Dongliang Yang; Changjiang Song; Gang Wu

Abstract: A palmprint classification method based on local sparse coding is presented. Firstly, patch-wise statistics based local binary patterns features are as input of local sparse coding classifier. And then the method finds k samples, which are similar with the test sample. Finally the sparse representation classifier is applied to the test sample and the k samples to determine the classification information of the test sample in the database. The experimental results on PolyU palmprint database show that new method has higher classification accuracy than the tradition methods.

Published in: 2018 IEEE 4th Information Technology and Mechatronics Engineering Conference (ITOEC)