"North-up" (as I understand it) refers to raster images stored with the northerly direction at top, and each subsequent line of pixels is a bit farther southward. Rasters' georeferencing metadata usually include

  • the coordinates of the origin; this is the top-left pixel for a north-up image.
  • the horizontal and vertical pixel size; the vertical size is negative for a north-up image. (ESRI World File reference; GDAL reference)

Every raster I've seen has followed this convention. Do rasters ever exist in the wild whose vertical pixel step size is not negative? If so, what motivates this orientation?

  • 1
    Many years ago I had problems with a TiffScanlineDirection flag - one piece of software honored this setting but another ignored it and made the images top left oriented, the end result is the image was correct in one but not in the other. Fortunately it was as easy as editing in photoshop to rectify the situation. Almost all rasters you will come across are rectified with top left origin and no rotation, this doesn't mean that unrectified imagery can't exist, it's just not very common. – Michael Stimson Nov 19 '14 at 5:54
  • 2
    Most rasters for arctic and antarctic areas are not north-up, and MODIS Swath imagery follows the route the satellite has taken, with a varying angle towards North. – AndreJ Nov 19 '14 at 6:11
  • North-up is different to variations in affine transform conventions, they are totally different concepts. Many projections are radial about a pole, besides many other counter examples to the OP – mdsumner Nov 19 '14 at 7:10
  • @mdsumner You clarified a problem with my question. I really intended to ask about the Affine transform's vertical scale size's sign; I don't really care about north vs south. – Jeff G Nov 19 '14 at 17:57

At first, you should distinguish "georeferenced rasters" from "non georeferenced raster".

The second group includes all the raw image acquisitions from remote sensing (aerial photographs, satellite images, UAV images) or scanners. Those raw images will be captured according to the sensor orientation, which is rarely North up. They are affected by geometric distortions and the coordinates of the pixels are relative to the sensor.

In the first group, the coordinates of the pixels relate to an absolute position on the ground in a given coordinate system. Most coordinate systems are North up (according to Wikipedia, this convention was established by Ptolemy), so you will find North up coordinate systems if you work at low latitude. However, north up coordinate systems are not good at the poles, so polar coordinates systems are not North up (see an example here). Once you have a coordiante system, you need an affine transform based on one corner coordinate (most of the time the upper left corner, but not always, it can depend on the software), the size in X and Y, and sometimes a rotation. The rotation is rare because it often increases the processing times, but it can help to avoid increasing the size of a raster by unnecessary NoData pixels, so you can still have some "non north up" rasters with a "north up" coordinate system.

  • It sounds like the coordinate reference system is a huge determiner of the affine transform. Good example about rotating to match the image capture orientation (not exactly my question, but rotation would have been my followup question). – Jeff G Nov 22 '14 at 3:19
  • Or as another example: gis.stackexchange.com/questions/174079/… – Matthias Kuhn Dec 19 '15 at 12:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.