Create Your First Project
Start adding your projects to your portfolio. Click on "Manage Projects" to get started
Bank Cheque Processing System
Date
Oct 2022
This project focuses on automating the manual and time-consuming task of cheque processing in financial institutions by leveraging image processing and machine learning techniques. The system simplifies and enhances the efficiency of cheque verification through key features such as field extraction, signature verification, handwriting recognition, and an integrated dashboard. A specialized component isolates and extracts critical fields like the date, payee name, amount, and cheque number from cheque images for seamless processing. The signature verification module, developed in MATLAB , employs analysis using Harris Corner Detection to compare signatures against a database for authenticity and connected via Matlab Raspi API's . Additionally, a handwriting recognition system decodes handwritten details, such as the amount in words and payee information, with high accuracy.
For the prototype version, we integrated a Raspberry Pi, Pi Camera, and MATLAB API to demonstrate the automated processing of a single cheque. Furthermore, we proposed a manufacturable hardware design capable of processing up to 78 cheques per hour, showcasing the system's scalability and potential for real-world applications. To further enhance usability, the system includes a centralized dashboard that aggregates all cheque data for quick and intuitive review. This solution reduces manual effort, minimizes human errors, and streamlines repetitive verification tasks, offering reliability and scalability for banks and other financial institutions. It also has the potential to be extended to other sensitive financial documents requiring automated analysis.







