PathPilot:OpenCV Based Lane Detection for Safe Navigation

Safe navigation is enabled by lane detection, which is an important component of autonomous vehicles and Advanced Driver-Assistance Systems (ADAS). Existing methods struggle with accuracy and robustness because of variations in lightning conditions, road-types and weather conditions. Due to these challenges there is a need to develop a efficient lane detection framework to overcome these problems. The aim is to develop a lane detection system using OpenCV which improves accuracy and robustness in various environments and different road types. Traditional computer vision techniques have been used. OpenCV is used for image preprocessing, applying canny edge detection and Hough transform to detect lanes. Lane region segmentation is used for accurate detection. It processes the images effec- tively and performs feature extraction, in addition to that it also improves performance. The model is trained by a dataset having various images including different road conditions and environmental variations like foggy scenarios, nighttime, daytime etc. Experimental results shows improved accuracy and robustness in various lightning conditions, including daytime, nighttime and foggy scenarios. The model achieves real-time performance, processing 63.66 frames per second with an accuracy of $92 \%$. This project efficiently and accurately detects lanes using OpenCV. The proposed approach shows potential to improve safety in autonomous navigation and ADAS applications. Future work will focus on integrating this system with other ADAS components, exploring its applications in realworld scenarios and exploring transfer learning techniques.