What is the difference between using a Digital Elevation Model (DEM) or a constant value for ortho-rectification?
The process of ortho-rectification uses height data in order to correct for errors introduced through the variation of terrain, known as height distortion. The higher the variation of the terrain, the more important is the use of an accurate, high resolution DEM.
However, it is not always possible to obtain accurate DEM information for the area of interest. This could be the case for countries where there is little or no ground information available or for areas, which are difficult to access. In these cases it is possible to use a constant height value instead of a DEM for the ortho-rectifcation process. This is useful in areas of minimal relief variation (<20 meters). However, it is important to note that when using a constant value the resulting ortho-rectified images will not be 'truly' ortho-rectified. The output images will not be as accurate as if a DEM had been used and there will still be some height distortion in the image. Therefore the use of a constant value for height should only be used in flat areas or in areas where it is not possible to obtain a DEM.
Note: LPS ATE allows users to generate Digital Terrain Models (DTMs) from triangulated image data, including raster DEMs. For areas where a DEM is not available, a DEM could be processed using LPS ATE (from the same LPS block file) and then could be used as the DEM for the ortho-rectification.