The primary DEM is the one to be fixed. Holes in the data are represented
by values less than or equal to the parameter THRESH.
The secondary DEM is used to supply values to fill the holes in the
primary DEM. Several adjustments are done to fit the correction to the
hole:
1. The patch is "boosted" by the average of the difference of edge
pixels around the hole.
2. The boosted patch is adjusted by the residual differences of
edge pixels around the hole. All the residuals are weighted
by 1.0/distance-squared from each pixel in the hole so the
patch will smoothly meet the primary DEM at all of the edges.
If the primary image has pixel values representing .1 meter and the secondary is in meters use demfac=10.0 to correct the values going from the secondary to the primary. Note that an offset (different values for sea level in the two images) would have no effect on this algorithm since the second image patch is "boosted" to agree with the primary image according to overlapping pixels.
The value 2 gives 1/distance-squared interpolation on edge pixels around the hole. This means that the residual differences between the primary and secondary DEM's are interpolated to each filled (hole) pixel with a weight of one over the distance squared (distance between the hole pixel and the edge pixel). The value 1 gives one over distance interpolation. Values between .05 and 3.0 are allowed. The default value is 2.0.
Use this if your DEM has bad pixels on the edge. For example some SRTM images have changes of 1000 meters in one pixel at the edge.