Is there an mistake here? the red rectangle
Zone 4 overlaps with two pixels, 3 and NoData. It looks like it doesn't quite explain it in the help, but the NoData is not counted in the COUNT output, nor considered in any of the other fields (e.g. MIN and MAX both equal 3).
Thank you very much
For more Explanation ,look at the following picture:
I draw overlaid illustration " ValRas + ZoneRas" with colors
as is the picture , the zone that has value zero ( 0 ) get blue color
,zone one (1) with green color , zone two (2) and zone four (4) with red color.
the field in the table that define the name of zone is " VALUE".
the field in the table that define count of pixels inside the zone is "COUNT".
and other filed for statistics of values raster of zone.
for example :
->zone 0 ( blue zone ) has 5 pixels , so the count is 5, and maximum raster value is 2 ( as you see in ValRas illustration) and minimum value 0 so the range is 2 ...etc
->zone 4 (red zone ) has 2 pixels one of them NoData , so the count is 1 , its value is 3 so the maximum value is 3 and that is minimum too. the range logically is 0 ...etc
I hope my explanation be cleared.
The observation by Darren is correct from the help topic
If any cell location on the Zone dataset is NoData, that location will be assigned NoData on the output.
Do note as well, that only certain functions are supported by both integer and floating point data types.
Retrieving data ...