Face Image Hallucination Under Illumination Variation
Image resolution and variable illumination conditions in the real word have significant effects on face recognition. In this paper, a new face image hallucination method under illumination variation is proposed. In order to hallucinate a high-resolution face image with frontal illumination from a low-resolution face image with non-frontal illumination, in the proposed method, the nonsubsampled contourlet transform is first used to estimate the reflectance component and features affected by illumination variation from an input low-resolution face image with non-frontal illumination, then a super-resolution approach via adaptively local sparse representation and global reconstruction constraint is presented to enhance the resolution of the reflectance component for the input low-resolution face image, and the low-resolution features affected by the illumination variation are normalized and enhanced by the Kernel Principal Component Analysis and bicubic interpolation techniques. Finally, the high-resolution face image with frontal illumination is hallucinated by combining the reconstructed high-resolution reflectance component and the enhanced illumination features. Experimental results on Extended Yale B face database show that under illumination variations, the proposed method can deal with uneven illuminations in face image and reconstruct the high-resolution face image with frontal illumination effectively compared with some other super-resolution methods.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
Document Type: Research Article
Publication date: November 1, 2014
More about this publication?
- Journal of Computational and Theoretical Nanoscience is an international peer-reviewed journal with a wide-ranging coverage, consolidates research activities in all aspects of computational and theoretical nanoscience into a single reference source. This journal offers scientists and engineers peer-reviewed research papers in all aspects of computational and theoretical nanoscience and nanotechnology in chemistry, physics, materials science, engineering and biology to publish original full papers and timely state-of-the-art reviews and short communications encompassing the fundamental and applied research.
- Editorial Board
- Information for Authors
- Submit a Paper
- Subscribe to this Title
- Terms & Conditions
- Ingenta Connect is not responsible for the content or availability of external websites