Novel Texture Description and Face Identification Methods by Defining Bridle Paths and Using Gabor Phases

Authors

1 Department of Computer Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran

2 Department of Computer Engineering, Sharif University of Technology, Tehran, Iran

3 Faculty of Engineering, Qaemshahr Branch, Islamic Azad University, Qaemshahr, Iran

Abstract

Identification based on faces is still a useful method for many applications and face recognition developing is an active research Field. In this paper, a novel face identification method is proposed. The proposed method (Bridle Path on Gabor Phase (BPGP)) is based on extracting texture patterns from phases of the Gabor wavelet. Also, in order to describe the textures, a novel texture descriptor method (Bridle Path) is proposed to extract the features from textures. The Bridle Path method inspired by LBP method and brings some advantages such as lower feature vector length and higher texture description capability in comparison with LBP. Experimental results shows that, Bridle Path texture descriptor is a powerful tool for describing textures and consequently proposed face recognition method (BPGP) yields high identification performance compared to other methods.

Keywords