Research Article
Face Recognition Voting System by using ABD-SVM Algorithm in Deep Learning
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357841, author={R. Madonna Arieth and Anush M A and Dheeraj Dheeraj and Berlin J H}, title={Face Recognition Voting System by using ABD-SVM Algorithm in Deep Learning}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={convolutional neural networks (cnn) support vector machine (svm) voter authentication machine learning}, doi={10.4108/eai.28-4-2025.2357841} }- R. Madonna Arieth
Anush M A
Dheeraj Dheeraj
Berlin J H
Year: 2025
Face Recognition Voting System by using ABD-SVM Algorithm in Deep Learning
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357841
Abstract
Today, election system serves the backbone of organization and democracy. India is among the largest developing country with an increase in rising Technologies. But the voting system is lacking. A new authentication a face recognition-based voting system is presented in this paper which is prepared to improve security and efficiency for voter verification. Our methods substitute conventional ID-based procedures with facial recognition to avoid fraud and simplification of voting process. Our approach uses OpenCV to take facial images, extracts identification features, and identifies faces with a Convolutional Neural Network (CNN). The features are then classified with a Support Vector Machine (SVM), which correctly authenticates registered voters while avoiding unauthorized access. The CNN is resistant to lighting, pose, and facial expression variations, while the SVM gives accurate classification. This combination does away with traditional ID-based approaches, enhancing accuracy and reliability. Secondly, we have been designed to incorporate a new algorithm named ABD-SVM algorithm to boost performance. Experimental findings present a high recognition rate and real-time efficiency that makes this system a feasible option for large-scale elections.
