Vehicle Detection Using Horizon and Base Approaches for the Real Time System
Bassam Daya, Al Hussain Akoum, Rwayda Abdallah · 2010
Vehicle detection is an important problem with application to driver assistance systems and autonomous, self-guided vehicles. The goal was to detect, as much as possible, in real time system the vehicles on images extra road scenes urban, resulting from a base of abundant data. These images are in levels of gray which eliminates the methods using spaces of color. In this paper we will de- scribe the classical methods: horizontal gradients and contour detection used to detect the positions of vehicles in an image. From this study, our work of devel- opment is to create, test and improve one of the detecting methods which is the method of horizontal gradient. In this context, we have developed two ap- proaches of horizon and basic and we have showed that the execution time de- creases and the detection percentage increases.