Feature extraction of palm prints using independent component analysis

摘要

Personal recognition systems using physical characteristics of humans, such as face images, fingerprints, and hand images, have a great deal of attractions recently. One of the solutions is the use of independent component analysis (ICA) which extracts independent components from original signals automatically. In this study, we apply ICA to feature extraction of palm prints. Experimental results showed that ICA was very useful to extract independent palm prints each of which well represented the features of individuals. Palm print recognition using the independent palm prints achieved high recognition accuracy.