I. Introduction
THE increasing interest on advanced driver assistance systems (ADAS) for the improvement of road safety has been reflected in the joint involvement of universities, research centers and car manufacturers, as well as in the deployment of national and international projects to address this issue. According to statistics, most of the accidents are caused by other cars, therefore vehicle detection arises as the key challenge for ADAS. In particular, the research on approaches based on image analysis for vehicle detection is gaining interest on account of its low cost and flexibility, and of the increased processing capabilities.