A Novel Feature Fusion Method Based on Two-Dimensional PCA
A novel feature fusion method, based on 2DPCA is proposed according to the idea of canonical correlation. As a method of comparing sets of vectors or matrices, canonical correlations offer many benefits in accuracy, efficiency and robustness compared to the classical methods. In the paper, based on 2DPCA, and its variants, the alternative 2DPCA, the proposed method can integrate the features extracted from 2DPCA and the alternative 2DPCA. Experimental results show that the proposed method significantly outperforms the popular methods.
Feature fusion Canonical Correlation 2DPCA Face recognition
Xu Zhang Jian Cao Yushu Liu Xiangqun Zhang
Beijing Laboratory of Intelligent Information Technology, School of Computer Science and technology, School of Computer Science and Technology, Xuchang University, Xuchang 461000, P.R.China
国际会议
The First World Congress on Global Optimization in Engineering & Science(第一届工程与科学全局优化国际会议 WCGO2009)
长沙
英文
407-412
2009-06-01（万方平台首次上网日期，不代表论文的发表时间）