Employing Vector Quantization In A Transform Domain For Face Recognition
Abstract
A face recognition system using an integration of Discrete Cosine Transform (DCT) and Vector quantization (VQ) is proposed in this paper. The system consists of two main phases, namely, Feature Extraction and Recognition. In the first phase, the input facial image is divided into blocks with dimensions equal to the codeword dimensions. Then, DCT is applied on each block. The codebook is initialized using the Kekre Fast Codebook Generation (KFCG) method. The Final Codebook computed using VQ algorithm efficiently represents the input facial image. The second phase aims to find the recognition rates based on the Euclidean distance criterion. The system is evaluated using four different databases, namely, ORL, YALE, FERET, and FEI that have different facial variations, such as facial expressions, illuminations, etc. The experimental results are analyzed using K-fold Cross Validation (CV). The proposed system is shown to improve the storage requirements, as well as the recognition rates.
Publication Date
12-7-2016
Publication Title
2016 IEEE 7th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2016
Document Type
Article; Proceedings Paper
Personal Identifier
scopus
DOI Link
https://doi.org/10.1109/UEMCON.2016.7777823
Copyright Status
Unknown
Socpus ID
85010333000 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/85010333000
STARS Citation
Alobaidi, Taif; Aldhahab, Ahmed; and Mikhael, Wasfy B., "Employing Vector Quantization In A Transform Domain For Face Recognition" (2016). Scopus Export 2015-2019. 3964.
https://stars.library.ucf.edu/scopus2015/3964