Potholes are a nuisance to society, especially to individuals who use public roads. The importance of road infrastructure for society is comparable to the importance of blood vessels in humans. Crevasse and gouge are the main factors that cause road accidents and damage to vehicles. They should be spotted and corrected before they become a hazard. Road conditions can thus be improved through the detection of potholes. It is the goal of engineers to continuously monitor and repair the roads to ensure that they are in good shape. Having objective and comprehensive data about the condition of the roadways is a promising method for achieving this objective. About two million kilometers of roads in India are surfaced with around one million kilometers being poorly constructed. The various problems that plague Indian roads are largely caused by the improper maintenance of the roadways. No matter where you go in India, you’ll find one or more roads with potholes. While Indians have learned to perfect their driving skills to compensate for disheveled roads, there are many accidents around the country. This paper compares various subjects such as YOLO, SSD, HOG, Neural network, Inception V3,2D LiDAR, CNN-DL, R-CNN, etc. for predicting well holes. This paper introduces the state of the art in well-known mining detection techniques that discuss a variety of methods and identify the best solutions for real-time implementation under extreme conditions and working conditions thereby ensuring human safety.