• Ms Somya Agrawal
• Bhavin Ameta
• Khushal Suthar

Lane detection is a fundamental component of autonomous driving and advanced driver assistance systems (ADAS), playing a crucial role in ensuring safe and efficient vehicle navigation. This paper presents a real-time lane detection framework that leverages computer vision techniques to identify and track lane boundaries from continuous video input. The system combines image processing methods with deep learning models to accurately extract and classify lane markings. The detection process starts with image enhancement to improve the visibility of lane lines, followed by the extraction of key visual features such as edges and angles to distinguish lanes from other road elements. To ensure real-time performance, the system uses algorithm optimization and parallel computing. A lightweight classification model, trained on annotated datasets, integrates both deep learning and traditional machine learning approaches to enhance detection accuracy and reduce false positives. The system’s performance is evaluated using benchmark datasets and real-world road conditions, accounting for variables such as lighting, weather, and surface types. Results show that the system delivers high precision and consistent performance, making it suitable for deployment in intelligent vehicle platforms.

Lane Detection, ADAS, Autonomous Driving, Computer Vision, Image Processing, Machine Learning, Real-Time Systems, Intelligent Vehicles.

"Real-Time Autonomous Lane Detection Using Computer Vision", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 6, page no.280-288, June-2025, Available :http://www.jetir.org/papers/JETIRGW06045.pdf

"Real-Time Autonomous Lane Detection Using Computer Vision", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 6, page no. pp280-288, June-2025, Available at : http://www.jetir.org/papers/JETIRGW06045.pdf