Face Recognition under Complex Illumination Based on Multi-scale Weberface and Local Binary Pattern

摘要

A novel method based on Multi-scale Weberface (MWF) and Local Binary Pattern (LBP) is proposed to reduce illumination effects for face recognition in this paper. The method starts with pre-processing face image by utilizing contrast stretching transformation that reduces the effect of different lighting to some extent. Then different scale Weber faces are extracted to calculate MWF based on weighted method as multi-scale illumination insensitive face. Finally the study abstracts LBP feature from MWF as face recognition feature under complex illumination and employs multi-class Support Vector Machine (SVM) for face authentication. The experiments are executed both on CMU-PIE and Yale B face databases. The experimental results have indicated that the proposed method can effectively eliminate the influence of face recognition under complex illumination and the recognition rates are superior to Weberface (WF), even single sample images with serious light as training sample images can also work well. Furthermore, this study also has shown that our method is more robust to noise and more suitable to the actual system for face authentication.