Adaptive Weber-face for robust illumination face recognition

作者：Yang, Chao^*; Wu, Shiqian; Fang, Hongping; Er, Meng Joo

来源：Computing, 2019, 101(6): 605-619.

DOI：10.1007/s00607-019-00706-7

• Face recognition
• Illumination normalization
• Adaptive Weber-face
• Gamma correct
• Local binary patterns

In this paper, an adaptive Weber-face method is presented to deal with varying lighting and extract illumination-insensitive representation for face recognition. A novel pixel-wise adaptive gamma transformation, which has been proved mathematically, is proposed to achieve a local uniform-illumination assumption in Weber-face. Next, the local binary patterns are extracted from the Weber-face images, and then further alleviate the effect of varying illumination. Experiments demonstrate that the proposed method achieves the high recognition rates of 99.55% and 96.63% on extended Yale B and CMU-PIE databases respectively, which outperforms the state-of-the-art methods.

• 出版日期2019-6
• 单位武汉科技大学; 南阳理工学院