Participants: Daniel Lopresti, George Nagy, Elisa Barney Smith

Funding: NSF "Collaborative Research: CT-T: Following the Paper Trail: Reliable Processing of Voting Records for Trustworthy Elections, " Grants CNS-0716647 (EBS), CNS-0716393 (GN) & CNS-0716368 (DL)

What was once seen as an expensive but straightforward transition to electronic voting is now viewed with increasing suspicion as a situation rife with potential risks and vulnerabilities. Computer security experts have stated unequivocally and with near unanimity that paper can play a fundamental role in guaranteeing safe and secure elections in the form of hand- or machine-marked ballots. However, the processing of such records during the initial counting of votes or in the conduct of recounts has raised its own set of problems which span technical and social boundaries.

We addressed the issues that currently make paper records more of a nuisance than an integral component in trustworthy voting systems. Specifically, we characterized the statistical distribution of mark sense errors as a function of ballot quality, created a system for unbiased visual auditing based on ballot images and studied its effectiveness, examining the possibility that a concept known as homogenous class display can facilitate manual recounts. We studied whether style-conscious recognition can replace isolated mark interpretation as a better way to model and capture voter intent, and tested a new method for anonymity-preserving image-level comparison of two independent scans of each ballot and reconciliation of the differences that may arise, developing testing procedures for the paper handling components of voting systems in accordance with operational constraints, including the modest training received by most poll workers.

Our work on voting technologies was supported by – and supports – a series of extensive user studies, surveys, and focus groups that we plan to run to measure voter understanding and acceptance.

Publications:

• IEEE Boise WIE talk, 17 February 2021, Video on ieee.tv

• EH Barney Smith, D Lopresti, G Nagy, Z Wu, "Towards Improved Paper-Based Election Technology," International Conference on Document Analysis and Recognition (ICDAR), 2011

• EH Barney Smith, S Goyal, R Scott, D Lopresti, "Evaluation of Voting with Form Dropout Techniques for Ballot Vote Counting," International Conference on Document Analysis and Recognition (ICDAR), 2011

• George Nagy, Daniel Lopresti, Elisa H. Barney Smith, Ziyan Wu, "Characterizing Challenged 2008 Minnesota Ballots," Proceedings SPIE Electronic Imaging Document Recognition and Retrieval, January 2011.

• D Lopresti, G Nagy, E Barney Smith, "Document analysis issues in reading optical scan ballots," Proceedings of the 9th IAPR International Workshop on Document Analysi (DAS), 2010

• G Nagy, B Clifford, A Berg, G Saunders, D Lopresti, E Barney Smith, "Camera-based ballot counter," 10th International Conference on Document Analysis and Recognition, pp. 151-155, 2009

• P Xiu, D Lopresti, H Baird, G Nagy, E Barney Smith, "Style-based ballot mark recognition," 10th International Conference on Document Analysis and Recognition, pp. 216-220, 2009

• George Nagy, Bryan Clifford, Daniel Lopresti, Elisa Barney Smith, Andrew Berg, Glenn Saunders, "Document Recognition in vitro," Proc. of Third International Workshop on Camera Based Document Processing, CBDAR, Barcelona, 2009.

• Elisa H. Barney Smith, George Nagy, Daniel Lopresti, "Mark Detection from Scanned Ballots," Proceedings SPIE Electronic Imaging Document Recognition and Retrieval, January 2009.

• Elisa H. Barney Smith, Daniel Lopresti and George Nagy, "Ballot Mark Detection," Proceedings International Conference on Pattern Recognition, December 2008.

• Daniel Lopresti, George Nagy, Elisa H. Barney Smith, "A Document Analysis System for supporting Electronic Voting Research," Proceedings Document Analysis Systems, Nara, Japan, September 2008.