Optimizing Gender and Age Group Classification through Spatial Domain Analysis of Fingerprint Features

Abstract

Biometric systems, particularly fingerprint recognition, have undergone significant evolution, extending beyond individual identification to encompass gender and age group recognition. This paper introduces a Spatial and Frequency Domain Fusion Approach for Gender and Age Group Recognition using fingerprints. The proposed algorithm integrates ridge segmentation, normalization, orientation, and frequency estimation, enhancing ridge features through Gabor filtering. Classification utilizing Support Vector Machines (SVM) results in robust gender and age group recognition. Application of the algorithm to both old and revised datasets demonstrate high accuracy, achieving 91.21% and 98.05%, respectively, in combined gender and age group classification. The approach addresses the limitations of existing systems, contributing to the advancement of fingerprint-based biometric technology.