uDIG

Worldimage off half a pixel

Details

  • Type: Bug Bug
  • Status: Closed Closed
  • Priority: Minor Minor
  • Resolution: Fixed
  • Affects Version/s: UDIG 1.1.1
  • Fix Version/s: UDIG 1.2.M6
  • Component/s: file format
  • Labels:
    None
  • Environment:
    XP, uDig 1.1.1 vs. GeoServer 1.7.2

Description

Images with worldfiles do not line up properly when mixing directly loaded images with a wms background.
My worldfiles contain the center coordinate of the top left pixel.
Using this Worldfile with GeoServer gives a perfectly aligned image.
Loading the Image+Worldfile into uDig results in an image that is half a pixel east relative to the real/wms location.

It seems as if the coordinate from the worldfile is taken as the left border of the top left pixel. (the vertical alignment is correct)

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
Jody Garnett added a comment -
It would be good to try this with a MRSID or GeoTiff image; although you mention GeoServer is handling it correctly.

Could you try testing your image with uDig 1.2-M2 ? The image handling in uDig has been rewritten ...
Show
Jody Garnett added a comment - It would be good to try this with a MRSID or GeoTiff image; although you mention GeoServer is handling it correctly. Could you try testing your image with uDig 1.2-M2 ? The image handling in uDig has been rewritten ...
Hide
Permalink
Karl Dietz added a comment -
Did some tests with udig-1.2-M2.win32.win32.x86:

a) JPG worldimage, does not find a suitable layer in the files
b) Tiff worldimage, does not find a suitable layer in the files
c) GeoTiff on UNC finds the layer but can't read it after adding it to the map ("unknown CRS" after manually setting the CRS I have a warning sign and a "rendering complete" message on the layer without anything on the map)
d) local GeoTiff works like a charm and is perfectly aligned

back to uDig 1.1.1:
e) local GeoTiff works but is off by half a pixel in both lat and lon as if data is taken as PixelIsPoint instead of PixelIsArea (see http://www.remotesensing.org/geotiff/spec/geotiff2.5.html#2.5.2)
Show
Karl Dietz added a comment - Did some tests with udig-1.2-M2.win32.win32.x86: a) JPG worldimage, does not find a suitable layer in the files b) Tiff worldimage, does not find a suitable layer in the files c) GeoTiff on UNC finds the layer but can't read it after adding it to the map ("unknown CRS" after manually setting the CRS I have a warning sign and a "rendering complete" message on the layer without anything on the map) d) local GeoTiff works like a charm and is perfectly aligned back to uDig 1.1.1: e) local GeoTiff works but is off by half a pixel in both lat and lon as if data is taken as PixelIsPoint instead of PixelIsArea (see http://www.remotesensing.org/geotiff/spec/geotiff2.5.html#2.5.2)
Hide
Permalink
Karl Dietz added a comment -
Two test images to illustrate UDIG-1474. Both show a black square aligned at 3500000,5500000 in EPSG:31467
Show
Karl Dietz added a comment - Two test images to illustrate UDIG-1474 . Both show a black square aligned at 3500000,5500000 in EPSG:31467
Hide
Permalink
Jody Garnett added a comment -
Implementation of world image is completely different now
Show
Jody Garnett added a comment - Implementation of world image is completely different now

People

  • Assignee:
    Simone Giannecchini
    Reporter:
    Karl Dietz
Vote (0)
Watch (2)

Dates

  • Created:
    Updated:
    Resolved: