REVISOR is a neural network-based software package that uses computer vision technology to monitor electoral procedures and count the number of actual voters.
This software can be deployed during election observation missions for a fraction of the cost of traditional methods. The Revisor system tracks physical objects, detects voting events, and distinguishes them from other activities at a polling station with high precision (up to 98% accuracy).
It is also a trainable neural network, which means that customers can teach it to work with different types of voting procedures, elections, and electoral systems in any country.
The system can detect multiple types of violations and speed up manual recounts. The software can be applied to operations based on video recordings, producing results immediately after an election and/or months and years later.
Revisor is a fast, reliable, and inexpensive system that is designed to detect ballot boxes on video records, count the number of voters who cast their ballots, identify polling stations with falsified turnout, draft a formal complaint, and speed up manual recounts.
Results from Revisor are evidenced in the success stories where it has detected discrepancies between the official and actual turnout at polling stations, reporting them to users who can resolve crimes that lead to the observed anomaly.
Overall, Revisor is an effective AI tool that leverages neural networks for election monitoring and compliance with electoral procedures.