FACE RECOGNITION USING CNN

Abstract

The use of face recognition is growing at an astounding rate these days. Currently, researchers are developing many methods
for how a facial recognition system functions. People often find themselves alone in their families in situations such as normal
disasters, kidnappings, accidents, missing persons cases, and many more situations. To reach the family of those refugees and
ensure their safety and support, it is imperative to identify their relatives. Every day, police departments enroll missing cases.
Some of these enlisted cases are resolved, but not all of them are resolved by employing the labor-intensive manual approach.
This study aims to address the time delay caused by current police examination procedures by leveraging the latest innovations.
Therefore, we implement a framework that makes use of the CNN (Convolutional Neural Network) technique and the VGG16
architecture. We begin with our input dataset, which consists of 84 photos taken from 21 different households. The final dataset,
obtained by applying the enhancement method, consists of 1512 photos, of which 80% are utilized for training and 20% are for
testing. This framework offers a quick and easy method for locating a refugee's personal and helps validate an individual's
identity by using their picture and family information with related models that are more accurate.