Enhanced Machine Learning Vehicle Plate Number Recognition System

Abstract

The Automatic Vehicle Number Plate Recognition System (AVNPRS) is a system that reads vehicle registration number plates from photographs using optical character recognition (OCR), which produces vehicle location data. Police departments all across the world employ (AVNPRS) to enforce the law, including determining whether a vehicle is registered or licensed. In this research work, we address the problem of car license plate detection. We developed a model for the Automatic Detection of Vehicle Number Pate Recognition. An Optical Character Recognition system which works with Convolutional Neural Network was employed. The system achieves approximately 95% accuracy license plate detection and recognition for majority of the classes of the dataset tested. The system is also tested with different condition complexities, such as rainy background, darkness and dimness, and different hues and saturation of images. It can be concluded from this work that the AVNPRS system herein develop is efficient in the recognition of vehicle plate Numbers, hence, and adaptable solution for the tolling of vehicles.