A New Approach for In-Vehicle Camera Obstacle Detection by Ground Movement Compensation

The purpose of this paper is to propose a new approach to detecting obstacles using a single camera mounted on a vehicle when the vehicle is backing or turning round at an intersection at a low speed. Using restrictions among feature point locations and their optical flows in geometrically converted top-view images, ground-movement information can be estimated. Our approach compensates for the ground movement between consecutive top-view images using the estimated ground-movement information and computes the difference image between the previous compensated top-view image and the current top-view image. Finally, a new angle histogram-based algorithm is processed to extract obstacle regions using the difference image. The actual in-vehicle experimental results show that our proposed approach has tolerance for various changing illumination conditions and different road textures.