Thank you for the quick, helpful response. Sounds like I need to dig into the NITF spec to determine how to best use the available precision. Thanks again.
From: Even Rouault <even.roua...@spatialys.com> Sent: Thursday, July 20, 2023 11:45 AM To: Nigus, Steve <steve.ni...@flightsafety.com>; gdal-dev@lists.osgeo.org Subject: Re: [gdal-dev] NITF geotransform data erratic CAUTION EXTERNAL EMAIL: Verify sender, links, and attachments are safe before taking action. NITF base georeferencing using the IGEOLO field sucks because there's not enough decimal digits to compute the full accuracy of a geotransform, and thus when GDAL tries to reconstruct the geotransform from the 4 corners (that's how georeferencing is stored), the threshold to detect if the computed affine geotransform isn't always met. I thought that a workaround would be to write the SDE TREs (GEOPSB/PRJPSB and MAPLOB). There's a SDE_TRE=YES creation option for that, but its implementation on the writing side is limited to geographic coordinates, not projected ones (could potentially be extended). So if you want to go with that route, you'll have to compose the content of the TREs by yourself from the NITF spec, which is a non trivial exercice. But actually, I'm wrong. MAPLOB cannot capture rotating terms of a geotransform. On the reading side, you can try to set the GDAL_GCPS_TO_GEOTRANSFORM_APPROX_OK config option to a value larger than the 0.25 pixel default value. cf https://gdal.org/api/raster_c_api.html#_CPPv422GDALGCPsToGeoTransformiPK8GDAL_GCPPdi Even Le 20/07/2023 à 18:18, Nigus, Steve a écrit : I am doing a real-time SAR spot simulation where the SAR orbits a target point and saves a series of the georeferenced spot image files. My initial implementation generated non-GDAL GeoTIFF or JPEG files. I use a least-squares algorithm to develop the 6 affine geotransform values. This all works fine, and the files display nicely in my GIS viewer. I have since brought in GDAL 2.2.3 to create NITF files. Unfortunately, over 30% of the NITF files lose the geotransform info. Rather, they show generic GCPs. The “good” NITF files display OK in my GIS viewer. As a sanity test, I took a series of non-GDAL GeoTIFF files and ran them through gdal_translate to NITF. Interestingly, around 30% of the NITF files again dropped the geotransform data. I’ve attached below gdalinfo reports of a gdal_translate input GeoTIFF file and the output NITF file for a failing example. My geotransform data appears to be well-behaved. Question: What could cause geotransform data to be rejected? Thanks….. Non-GDAL gdal_translate input GeoTIFF file: $ gdalinfo sar2023071722525322.tif Driver: GTiff/GeoTIFF Files: sar2023071722525322.tif Size is 1024, 1024 Coordinate System is: PROJCS["VITAL SAR Simulation", GEOGCS["WGS 84", DATUM["WGS_1984", SPHEROID["WGS 84",6378137,298.257223563, AUTHORITY["EPSG","7030"]], AUTHORITY["EPSG","6326"]], PRIMEM["Greenwich",0], UNIT["degree",0.0174532925199433], AUTHORITY["EPSG","4326"]], PROJECTION["Transverse_Mercator"], PARAMETER["latitude_of_origin",0], PARAMETER["central_meridian",-81], PARAMETER["scale_factor",0.9996], PARAMETER["false_easting",500000], PARAMETER["false_northing",0], UNIT["metre",1, AUTHORITY["EPSG","9001"]]] GeoTransform = 586659.1336352756, -0.2027675805429681, -0.2274441341933198 2969181.096499263, -0.2274441341933198, 0.2027675805429681 Metadata: AREA_OR_POINT=Area Image Structure Metadata: INTERLEAVE=BAND Corner Coordinates: Upper Left ( 586659.134, 2969181.096) ( 80d 7'40.06"W, 26d50'29.50"N) Lower Left ( 586426.231, 2969388.731) ( 80d 7'48.44"W, 26d50'36.30"N) Upper Right ( 586451.500, 2968948.194) ( 80d 7'47.64"W, 26d50'21.98"N) Lower Right ( 586218.597, 2969155.828) ( 80d 7'56.02"W, 26d50'28.78"N) Center ( 586438.865, 2969168.462) ( 80d 7'48.04"W, 26d50'29.14"N) Band 1 Block=1024x8 Type=Byte, ColorInterp=Gray gdal_translate output NITF file: $ gdalinfo sar2023071722525322.nitf Driver: NITF/National Imagery Transmission Format Files: sar2023071722525322.nitf Size is 1024, 1024 Coordinate System is `' GCP Projection = PROJCS["UTM Zone 17, Northern Hemisphere", GEOGCS["WGS 84", DATUM["WGS_1984", SPHEROID["WGS 84",6378137,298.257223563, AUTHORITY["EPSG","7030"]], AUTHORITY["EPSG","6326"]], PRIMEM["Greenwich",0, AUTHORITY["EPSG","8901"]], UNIT["degree",0.0174532925199433, AUTHORITY["EPSG","9122"]], AUTHORITY["EPSG","4326"]], PROJECTION["Transverse_Mercator"], PARAMETER["latitude_of_origin",0], PARAMETER["central_meridian",-81], PARAMETER["scale_factor",0.9996], PARAMETER["false_easting",500000], PARAMETER["false_northing",0], UNIT["Meter",1]] GCP[ 0]: Id=UpperLeft, Info= (0.5,0.5) -> (586659,2969181,0) GCP[ 1]: Id=UpperRight, Info= (1023.5,0.5) -> (586451,2968948,0) GCP[ 2]: Id=LowerRight, Info= (1023.5,1023.5) -> (586219,2969156,0) GCP[ 3]: Id=LowerLeft, Info= (0.5,1023.5) -> (586426,2969389,0) Metadata: NITF_ABPP=08 NITF_CCS_COLUMN=0 NITF_CCS_ROW=0 NITF_CLEVEL=03 NITF_ENCRYP=0 NITF_FBKGC= 0, 0, 0 NITF_FDT=20021216151629 NITF_FHDR=NITF02.10 NITF_FSCATP= NITF_FSCAUT= NITF_FSCLAS=U NITF_FSCLSY= NITF_FSCLTX= NITF_FSCODE= NITF_FSCOP=00000 NITF_FSCPYS=00000 NITF_FSCRSN= NITF_FSCTLH= NITF_FSCTLN= NITF_FSDCDT= NITF_FSDCTP= NITF_FSDCXM= NITF_FSDG= NITF_FSDGDT= NITF_FSREL= NITF_FSSRDT= NITF_FTITLE= NITF_IALVL=0 NITF_IC=NC NITF_ICAT=VIS NITF_ICORDS=N NITF_IDATIM=20021216151629 NITF_IDLVL=1 NITF_IGEOLO=175866592969181175864512968948175862192969156175864262969389 NITF_IID1=Missing NITF_IID2= NITF_ILOC_COLUMN=0 NITF_ILOC_ROW=0 NITF_IMAG=1.0 NITF_IMODE=B NITF_IREP=MONO NITF_ISCATP= NITF_ISCAUT= NITF_ISCLAS=U NITF_ISCLSY= NITF_ISCLTX= NITF_ISCODE= NITF_ISCRSN= NITF_ISCTLH= NITF_ISCTLN= NITF_ISDCDT= NITF_ISDCTP= NITF_ISDCXM= NITF_ISDG= NITF_ISDGDT= NITF_ISORCE=Unknown NITF_ISREL= NITF_ISSRDT= NITF_ONAME= NITF_OPHONE= NITF_OSTAID=GDAL NITF_PJUST=R NITF_PVTYPE=INT NITF_STYPE=BF01 NITF_TGTID= Corner Coordinates: Upper Left ( 0.0, 0.0) Lower Left ( 0.0, 1024.0) Upper Right ( 1024.0, 0.0) Lower Right ( 1024.0, 1024.0) Center ( 512.0, 512.0) Band 1 Block=1024x1 Type=Byte, ColorInterp=Gray _______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org<mailto:gdal-dev@lists.osgeo.org> https://lists.osgeo.org/mailman/listinfo/gdal-dev -- http://www.spatialys.com My software is free, but my time generally not.
_______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev
