| 摘要 |
The disclosed vehicle surroundings monitoring device is provided with: an imaging step, in which images of the vehicle surroundings are acquired; a step in which the acquired greyscale image is converted to a multi-value image; a step in which a first edge extraction mask is applied to regions in the multi-value image that contain target candidates and multiple horizontal edges are detected; a head upper edge detection step, in which the position of the horizontal edge with the greatest edge intensity, from among the multiple horizontal edges, is detected as the position of the upper edge of the head of the target; a step, in which a second edge extraction mask that extends down from the position of the upper edge of the head of the target is applied and multiple vertical edges are detected; a head lower edge detection step, in which the position of the lower edge of the head of the target is detected on the basis of changes in the position of the multiple vertical edges; a step in which the head region of the target is determined on the basis of the interval between the position of the upper edge of the head of the target and the position of the lower edge of the head of the target; and a step in which whether or not the target is a specified monitoring target is determined on the basis of a multi-value image containing at least the upper region of the head of the target. |
