A Novel Method of Fingerprint Classification Using Image Parameters on ANFIS

Date Added: Oct 2009
Format: PDF

Numerous algorithms have been developed to improve the speed of fingerprint recognition and this assumes prime importance when dealing with a large database of fingerprint images. This paper proposes a novel technique of using the image parameters like mean, median, variance, standard deviation and root mean square value to train an Adaptive Network Based Fuzzy Inference System (ANFIS) to classify the test image into one of the six popular categories of fingerprints namely arch, tented arch, left loop, right loop, whorl and twin loop. This classification allows to proceed with the actual matching algorithm only with the images which fall in that particular category of ridge structure thereby saving time.