Robust face recognition in a polar coordinate system using LDAr
Jae Hyun Oh, Nojun Kwak · 2010
In this paper, we propose a novel method for face recognition which uses a polar coordinate system instead of the conventional cartesian coordinate system. The polar coordinate system delineates near-pole area more vividly than the area far from the pole. Therefore, a polar face image can achieve more vivid representation of the important central facial regions compared to the conventional cartesian face image. To cope with a small amount of rotation and scaling change in a frontal face image, we applied a method based on regressional version of linear discriminant analysis (LDAr). The experimental results show that the proposed method achieves a good recognition performance under varying rotation and scaling.