Deep Learning Based Gender and Age Estimation from Human Iris
9 Pages Posted: 17 Apr 2020
Date Written: June 2019
Abstract
The existing biometric systems recognize and accept a person, else simply reject a person if not enrolled. These systems do not give software at-tributes of a person which is needed to search impostor. In this context, the work illustrated in this paper is carried out to show the age and gender attributes of a person at the time of iris recognition. We propose deep convolution neural network for age estimation and pre-trained deep CNN AlexNet and GoogLeNet for gender prediction from iris. Deep learning pre-trained networks are adopted to extract features from iris images. Further, these features are trained and clas-sified using the multi-class SVM model for performance evaluation of the sys-tem. The hypothesis stating ‘the iris has age and gender-related information’, is proven correct from experimental results.
Keywords: Age and Gender Estimation, Deep Learning, Iris Feature Extrac-tion, Classification.
Suggested Citation: Suggested Citation