A Comparative Dimensionality Reduction Approach For Face Recognitionunder Uncontrolled Illumination Variations
Neenu Prasad

Neenu Prasad, Information Technology, M.G University,Thiruvalla, India.
Manuscript received on May 11, 2013. | Revised Manuscript received on May 15, 2013. | Manuscript published on May 25, 2013. | PP: 50-53 | Volume-1 Issue-7, May 2013. | Retrieval Number: G0316051713/2013©BEIESP

Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Most of the method developed for face recognition with face images collected under relatively well controlled conditions have difficulty in dealing with the range of appearance variations that commonly occur in unconstrained natural images due to different factors. A simple, efficient image processing chain is used whose practical recognition performance is comparable to or better than current methods, a rich descriptor for local texture called Local Binary Pattern(LBP), is used for feature extraction, dimensionality reduction is performed using Principal Component Analysis(PCA) and Linear Discriminant Analysis (LDA). To demonstrate the effectiveness of the proposed method, we give results on the Extended Yale-B dataset which contains images of 38 subjects under 64 illumination and 9 poses.
Keywords: Face recognition, Feature extraction, illumination normalization, linear discriminant analysis, local binary patterns, principal component analysis.