Cutting-Edge Deep Learning Models for Real-Time Pothole Detection and Localization in Challenging Road Conditions

Potholes are a serious issue with the road network that endangers drivers' safety and seriously damages cars. Timely repairs and upkeep of roads depend on the accurate and real-time identification of potholes. Numerous studies have b een conducted on pothole identification among other pavement problems since they can cause injuries to cars and passengers and result in an accident. The research study explores how potholes can be detected using deep learning (DL. The study depends mainly on the well-known (You Look Only Once) YOLO model, examined on a Publicly available dataset from Kaggle, For Efficient detection of potholes canny edge detection techniques are also used to enhance the model accuracy. The research depicts that the YOLOv8 model is very effective when it comes to finding potholes as it has an accuracy of 99.2%. The results show that the issue of potholes can be detected and fixed using YOLOv8 in a way where the road is safe for everyone.