Determination of gender from fingerprints using dynamic horizontal voting ensemble deep learning approach

Abstract

Despite tremendous advancements in gender equality, there are still persistent gender disparities, especially in important human activities. Consequently, gender inequality and related concerns are a serious problem in our global society. Major players in the global economy have identified the gender identity system as a crucial stepping stone for bridging the enormous gap in gender-based problems. Extensive research conducted by forensic scientists has uncovered a unique pattern hidden in the fingerprint and these distinguishing characteristics of fingerprints can be utilized to determine the gender of individuals. Numerous research has revealed various fingerprint-based approaches to gender recognition. The purpose of this research is to present a novel dynamic horizontal voting ensemble model with hybrid Convolutional Neural Network and Long Short Term Memory (CNN-LSTM) deep learning algorithm as the base learner to automatically determine human gender attribute based on fingerprint pattern. More than four thousand Live fingerprint images were acquired and subjected to training, testing and classification using the proposed model. Result of this study indicated over 99% accuracy in predicting person’s gender. The proposed model also performed better than other state-of-the-art model such as ResNet-34, VGG-19, ResNet-50 and EfficientNet-B3 model when implemented on SOCOFing public dataset.