Face recognition by using characteristic matrix and ring rotation invariant transform

作者:

Highlights:

摘要

In this research, the characteristic matrix is employed to generate the various discriminators. By employing these newly created discriminators, one can identify the different faces. The “ring rotation invariant transform” technique can cope with the face rotation problem and the “ring rotation invariant transform” technique will also generate the various transform signals. By these newly created transformed-signals, the algorithm can also identify different faces. In this research, the transformed-signals, which are generated by the “ring rotation invariant transform”, will be treated as the input signals and these transformed-signals will be fed to the characteristic matrix to generate the various discriminators. For coping with the image-shifting problem, one pixel in the sample image is compared with the surrounding pixels of the corresponding pixel in the unknown image. The comparison approaching in this research is by the basis of pixel-to-pixel-comparison.

论文关键词:Face image recognition,Characteristic matrix,Orientation invariant transform,Phase-based recognition

论文评审过程:Available online 16 February 2007.

论文官网地址:https://doi.org/10.1016/j.amc.2006.11.162