL coefficients. The support vectors (SV) are obtained from support vector machine (SVM) for LL coefficients.\udThe LL and SVs are fused using arithmetic addition to generate final features. The face database and test face image features are\udcompared using Euclidean Distance (ED) to compute the performance parameters.\ud
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom