FINGERPRINT MATCHING BASED ON STATISTICAL TEXTURE FEATURES?

Fingerprint is a reliable biometric which is used for personal verification. Three arithmetical features are extracted from fingerprint images and represented using a mathematical reproduction. These features are (1) an entropy coefficient, computed from the intensity histogram of the image, (2) a relationship coefficient, computed by correlation operation between the novel image and a filtered version of the image obtained using a 2D wiener filter, and (3) an energy coefficient, obtained by first subjecting the image to a 5-level wavelet decomposition and thereafter computing the proportion energy of the approximation coefficient obtained after the 5th level decomposition. Results show that fingerprint texture feature can be convincingly used for discrimination and for personal verification.