3D Face Recognition Using Weiner Filter and DFT Based On Optimized Directional Faces
Shilpa S. Nair1, Naveen S.2, Moni R.S3
1Shilpa S. Nair, Department of Electronics and Communication, LBS Institute of Technology for women, Thiruvananthapuram (Kerala), India.
2Naveen S., Department of Electronics and Communication, Kerala LBS Institute of Technology for women, Thiruvananthapuram (Kerala), India.
3Dr R.S Moni, Professor, Department of ECE, Marian Engineering College, Thiruvananthapuram (Kerala), India.
Manuscript received on 15 August 2015 | Revised Manuscript received on 25 August 2015 | Manuscript Published on 30 August 2015 | PP: 64-70 | Volume-4 Issue-6, August 2015 | Retrieval Number: F4163084615/15©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. 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: Traditional 2D face recognition methods based on intensity or color images, face challenges in dealing with pose variations or illumination changes. The face recognition based on combination of 3D shape information and 2D intensity/color information is a novel approach, which provides an opportunity to improve the face recognition performance. This paper proposes an efficient multimodal face recognition method by combining the textural as well as depth features, extracted from directional faces of input image. To overcome problems occurred due to low quality image, pre-processing is done before extracting features from the image. The directional faces captured using Local Polynomial Approximation (LPA) filters are adaptively optimized. The modified LBP (mLBP) is used for the feature extraction from these directional faces. The spectral transformation of the concatenated block histogram of mLBP feature image acts as the robust face descriptor. Discrete Fourier Transform (DFT) is used as the transformation tool. The fusion of both modalities is performed at score level. The experimental results shows that the proposed method gives better performance than single modality.
Keywords: DFT, MLBP, Multimodal, ODF, Weiner Filter.
Scope of the Article: Pattern Recognition