Lip Pattern Recognition based on Local Feature Extraction

Abstract

Lip is claimed to be a unique organ in human body and hence a candidate for being a biometric. The uniqueness of lip has been proven by the researchers by using color information and shape analysis. These measures of lip are unique for every person. This paper proposes that grayscale lip images constitute local features. The claim has been experimentally established by extracting local features from 23 grayscale lip images from 10 different subjects and measuring the accuracy of match. There are two techniques applied for extraction and matching of local features from lip, viz. SIFT and SURF. The accuracy turns out to be very high (>90%) in both the experiments conducted.