Dear Joelle,

Curious by your question, I tried it immediately on a simple 3 by 3 pixel raster map 
with your pixelvalues in it. I have the most recent ILWIS 3.12 in use.
I applied the avg3x3 filter and I obtained the expected  1765.667 for the central 
pixel (in case of course,  I put the decimal output precision on 0.001)
Hence I cannot yet explain your outcome.

On the other hand, if I put the precision on 100 (so the outcome will be rounded to 
the nearest multiple of 100), I get indeed the central average equal to 1800. Could 
this explain your results?

Succes

Jan Hendrikse

ITC / ILWIS team

-----Original Message-----
From: Joelle Dagnelies [mailto:[EMAIL PROTECTED]
Sent: Tuesday, September 23, 2003 09:56
To: Ilwis list
Subject: (ilwis-list) filter avg 3x3


Dear list members,

I have a question concerning the avg3x3 filter : I used this filter on an 
image, and wanted to verify the result. For this, I choosed a pixel in the 
original image (in the central part of the image) (with DN = 2507, see 
below), made the mean with it and its 8 surrounding pixels (result : 
1765.667) and compared the result with the same pixel in the filtered image 
(result : 1800 !?). Can somebody explain me where this value is coming from 
? maybe I didn't understand the filter avg3x3 principle ?!...

pixel and its surroundings : 1050  1948  1771
                                     1204  2507  2641
                                     827   1782  2161

Thank you,

Joelle

_________________________________________________________________



________________________________________________________

To leave the ILWIS-LIST discussion list, send a email with
'unsubscribe ilwis-list' in the body to '[EMAIL PROTECTED]'

Reply via email to