Abstract:Illumination processing is a challenging task in face recognition. Although numerous techniques have been proposed to tackle this problem, none of them can achieve high performance under different illumination variations. In this paper, we propose three closely related illumination processing techniques to address different illumination variations. First, we employ singular value decomposition (SVD) bases with frequency interpretation and corresponding contribution coefficients (i.e. normalized nonlinear singular values) to obtain the high-frequency facial features of a face image in logarithm domain, which is termed as the high-frequency SVD face (HFSVD-face). Moreover, we propose the adaptive HFSVD-face (AHFSVD-face), which adaptively selects a special nonlinear parameter to generate HFSVD-face according to the face image illumination level. The illumination level of a face image is estimated by its singular values. Second, we construct the local near neighbor feature (LNN-feature) based on the local region, which is an illumination invariant measure. Then the discriminative weights (DWs) and Gaussian weights (GWs) respectively combine with several LNN-features to generate the fusion LNN-feature (FLNN-feature), which can be named as DWFLNN-feature and GWLNN-feature. The bipolar sigmoidal function removes the high-frequency interference of FLNN-feature to form LNN-face (including DWLNN-face and GWLNN-face). Third, we develop the H & L_SVD as a preprocessing technique, which consists of the normalized low-frequency components and the corrected high-frequency components divided by SVD. And then the illumination insensitive measure can be extracted from H & L_SVD by LNN-face to robustly tackle the face image with severe illumination variations. The performances of the proposed methods are verified using Yale B, CMU PIE, LFW and our self-built driver face databases. The experimental results indicate that the proposed methods can obtain high performances under different illumination variations and outperform several state-of-the-art approaches which are proposed to address face recognition under varying illumination.

Illumination invariant face recognition using nonlocal total variation in logarithmic domain

Adaptive Illumination-Invariant Face Recognition Via Local Nonlinear Multi-Layer Contrast Feature.

A novel illumination normalization method in face recognition based on logarithmic total variation

Adaptive Illumination Normalization Via Adaptive Illumination Preprocessing and Modified Weber-Face

Face Recognition under Varying Illumination Based on Gradientface and Local Features

Illumination Invariant Face Recognition by Local Image Descriptor in Logarithm Domain

Single-sample Face Recognition under Varying Lighting Conditions Based on Logarithmic Total Variation.

Face illumination normalization on large and small scale features

A Fusion Framework For Face Recognition Under Varying Illumination Based On Multi-Scale Analysis

Spatially Adaptive Logarithmic Total Variation Model for Varying Light Face Recognition

Illumination Invariant Face Recognition Based on Improved Local Binary Pattern

Illumination Invariant Extracting Algorithm Based on Nonsubsampled Contourlet Transform

Illumination normalization based on correction of large-scale components for face recognition.

Joint Features for Face Recognition under Variable Illuminations

Singular Value Decomposition and Local Near Neighbors for Face Recognition under Varying Illumination.

An Illumination Processing Algorithm Based on Total Variation Model

Face Recognition under Varying Illumination Based on Singular Value Decomposition and Retina Modeling

Extraction of illumination invariant facial features from a single image using nonsubsampled contourlet transform

An efficient illumination normalization method for face recognition

Illumination Normalization via Merging Locally Enhanced Textures for Robust Face Recognition