DWT-ELBP based model for face recognition

Abstract

The extraction of exact facial landmarks plays a significant role in a face recognition (FR) system. This paper presents an approach for recognition of both frontal and non frontal faces using symmetric Elliptical Local Binary Pattern (ELBP) and Discrete Wavelet Transform (DWT) under different illumination conditions and facial expressions. The facial features are extracted using ELBP and further features are decomposed by DWT. For classification K-NN classifier is used here. The approach implemented on Essex, AR and color FERET face database respectively signifies 99.8%, 98.4% and 96.3% rate of recognition rate.