All Round Blind Spot Detection by Lens Condition Adaptation based on Rearview Camera Images

This paper deals with a vehicle detection method for realizing a blind spot warning function, under various environmental conditions. We introduced a method that is capable of discriminating the target object vehicles, under poor lighting conditions and in cases where the lens may be exposed to splashes in wet, snow and dirt roads. The image sensing of the vehicle detection consists of four functional components: obstacle detection, velocity estimation, vertical edge detection, and final classification. Such componets allow robust performances resembling geometry based approaches, with low calculation power as an apperance based approach. This paper describes the functional components, and furthermore methods to enhance the performances under low contrast conditions and also suppress false detections caused by residue on the lens, which becomes essential for installation on vehicles driven in actual road conditions.