Practical Procedures to Improve Face Recognition Based on Eigenfaces and Principal Component Analysis 1
Paulo Quintiliano, Renato Da Veiga Guadagnin, Antonio N. Santa-Rosa · 2001
Face recognition is a vital research area with some important challenges mainly concerning images taken in nonideal illumination conditions. We propose an approach based on eigenfaces and principal compo- nent analysis in which we use symmetrization techniques for poorly illuminated faces from The Yale Face Data- base. The obtained recognition rates are above 72% when we work with images with illumination problems.