Select to view content in your preferred language

Normalizing Values within Rasters using Raster Calculator

2279
3
04-28-2011 08:21 AM
DavidEvans2
Deactivated User
I am a University undergrad working on a project to study the density of different classes of storms in South America. I am trying to compare two raster based Kernel Density maps to determine the difference in distribution of two classes of storm.

The problem is that the value ranges (difference between the minimum and maximum density values) is quite different between the two due to the fact there were significantly more storms in Storm Class A then Storm Class B and therefore cannot directly compare the two density maps with each other.

What I would like to do (if possible) is apply some form of Normalization to each raster to transform the values to use the same range (0 - 1).

The Min-Max Normalization method I have learned in during my course works as follows

v???(i) = (v(i) - minA) / (maxA - minA) 

where v'(i) is the normalized observation value, v(i) is the original observation value, maxA is the highest observed value and minA the lowest observed value.

If someone could provide me with a formula for use within the Raster Calculator or suggest some other method for performing this Normalization operation I would be hugely grateful.

Thank you in advance for your assistance
0 Kudos
3 Replies
curtvprice
MVP Alum
Which version of ArcGIS?
0 Kudos
VitorVasconcelos
Deactivated User
It is worth to look a more general discussion about this topic in:

http://forums.arcgis.com/threads/48697-How-do-I-normalize-raster-data?p=348623#post348623
0 Kudos
MarkBoucher
Honored Contributor
A David Curtis, PhD with West Consltants had done extensive work with radar and storms. I found his resume at http://www.ametsoc.org/boardpges/cwce/docs/profiles/CurtisDavidC/resume.pdf.  I don't know if he would give you a GIS answer, but there might be something in his work that could help.
0 Kudos