A new vehicle detection algorithm using gray-level features

摘要

Vehicle detection has become a necessary part of advanced driver assistance systems (ADAS). In this paper, our main focus is on improving the performance of single camera based vehicle detection system. A new gray-level feature (GLF) obtained from gray-scale map is a description of the difference of potential target area (PTA) and potential background area (PBA). According to the result of density filtering of GLF, the existence of vehicle in extracted block can be verified. Experiments prove that in different scenes, this algorithm utilizing GLF is extremely fast and effective.