Changes between Version 7 and Version 8 of Processing/PixelSize

Timestamp:

Oct 9, 2009, 5:12:35 PM (15 years ago)

Author:

benj

Comment:

--

Processing/PixelSize

@@ -3 +3 @@
 == Quick reckoner ==
 
-This graph gives the theoretical pixel size at nadir for Eagle and Hawk. Pixel size will be larger at the edges of the swath, so at least to start with you should use a slightly higher value than that given by this graph. Also note that this is primarily across-track pixel size, along-track will vary with the speed of the aircraft (but we suggest you use square pixels based on the across-track pixel size).
+This graph gives the theoretical pixel size at nadir for Eagle and Hawk. Pixel size will be larger at the edges of the swath, so at least to start with you should use a slightly higher value than that given by this graph. Also note that this is primarily across-track pixel size, along-track will vary with the speed of the aircraft. We suggest that most users should use square pixels based on the across-track pixel size, but see the note below regarding along-track pixel sizes.
 
 This javascript calculator will give the exact value for pixel size given an altitude using the formula from the graph below
@@ -12 +12 @@
 
 Note this is the altitude over the ground, not the altitude from GPS (which is over the spheroid and takes no account of local terrain variation).
+
+The along-track IFOV is the same as the across-track one for Hawk, and is either 2.5 (for datasets with spatial binning 1, ie most of them) or 1.25 (for datasets with spatial binning 2) times the IFOV for Eagle. This means that the along-track pixel size scales by that much relative to the across-track size. However, the along-track pixel size is also affected by the aircraft speed and the integration time used, so for high speeds or long integration times you may find you need to use longer pixels.
 
 Logsheets provided by ARSF contain the altitude from GPS.  One must compensate for the average terrain height when using the reckoner above.  For example, if your data is over coast areas and thus at sea level, the numbers are fine as is.  If your data is over a mountain valley, one should subtract the average height of the terrain from the logsheet altitude, then use that ground-altitude in the table above.