AI Car With Real-Time Detection of Damaged Road and Lane Detection Using Deep Learning

This project brings a smart and practical AI-based approach to improving road safety and efficiency. It merges real-time detection of road damage with accurate lane identification, all packaged in an intuitive and easy-to-use system. At its core are two powerful YOLOv8 models—one trained to spot issues like potholes, cracks, and surface damage, and the other focused on identifying lane lines, even in low light or bad weather. A major strength of this system is its modern interface, built with CustomTkinter, which makes it accessible for users from different backgrounds, whether they’re engineers, city officials, or tech enthusiasts. It can analyse both live camera feeds and saved videos, displaying real-time results directly on the screen with clear overlays. Thanks to multithreading, the system handles video processing and detection tasks simultaneously, delivering smooth, lag-free performance. In practical terms, it can support local governments in scheduling timely road maintenance, help self-driving cars navigate safely, and provide useful insights for traffic and infrastructure planning. With potential for future upgrades like GPS integration for location-based reporting, this project shows how combining AI with thoughtful design can lead to safer roads and smarter cities. KEYWORDS AI Car, Road Damage Detection, Lane Detection, YOLOv8, Real-time Detection, Deep Learning, CustomTkinter, OpenCV, Multithreading, Computer Vision, Object Detection, Pothole Detection, Crack Detection, Lane Marking Identification, Graphical User Interface (GUI), Smart Transportation, Autonomous Vehicles, Traffic Infrastructure Monitoring