edge handling of rasters when reprojected #888

Open
artemp opened this Issue Oct 11, 2011 · 3 comments

Comments

Projects
None yet
3 participants
Owner

artemp commented Oct 11, 2011

The foss4g style is composed of a variety of adjacent rasters. When reprojection occurs various things can go wrong at edges. This is understandable and we likely can't handle every edge case but this one is severe.

[[Image(normal.png)]]

[[BR]]

[[Image(seams_epsg3153.png)]]

[[BR]]

[[Image(clipping_epsg3857.png)]]

Owner

artemp commented Oct 11, 2011

[albertov] Hmm, I think that this might be fixed by changing {{{reproject_raster()}}} to use map coordinates instead of screen coordinates to determine the mesh points. I'll try to fix this later today or ASAP... Can you please write down the bboxes and srss of those adjacent rasters (gdalinfo output would be more than enough), as well as for the query that visualizes them, so I can synthetise some test data?

Alberto

Owner

artemp commented Oct 11, 2011

[springmeyer] gdalinfo output on the rasters is here: https://gist.github.com/1262581

Owner

springmeyer commented Oct 12, 2011

@albertov - do you need anything else? I can track down the source images if you need, just let me know.

@ghost ghost assigned albertov Oct 12, 2011

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment