Premium
Performance evaluation of face recognition using feature feedback over a number of Fisherfaces
Author(s) -
Jeong GuMin,
Choi SangIl
Publication year - 2013
Publication title -
ieej transactions on electrical and electronic engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.254
H-Index - 30
eISSN - 1931-4981
pISSN - 1931-4973
DOI - 10.1002/tee.21848
Subject(s) - pattern recognition (psychology) , facial recognition system , feature (linguistics) , artificial intelligence , computer science , face (sociological concept) , feature extraction , feature vector , feature recognition , machine learning , social science , philosophy , linguistics , sociology
In this paper, we present a performance evaluation of face recognition using feature feedback over the number of Fisherfaces and propose a guideline to determine an appropriate number of Fisherfaces. Feature‐feedback‐based pattern recognition has been proposed to distinguish the important part from the whole dataset. The selected data by feature feedback is applied to extract features for classification and consequently enhance the performance of the classification. However, some parameters should be adjusted when using the feature feedback for pattern recognition depending on the applications. The number of Fisherfaces is one of the most important parameters that affect the recognition rates. In this paper, we propose a method to determine the number of Fisherfaces when using feature feedback for face recognition. Experimental results show that the proposed method is useful in the effective application of feature feedback for the improvement of the recognition performance. © 2013 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.