Vehicle Type Detection and Classification System To Determine Parking Rates Based On Image Recognition

Abstract

This study aims to develop a system for detecting and classifying vehicle types using the Convolutional Neural Network (CNN) model YOLO V5 based on image recognition. This research consists of several stages, from the potential and problem stages, needs analysis, literacy studies, prototyping, system design, and system testing. The collected datasets were taken using smartphone cameras and webcams with a total of 800 image datasets, divided into two categories: training data and validation data. System testing is carried out in day and night conditions. The classification test results in daytime conditions obtained an accuracy of 93, an accuracy of 80%. The system's design for detecting and classifying vehicle types for determining parking rates based on image recognition works well. Each type of vehicle can be seen and ranked by the system.