A STUDY ON PRE AND POST PROCESSING OF FINGERPRINT THINNED IMAGE TO REMOVE SPURIOUS MINUTIAE FROM MINUTIAE TABLE

In Fingerprint recognition, after the initial preprocessing, the feature is extracted from the Fingerprint thinned image. Extraction of crucial and beneficial capabilities or features of interest from a fingerprint image is an essential venture during recognition. Feature extraction algorithms pick handiest or only applicable features important for enhancing the performance of matching and recognition rate and outcomes with the feature vector. The feature extraction algorithms or techniques require only relevant features like minutiae details and do not require any background details or domain-specific details. They need to be smooth or easy to compute with a purpose to gain a viable or practicable technique for a huge image series. Minutiae details or fingerprint ridge ending or bifurcation details using skeletonized or thinning approach is a very popular method for feature extraction. The preprocessed thinned image is further post-processed to remove some false minutiae from minutiae table and which is generated through crossing number theory. One more purpose of post-processing is to reduce the number of minutiae points by removing false minutiae structures like spurs, ride breaks, short ridge, holes or islands, bridges, and ladders. In this paper w × w window neighborhood is considered for each minutia in Minutiae Table. Minutiae Table contains Ridge ending or bifurcation code as 1 or 3 with its location details means x and y position in two columns and the sum of these details as its fourth column. These Minutiae tables are used for generating Fingerprint Hash code, which can be used as index-or identity key in order to uniquely identify an individual person or as one factor in Multifactor Authentication Model