PROJECT TITLE :

Lane Detection of Curving Road for Structural High-way with Straight-curve Model on Vision

ABSTRACT:

Curve is a traffic accident-prone place in the structural road's traffic system. A problematic aspect for assisted safe driving is detecting the lane-line and providing traffic information to drivers in a timely manner. In curved road conditions, typical lane detection technology isn't very useful. In this research, a straight-curve model-based curve detection algorithm is given, and this method has good application for most curve road circumstances. By assessing the basic properties of the road picture, the approach splits the road image into the zone of interest and the road background region. The region of interest is further separated into two sections: straight and curve. Simultaneously, the mathematical model of a straight curve is established. The improved Hough transform is used to obtain the straight model's mathematical equation. The continuity of the road lane-line and the tangent relationship between the straight and curve models are used to create the polynomial curve model. The curve fitting method is then used to solve the parameters of the curve model equation. Finally, the straight and curve detection and identification are accomplished, and the road lane-line is recreated. Experiments demonstrate that this system can reliably detect the curve lane-line, provide useful traffic information, issue early warnings, and is ubiquitous.