Open AccessLDA BASED FACE RECOGNITION USING DCT AND HYBRID DWT
Author(s) -
J. Lakshmi Tirupathamma,
M. Koteswara Rao,
K. Veera Swamy
Publication year - 2012
Publication title -
international journal of electronic signal and systems
Language(s) - English
Resource type - Journals
ISSN - 2231-5969
DOI - 10.47893/ijess.2012.1057
Subject(s) - discrete cosine transform , artificial intelligence , computer science , pattern recognition (psychology) , linear discriminant analysis , facial recognition system , face (sociological concept) , matlab , feature extraction , discrete wavelet transform , similarity (geometry) , feature (linguistics) , similarity measure , computer vision , image (mathematics) , wavelet , wavelet transform , social science , linguistics , philosophy , sociology , operating system
In this paper we present a hybrid approach for efficient human face recognition. The proposed method is based on linear discriminant analysis of image in DCT domain with a combination of details of DWT. And the similarity measure Minkowshi is used here. This approach reduces the storage requirement and computation time while preserving the data. The approach LDA -DCT-hybrid DWT is evaluated on Matlab using ORL face database. Compared to previous methods the proposed method improves feature extraction and retrieval rate.