Computer vision for asphalt cracks detction using YOLOv5

Recent studies have shown that researchers proposed various techniques for Pothole detection using data collected from different parts of the world. Automating pothole will go a long way in providing safe driving road users and intelligent transportation systems. This is not only necessary to guarantee adequate performance, but also adjust drivers’ needs, potentiate their acceptability, ultimately meet preferences bad roads. paper presents computer vision model assists drivers by detecting predicting potholes while on curb accidents. The datasets used this research were images extracted kaggle which classified into two; normal object algorithm was evaluate YOLO 5. results parallel testing provided good roads potholes. predicted values all positive. two classifiers detected perfectly without being perverse. system its value percentage, therefore showing level adherence each classes detected.