A Novel Ear Recognition Method Using Feature Combination

Document Type : Original Article

Abstract

In this paper we aim to improve the accuracy of the recognition rate in the ear recognition by developing theoretical framework for combining (serial or parallel) statistical feature extraction methods (PCA, LDA, and DCT). Experimental comparisons of the combining methods demonstrate that the combination methods outperform other single feature extraction methods.

Keywords

Main Subjects

Files

Share

How to cite

Statistics