Effect of eyelid and eyelash occlusions on iris images using subpattern-based approaches

Abstract

The effect of eyelid and eyelash occlusions on iris images is investigated in this study using subpattern-based approaches. Principal Component Analysis (PCA), subpattern-based PCA (spPCA) and modular PCA (mPCA) methods are used as feature extractors to recognize occluded iris images. In order to eliminate the effect of illumination changes, histogram equalization and mean-and-variance normalization techniques are used. Various experiments are carried out on UBIRIS, CASIA and MMU iris databases to demonstrate the effect of eyelid and eyelash occlusions on iris images. The results of the experiments are consistent with the results of other biometrics systems using PCA, spPCA and mPCA approaches. ©2009 IEEE.