The economy of the nation is significantly boosted by well-maintained roadways. Finding pavement problems like potholes helps drivers prevent collisions or vehicle damage and also aids in road maintenance. Numerous ongoing efforts in the field of transport networks aim to give drivers pertinent information about the roads and traffic patterns. Collecting data sets is one of the most crucial steps to create any recognition system. Labeling an image means pinpointing the subject which we will be trying to find. Training the algorithm through those images to detect the subjects is critical in detecting potholes. To detect potholes from real-time videos, firstly, we collected data sets containing more than 600 images of potholes. After that, we labeled those images through labeling software. We used those images to train the model which was detecting potholes from still photos given to it. Next, we used YOLOv5 to detect potholes from real-time feeds help the masses to detect potholes on roads to avoid accidents, and it will also help people related to the road works to find the potholes for further road maintenance. Deep learning and image processing methodologies are integrated with hardware tools to formulate a refined set of requirements for the pothole detection system. The system will use a camera to capture images of the road, and a deep learning model will be used to detect objects such as potholes, depth and size. We propose a new solution to automatically detect potholes on the road surface from dash camera images using a state-of-the-art deep learning based object detection algorithm, namely, You Only Look Once version 5 (YOLOv5). Cost of pothole estimation typically depends on its height, width and length. System that detecting and measuring potholes on the road after it providing accurate cost estimates for pothole repair and road construction using image processing and deep learning and accurately detects potholes on roads in real-time and displays their live location for immediate action. To achieve this goal, we used deep learning techniques and image processing to detect potholes in real-time and relay this information to via an internet connection. Potholes are a common road hazard that can cause significant damage to vehicles and pose a safety risk to drivers and pedestrians. Manual inspection of roads to identify and assess potholes is time-consuming and labour-intensive.