Corner Coordinates for a PixelisPoint type GeoTIFF
|
|
Hermann Peifer
|
Hi All,
I am somewhat puzzled by the different sets of corner coordinates that gdalinfo and listgeo report for the same PixelisPoint GeoTIFF, see [1] and [2]. Can both be true at the same time? Thanks, Hermann [1] $ gdalinfo cdda100.tif Driver: GTiff/GeoTIFF Files: cdda100.tif Size is 56392, 44743 Coordinate System is: PROJCS["ETRS_1989_LAEA", GEOGCS["GCS_ETRS_1989", DATUM["European_Terrestrial_Reference_System_1989", SPHEROID["GRS_1980",6378137.0,298.257222101]], PRIMEM["Greenwich",0.0], UNIT["Degree",0.0174532925199433]], PROJECTION["Lambert_Azimuthal_Equal_Area"], PARAMETER["False_Easting",4321000.0], PARAMETER["False_Northing",3210000.0], PARAMETER["longitude_of_center",10.0], PARAMETER["latitude_of_center",52.0], UNIT["Meter",1.0]] Origin = (1547200.000000000000000,5416000.000000000000000) Pixel Size = (100.000000000000000,-100.000000000000000) Metadata: AREA_OR_POINT=Point TIFFTAG_SOFTWARE=Arc/Info TIFFTAG_XRESOLUTION=0.0099999998 TIFFTAG_YRESOLUTION=0.0099999998 TIFFTAG_RESOLUTIONUNIT=1 (unitless) Image Structure Metadata: COMPRESSION=LZW INTERLEAVE=BAND Corner Coordinates: Upper Left ( 1547200.000, 5416000.000) ( 48d 6'29.49"W, 60d36'53.91"N) Lower Left ( 1547200.000, 941700.000) ( 17d54'14.25"W, 26d55'37.74"N) Upper Right ( 7186400.000, 5416000.000) ( 69d13'59.10"E, 59d59'56.41"N) Lower Right ( 7186400.000, 941700.000) ( 38d46'25.90"E, 26d37'44.54"N) Center ( 4366800.000, 3178850.000) ( 10d39'45.86"E, 51d43'5.40"N) Band 1 Block=56392x2 Type=Byte, ColorInterp=Palette (...) [2] $ listgeo cdda100.tif Geotiff_Information: Version: 1 Key_Revision: 1.0 Tagged_Information: ModelTiepointTag (2,3): 0 0 0 1547250 5415950 0 ModelPixelScaleTag (1,3): 100 100 0 End_Of_Tags. Keyed_Information: GTModelTypeGeoKey (Short,1): Undefined GTRasterTypeGeoKey (Short,1): RasterPixelIsPoint GTCitationGeoKey (Ascii,304): "IMAGINE GeoTIFF Support\nCopyright 1991 - 2005 by Leica Geosystems Geospatial Imaging, LLC. All Rights Reserved\n@(#)$RCSfile: egtf.c $ IMAGINE 9.0 $Revision: 10.0 $ $Date: 2005/07/26 15:10:00 EST $\nGeocoding information not available\nProjection Name = ETRS_1989_LAEA\nUnits = meters\nGeoTIFF Units = meters" ProjLinearUnitsGeoKey (Short,1): Linear_Meter End_Of_Keys. End_Of_Geotiff. Projection Linear Units: 9001/metre (1.000000m) Corner Coordinates: Upper Left ( 1547250.000, 5415950.000) Lower Left ( 1547250.000, 941650.000) Upper Right ( 7186450.000, 5415950.000) Lower Right ( 7186450.000, 941650.000) Center ( 4366850.000, 3178800.000) [3] $ cat cdda100.tfw 100.00000000000000 0.00000000000000 0.00000000000000 -100.00000000000000 1547250.00000000000000 5415950.00000000000000 _______________________________________________ gdal-dev mailing list [hidden email] http://lists.osgeo.org/mailman/listinfo/gdal-dev |
||||||||||||||||
Daniele Romagnoli-2
|
Hi Hermann,
as far as I know, GDAL always returns coordinates in the pixel_corner convention. Take a look here about conventions: http://www.vterrain.org/Misc/cellreg.html Therefore, I guess from gdalinfo you see the AREA_OR_POINT = POINT (Center_extent) metadata whilst the coordinates are exposed in Corner_extent (instead of center_extent). I don't know how listgeo works but I guess it shows AREA_OR_POINT=POINT and it produces coordinates in that convention (center_extent). This will explain why you see a 50 meters (half pixel) difference between the 2 outputs which seem coherent. Gdal shows the coordinate of the corner of the pixel whilst listgeo shows the coordinate of the center of the pixel. Also note that world files should always be in pixel is center (indeed, your world file contains 1547250 and 5415950 grid origin). Hope this helps. Best Regards, Daniele -- ------------------------------------------------------- Eng. Daniele Romagnoli Software Engineer GeoSolutions S.A.S. Via Carignoni 51 55041 Camaiore (LU) Italy phone: +39 0584983027 fax: +39 0584983027 mob: +39 328 0559267 http://www.geo-solutions.it ------------------------------------------------------- On Tue, Nov 3, 2009 at 3:11 PM, Hermann Peifer <[hidden email]> wrote: Hi All, _______________________________________________ gdal-dev mailing list [hidden email] http://lists.osgeo.org/mailman/listinfo/gdal-dev |
||||||||||||||||
Daniele Romagnoli
|
In reply to this post
by Hermann Peifer
Hi Hermann,
as far as I know, GDAL always returns coordinates in the pixel_corner convention. Take a look here about conventions: http://www.vterrain.org/Misc/cellreg.html Therefore, I guess from gdalinfo you see the AREA_OR_POINT = POINT (Center_extent) metadata whilst the coordinates are exposed in Corner_extent (instead of center_extent). I don't know how listgeo works but I guess it shows AREA_OR_POINT=POINT and it produces coordinates in that convention (center_extent). This will explain why you see a 50 meters (half pixel) difference between the 2 outputs which seem coherent. Gdal shows the coordinate of the corner of the pixel whilst listgeo shows the coordinate of the center of the pixel. Also note that world files should always be in pixel is center (indeed, your world file contains 1547250 and 5415950 grid origin). Hope this helps. Best Regards, Daniele -- ------------------------------------------------------- Eng. Daniele Romagnoli Software Engineer GeoSolutions S.A.S. Via Carignoni 51 55041 Camaiore (LU) Italy phone: +39 0584983027 fax: +39 0584983027 mob: +39 328 0559267 http://www.geo-solutions.it On Tue, Nov 3, 2009 at 3:11 PM, Hermann Peifer <[hidden email]> wrote: Hi All, -- ------------------------------------------------------- Eng. Daniele Romagnoli Software Engineer GeoSolutions S.A.S. Via Carignoni 51 55041 Camaiore (LU) Italy phone: +39 0584983027 fax: +39 0584983027 mob: +39 328 0559267 http://www.geo-solutions.it ------------------------------------------------------- _______________________________________________ gdal-dev mailing list [hidden email] http://lists.osgeo.org/mailman/listinfo/gdal-dev |
||||||||||||||||
| Free Embeddable Forum Powered by Nabble | Help |
