9P-E-1 Extracting Obstacles from a Video Taken by a Moving Camera(Room E International session)

抄録

This paper proposes a method to detect obstacles from a video taken by a moving camera mounted on a vehicle by background subtraction. The background subtraction is often used to detect moving objects when camera is static. But according to the characteristics of a road, we can also employ Gaussian mixture model to detect all objects (either static or moving objects) on the road in the case of moving camera. After background subtraction, we get a foreground image which includes all objects on the road, buildings outside the road, shadows and road marks. Our goal is to detect obstacles on the road, so next we use Hough transform to detect road area to delete buildings ; we use NCC to delete shadows; and we use triplets to detect a zebra crossing and delete it. In our experiments, it is shown that proposed method is able to detect obstacles such as vehicles and pedestrians on the road.