geotransform for resolution in aux file

Question

I'm trying to use (display) some files that are in ecw format and have an associated geotransform in an aux file. The images themselves are in the correct projection as matches the srs in the aux.xml file but there is a geotransform that appears to just alter the resolution so you end up needing to sort of double the stretch of the image to get it to match data in the projection. I may not be using the correct terms here but thats how it appears if you try to display the image with out the additional stretch from the geotransform in the aux file. Here is how one looks: 2.0750000000000000e+005, 5.0000000000000000e-001, 0.0000000000000000e+000, 1.3975000000000000e+006, 0.0000000000000000e+000,-5.0000000000000000e-001 My question is why what is the motivation for producing images with this additional tweak to be done on the fly in the display? Does it allow the image size to be smaller? I just don't get it. How can this data be used with any other data? Seems like the resolution change will only degrade the image. What am I missing?

ValentinWittich · Answer

I don't really get the problem, but in general GeoTransform works like that: X_COORDINATE_OF_TOP_LEFT_CORNER, (X_SIZE_IN_MAP/X_PIXEL_OF_IMAGE),-0.0000000000000000e+000, Y_COORDINATE_OF_TOP_LEFT_CORNER, 0.0000000000000000e+000,-(Z_SIZE_IN_MAP/Z_PIXEL_OF_IMAGE) So if you have any problem with the import you can check the values an see what's going wrong... or if you want to write the header file manually. My sample header file looks like that: PROJCS["DHDN_3_Degree_Gauss_Zone_3",GEOGCS["GCS_Deutsches_Hauptdreiecksnetz",DATUM["Deutsches_Hauptdreiecksnetz",SPHEROID["Bessel_1841",6377397.155,299.1528128]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]],PROJECTION["Transverse_Mercator"],PARAMETER["False_Easting",3500000.0],PARAMETER["False_Northing",0.0],PARAMETER["Central_Meridian",9.0],PARAMETER["Scale_Factor",1.0],PARAMETER["Latitude_Of_Origin",0.0],UNIT["Meter",1.0]] 3475000e+000, 8.8183421516755e+000,-0.0000000000000000e+000, 5480000e+000, 0.0000000000000000e+000,-8.8183421516755e+000 PIXEL NEAREST Regards Valentin

lowolhar · Answer

Bump...is there anyone out there able to comment on this? Really curious as to why images would be distributed in this manner.