Color space construction by optimizing luminance and chrominance components for face recognition

作者：

Highlights：

• We propose a framework of constructing effective color spaces for face recognition tasks.

• The luminance component is selected from four candidates by R, G, B coefficient analysis and color sensor analysis.

• Two chrominance components are extracted from the RGB color space by discriminant analysis and correlation analysis.

• The proposed color space consistently performs better than state-of-the-art color spaces on four benchmark databases.

摘要

•We propose a framework of constructing effective color spaces for face recognition tasks.•The luminance component is selected from four candidates by R, G, B coefficient analysis and color sensor analysis.•Two chrominance components are extracted from the RGB color space by discriminant analysis and correlation analysis.•The proposed color space consistently performs better than state-of-the-art color spaces on four benchmark databases.

论文关键词：Color face recognition,Color space,Color sensor analysis,Chrominance subspace,Discriminant analysis,Covariance analysis

论文评审过程：Received 19 June 2017, Revised 2 May 2018, Accepted 19 June 2018, Available online 20 June 2018, Version of Record 27 June 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.06.015