Automated Bank Cheque Field Extraction Using Cutting-Edge Object Detection AI Models

The proposed research project aims to tackle the issue of bank cheque fraud. The project will use advanced object detection AI models to accurately extract and segment bank cheques into their component fields, a crucial step in detecting fraudulent activity. These AI models will be trained to recognize specific parts of a cheque, such as the payee name, date, MICR code, and amount. This approach is more flexible and robust than existing methods, as it does not rely on knowing the exact location of each part of the cheque. The enhanced fraud detection capabilities will allow Verafin to better protect its clients from bank cheque fraud, potentially leading to increased customer trust and satisfaction. Additionally, the automated system developed through this research could streamline Verafin’s operations, saving time and resources that can be redirected toward other areas of the business. This aligns with Verafin’s mission to provide innovative, effective solutions for fraud detection.

Faculty Supervisor:

Karteek Popuri;Lourdes Peña-Castillo

Student:

Partner:

NASDAQ Canada Inc

Discipline:

Computer science

Sector:

Finance and Insurance; Information and Communication Technology; Artificial Intelligence

University:

Memorial University of Newfoundland

Program: