Revolutionizing Electoral Integrity With A CNN-Powered Smart Voting System Using Advanced Facial Recognition Techniques
M. Kathiravan, Lubin Balasubramanian, S. Sreesubha, K. Venkataramana, A. Irumporai, V Sathyadurga · 2024
A Smart Voting System is a sophisticated web service that offers a straightforward and secure way to conduct elections using blockchain technology. It guarantees the accuracy of votes, even in real-time voting, with fraud nearly eliminated, making the electoral process more straightforward and democratic. This research proposed CNN (convolutional neural network) algorithms, which provide a secure and easy-to-use face recognition-based user interface. The system compares a facial scan with data stored on central servers and confirms your identity if a match is found. The system uses the acronym UDD(Universal Democratic Delivery) to represent a voter’s ability to vote on e-Enabled days, DDM(Direct Democracy Machine) and DRE(Direct Recording Electronic) machines, dead-end checking with removable inking pins, discounted absentee ballots, and more. The paper thoroughly discusses the claimed 99.5% accuracy of the CNN algorithm and explores the potential for private information leakage if implemented directly on user devices. This approach enables a more democratic process to emerge, ensuring that elections are both authentic and convenient. The experimental results demonstrated higher accuracy than traditional methods.