Deep Learning Algorithm for Face Recognition Using a Hybrid Model (original) (raw)

2024

Abstract

This study presents a robust hybrid model for face recognition, which synergistically integrates the VGG16 convolutional neural network (CNN) for feature extraction with an autoencoder for dimensionality reduction and representation learning. This paper proposed VGG16 and Autoencoder architecture efficiently extracts high-level features from images, much reducing computational complexity while the classify of the extracted features used Support Vector Machine (SVM). The proposed hybrid model has achieved a high accuracy of 98% in face recognition tasks on benchmark datasets. This high accuracy highlights efficiency of combination VGG16-based feature extraction with autoencoder of a dimensionality reduction technique and SVM classification in advancing the state-of-the-art in face recognition approaches .

Ahmed Adil Nafea hasn't uploaded this paper.

Let Ahmed Adil know you want this paper to be uploaded.

Ask for this paper to be uploaded.