Face recognition: using principal components and independent components
Sikander M. Mirza Zahid Riaz · 8th International Multitopic Conference, 2004. Proceedings of INMIC 2004. · 2005
This paper describes the novel approach of face recognition using the combination principal component analysis (PCA) and independent component analysis (ICA). The results produced, are up to 99.3% accurate on the subjects of ORL database. While previously, independent use of either PCA or ICA resulted in lower percentage of accuracy. Also PCA and ICA are considered equivalent. The computational strategy presented in this work yields high accuracy while maintaining efficiency.