Composite image simulation involves merging two or more images that have the same viewing
station/location, as defined by
x
,
y
,
z
coordinates and perspective parameters. Composite static
images most frequently involve merging photographic or video base images for which the
camera location and settings are known and/or calculated. These are then merged or overlaid
with a 3-D computer image that has been generated using the real world location and settings for
the virtual camera (computer camera). In this process, the background image must be taken so
that the geographic location parameter, the viewpoint, the camera settings, and the actual size of
some of the objects in the image are known. This is very important so that when changes are
introduced (such as proposed transportation features), they can be matched to provide
geometrically and dimensionally accurate images. A three-dimensional computer model is used
to generate a perspective using the same coordinate camera location and camera settings as those
of the original background or base image. These two images are then overlaid using control
points to create the final composite image. This method provides the most visually correct,
accurate, and defensible representation possible. With proficiency and care, the range of error is
generally less than two percent. Several additional steps are also involved, such as material
definitions and material rendering, to complete the final composite image.