A surrounding area of a vehicle is imaged, and target objects in the area are detected based on captured images. A driver is warned of the presence of the detected target objects by displaying a warning indication on the displayed image. If a total number of simultaneously detected target objects is equal to or less than a specified number, the warning indication of a first display mode is applied in which each target object is emphasized individually. If the total number is more than the specified number, the warning indication of a second warning display mode is applied in which one emphasized indication predefined is displayed for the whole displayed image without displaying the warning indication of the first display mode. The warning image becomes easier to understand for a driver, compared to an image where all detected objects are always emphasized individually regardless of the total number of them.