Paper Title
Driver’s Assistance Technique based on Images Processing and Target Tracking

Abstract
This paper presents an algorithm that may be used in a Driver’s Assistance System (DAS). Based on the images captured with a camera mounted on the vehicle, the algorithm can detect the lane boundaries and at the same time provide information about the curvature of the road, such as its degree and orientation. This information can be used to trigger the generation of a warning signal for the driver. The algorithm is based on the formulation of the lane detection problem as a tracking problem and the use of an efficient tracking algorithm, namely the Interacting Multiple Models (IMM) with suitable models, to solve it. Experimental results show that the proposed algorithm can be applied effectively to detect the lane boundaries of the road and the changes in its geometry. Keywords - Lane Detection, Curvature Detection, Multiple Models, Kalman Filter.