ROAD LANE LINE DETECTION USING DEEP LEARNING

Abstract

This model introduces a many technical advancements have recently been made in the field of road safety, as accidents are being increasing day by day, and one of the main causes of such accidents is a driver's lack of attention. To reduce the incidence of accidents and keep safe, technological improvements should be made. One model is to use Lane Detection Systems, which function by identifying lane borders on the road and alerting the driver if he switches to an wrong lane marking. A lane detection system is an crucial part of many technologically advanced transportation systems. Anyhow it is a difficult goal to fulfil due to the various road conditions that a person finds, particularly when driving at night or in daytime. A camera positioned on the front of the car catches the view of the road and detects lane boundaries. The model used in this research divides the video image into a series of sub-images and generates image-features for each of them, which are then used to recognize the lanes on the road. Several methods for detecting lane markings on the road have been presented. Keywords: open cv , CNN, Deep Learning.