Image-Based Pothole Detection System Using YoloV8 Algorithm

A pothole is a significant disadvantage in the roadways, resulting from depressions formed due to the improper quality of road materials and external conditions, such as extreme weather forces and the heavy vehicles passing over them. This leads to soil erosion beneath the road surface. Potholes are responsible for various damages including punctures, wheel damages, dents on vehicles, damage to the vehicle floor, collisions, and even severe accidents. The number of deaths and injuries stands at 1481 and 3103 respectively. Therefore, accurate and fast pothole detection is crucial for improving ITS (Intelligent Transport Systems). Hence, this proposed study aims to overcome this major issue by implementing the YOLOv8 algorithm to detect the potholes present on the roads, which has the highest accuracy level of 98.18 percent, compared to existing algorithms. The study under consideration includes a comparison with the YOLOv5 algorithm, and the results demonstrate that the proposed approach outperforms the current method.