Smart gate ticketing system using OCR and convolutional neural networks with real-time edge inferencing: A case study of Federal University Dutsin-Ma, Katsina state

Abstract

This research proposal aims to develop a Smart Gate Ticketing System utilizing Optical Character Recognition (OCR) and Convolutional Neural Networks (CNNs) with real-time edge inferencing, specifically tailored for the Federal University Dutsin-Ma in Katsina State, Nigeria. The proposed system seeks to address the significant operational inefficiencies and security vulnerabilities inherent in the current manual gate ticketing process. By automating the recognition and verification of vehicle license plates, the system aims to enhance operational efficiency, reduce long wait times, and minimize the risk of fraud. Unique to this research is the focus on environmental resilience, taking into account the prevalent weather conditions in Northern Nigeria, such as heavy rainfall, lightning, and sandstorms, which have previously been neglected in similar studies. The integration of OCR and CNN technologies will enabled high-accuracy real-time recognition and validation of vehicle entries. Edge computing will ensure low latency and high efficiency by processing data locally at the source. The study involved comprehensive data collection, model training, and system evaluation to ensure robustness and reliability. The anticipated outcome is a scalable, cost-effective solution that not only meets the current needs of the university but also sets a precedent for similar applications in other institutions and public facilities. This research contributed significantly to the advancement of smart technology integration in ticketing and access control systems.