<div>Frank,</div> <div> </div> <div>I have been successful in using GDALGCPsToGeoTransform, but</div> <div>I would like to have zero error at the GCPs. I was hoping to use</div> <div>TPS transformer for this purpose. My thought was to first get</div> <div>the GDAL GCP transformer working, then to get brave and try</div> <div>the TPS transformer. I was unfortunately not able to get good results</div> <div>with the GCP transformer, and got nothing when I used the TPS transform</div> <div>method. (My geotransform had parameters on the order of 10e67 in it.)<BR></div> <div>I guess my main question would be if this is the correct approach to the</div> <div>problem. If it is, I'll keep working at it. If not, I can live with using the </div> <div>least squares transform used in GDALGCPsToGeoTransform.</div> <div> </div> <div>Thanks again,</div> <div> </div> <div>Chris Pozsgai</div> <div><BR><B><I>Frank Warmerdam
<warmerdam@pobox.com></I></B> wrote:</div> <BLOCKQUOTE class=replbq style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">chris pozsgai wrote:<BR>> Hello again..<BR>> <BR>> I would like to repost my query to the mailing list regarding the thin plate<BR>> splines... (I now have access to my code and the internet at the same time.)<BR>> <BR>> I was able to get reasonable results using the<BR>> GDALGCPsToGeoTransform function to generate a geo transform, but I would<BR>> like to have an exact fit at the GCPs. That drove me to delve further <BR>> into the<BR>> GDAL functionality where I ran across the warping functions.<BR>> <BR>> I tried to first use the gdal warper to be able to reproduce the results <BR>> I got<BR>> using the GDALGCPstoGeoTransform function with the warper. I was<BR>> not able to do that, but here is the code that I wrote. My intention was to<BR>> look at the TPS transform's
ability to preserve the GCP values, but I <BR>> could not<BR>> get that to work, or even to reproduce the results from <BR>> GDALGCPsToGeoTransform<BR>> using regular GCP warping.<BR>> <BR>> I apologize for putting this much code on the mailing but I guess it's <BR>> the best way....<BR><BR>Chris,<BR><BR>I'm sorry for being dense, but I've lost track of what you are asking us to<BR>look for in the following code. Are you having trouble getting the warp<BR>api to work with the GCP transformer (using polynomials)? Or is it just<BR>the TPS that wasn't working for you? Is it that you get a poor warp or<BR>that nothing works?<BR><BR>I hesitate to try and grok a substantial amount of code in the hopes of<BR>then realizing what it is you wanted me to look for. This is very time<BR>and frustration intensive.<BR><BR>Best regards,<BR><BR>-- <BR>---------------------------------------+--------------------------------------<BR>I set the clouds in motion - turn up
| Frank Warmerdam, warmerdam@pobox.com<BR>light and sound - activate the windows | http://pobox.com/~warmerdam<BR>and watch the world go round - Rush | President OSGeo, http://osgeo.org<BR><BR></BLOCKQUOTE><BR><p> 
<hr size=1>Don't pick lemons.<br>
See all the <a href="http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE0OGRsc3F2BF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDbmV3Y2Fycw--">new 2007 cars</a> at <a href="http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE0OGRsc3F2BF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDbmV3Y2Fycw--">Yahoo! Autos.</a>