Hi David,
On 2 November 2012 23:30, David Glenn <[email protected]> wrote: > Greetings All! > > I did some analytical research on the files that VPB makes and confirmed > that the software builds a map to a LOD that is equal to the resolution of > the texture being used and stops at that point. I also discovered that if add two different types of textures shay one that > has 15 meters of resolution per pixel and another texture that has say 1 > meter of resolution per pixel > that it process all areas up to the point that the 15 meter data is > exposed then continue on only the areas that only the 1 meter data is > exposed. > > That’s very cool! :) > > The Question I have is the question that I know I'm going to get asked, is > what the function that does this is? How does it determine where to stop > relative to the size that makes up the area of the terrain? Maybe easier to > ask, is there any paper that explains how this works? > The way it works is to keep descending the quad tree LODs level till a LOD level matches or exceeds the resolution of the source data (imagery or DEM) that covers that particular tile. The -l command option will provide a maximum number of level to descend down to so if this is less than what is required then you will not get the full resolution required. I haven't written a paper on VPB's algorithms. Will you be publishing yours? Perhaps myself and others can help edit/make suggestion clarifications. Robert,
_______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
