Voter’s Recognition and Fake Using Digital Image Processing and Deep Learning in Multiple Voters Real Time

Computer Engineering

The fundamental right to vote in elections is a cornerstone of democratic citizenship. In the modern era, Electronic Voting Machines (EVMs) have been introduced, marking a significant departure from the traditional voting system in India that used paper ballots and ballot boxes. Creating a secure voting system that maintains the privacy of conventional voting methods, ensures proper voter authentication, and promotes transparency has been a long-standing challenge. Previously, the use of paper ballots was time-consuming and susceptible to malpractices like booth-capturing and ballot-box stuffing, leading to disputes and delayed election results. In this project, we propose an EVM system that employs a deep Convolutional Neural Network (CNN)-based face recognition technology to capture a voter's facial image. This image is then verified against pre-captured images in the database. If the verification is successful, the system identifies the voter as valid and allows them to cast their vote for a political party. After voting, the voter's facial data is removed from the system, ensuring that each voter can only vote once. Face recognition is the process of identifying an individual from an image of their face by comparing it to a database of known faces. While this is a relatively straightforward task for most humans, "unconstrained" face recognition by machines, particularly in settings like malls, casinos, and transport terminals, remains an ongoing and active area of research. In recent years, the availability of a vast amount of photos crawled by search engines and uploaded to social networks, containing various unconstrained elements such as objects, faces, and scenes, has spurred advances in the field of image classification, facilitated by increased computational resources and more powerful statistical models.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.