题目：Advanced Variations of Two-Dimensional Quaternion Principal Component Analysis for Face Recognition
报告人：贾志刚 教授 （江苏师范大学）
时间：2020/10/29 （周四）16:00-17:00
地点：数学学院425会议室
摘要：The two-dimensional quaternion principal component analysis (2DQPCA) has been one of the basic methods of developing artificial intelligent algorithms. To increase the feasibility, we propose a new general ridge regression model for 2DQPCA and variations, with extracting low dimensional features under two projection subspaces. A new relaxed 2DQPCA is proposed to utilize the label (if known) and color information to compute the essential features of generalization ability with optimization methods. The 2DQPCA-based approaches for face recognition are also improved by weighting each principle component a scatter measure, which increases efficiently the rate of face recognition. We also develop new structure-preserving algorithms to solve the corresponding quaternion eigenvalue problems and quaternion optimization problems. In numerical experiments on well-known standard databases, the relaxed 2DQPCA approach has high generalization ability and achieves a higher recognition rate than the state-of-the-art 2DPCA-like methods.
