Pothole Detection and Prevention using YOLOv5 in ITS (Intelligent Transportation System)

Abstract

Potholes on roads can cause serious damage to vehicles and accidents, so it is essential to detect them quickly and accurately. Determining appropriate strategies for ITS (Intelligent Transportation System) service is critical. In this study, the proposed solution employs YOLOv5 to perform real-time detection of potholes in images and videos. The dataset of annotated images and videos containing potholes, were used to train and fine-tune the algorithm. The proposed approach exhibits exceptional accuracy in detecting potholes, highlighting its capacity to boost road maintenance efforts while reducing the occurrence of accidents related to potholes.