Automatic fingerprint classification system using fuzzy neural techniques
Samia Mohamed, Henry O. Nyongesa · 2003
The paper presents a fingerprint classification system and its performance in an identification system. The classification scheme is based on fingerprint feature extraction, which involves encoding the singular points (Core and Delta) together with their relative positions and directions obtained from a binarised fingerprint image. Image analysis is carried in four stages, namely, segmentation, directional image estimation, singular-point extraction and feature encoding. A fuzzy-neural network classifier is used to implement the classification of input feature codes according to the well known Henry system. Fingerprint images from NIST-4 database were tested and, 98.5% classification accuracy was obtained for the five class-problem.