摘要 |
An object of the present invention is to detect a three-dimensional object by use of a single-eyed camera (201) with a higher degree of accuracy by eliminating false detection of an obstacle, which is caused by a detection error of a vehicle speed sensor or that of a rudder angle sensor. A top view of a first image including a road surface image is created. The first image is imaged by a camera mounted to a vehicle. Then, a top view of a second image is created. The second image is imaged at a timing different from the timing when the first image is imaged. The two top views are associated with each other on the basis of a characteristic shape on the road surface. In each overlapped portion of the two top views, an area in which a difference occurs is identified as an obstacle. |