Securing Balloting Systems through Iris Scan and Machine Learning

The current voting systems used in political elections in India have appeared to be prone to security breaches, with individuals voting more than once to create biased results in favor of their favorite candidates which eventually results in eroding and loosening up the common public’s trust. The objective of this paper is to propose a verification system for voting booths. This system will detect if the individual about to cast a vote has already cast it once and is trying to do it again to help create a biased result by tampering with the votes. By using Haar cascade, Hough Circles Transform, LBP, and Bhattacharya Distance, this paper aims to provide a more secure and robust verification system for voting booths.