Fast Fingerprint Identification based on the DoG Filter

Biometric recognition is the study of recognizing individuals through their distinct physiological or behavioral traits, which encompass iris, facial features, fingerprints, retina patterns, vein structures, hand geometry, handwriting, human gait, signatures, keystrokes, and vocal characteristics. Fingerprint images comprise various whorls, arches, loops, ridge ends, and ridge bifurcations on each digit. The extraction of these features with some transformations can uniquely identify an individual. Fingerprints are extensively utilized in numerous biometric identification systems; finger scanning is efficient, user-friendly, and a well-established technique. This study examines the extraction of the fingerprint features and enhances the efficacy of the fingerprint authentication system by implementing a novel extractor for fingerprint features utilizing the DoG(Difference of Gaussian) Filter. The extracted features are encoded and prepared for the matching process to get the system identification and verification results.