Employing Vector Quantization Algorithm In A Transform Domain For Facial Recognition

Abstract

A Vector Quantization (VQ) algorithm in the Discrete Cosine Transform (DCT) domain is proposed for facial recognition. There are three main phases in the proposed system, namely, Preprocessing, Feature Extraction, and Recognition. Cropping and choosing an appropriate dimension are performed in the preprocessing step. Then, DCT with appropriate truncation dimensions is applied to the processed faces for dimensionality re- duction. For further feature compaction, VQ algorithm employing Kekre Fast Codebook Generation (KFCG) approach for codebook initialization is applied to the transformed truncated features. Finally, the proposed system is extensively evaluated using four different databases, namely, ORL, YALE, FERET, and FEI that have different facial variations, such as illuminations, rotations, facial expressions, etc. Euclidean distance criterion is used to calculate the recognition rates. Then, the results are analyzed using K-fold Cross Validation (CV). The proposed approach is shown to improve the recognition rates as well as the storage requirements in comparison with some of the existing state-of-The arts approaches.

Publication Date

7-2-2016

Publication Title

Midwest Symposium on Circuits and Systems

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

DOI Link

https://doi.org/10.1109/MWSCAS.2016.7869957

Socpus ID

85015974161 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85015974161

This document is currently not available here.

Share

COinS