Kernel Density and Reclassified Integers

TonySalomone — Tue, 18 Oct 2011 15:37:13 GMT

I'm using kernel density to analyze the distribution of various land marks and transit stops in a city. I am reclassifying the density rasters on a scale of 1 - 5. I made 4 or 5 of these reclassified rasters so I could make a composite raster model. After creating the composite raster model, I want to normalize on a scale of 1 - 5. I use this formula to do so - (Input Raster/Highest # in Input Raster)* 5.

My problem is that I only get two resulting bands...meaning a raster band of "0" and one of "5." I'm pretty sure this is because the resulting raster type can only hold short integers and it is rounding up and down my inputted numbers. Does anyone know how to fix that?

Thank you!

Tony

Re: Kernel Density and Reclassified Integers

Luke_Pinner — Wed, 19 Oct 2011 00:54:08 GMT

Convert input or highest raster to floating point before dividing.

(Input Raster/float(Highest) # in Input Raster)* 5

If you want the output to be integer

int((Input Raster/float(Highest) # in Input Raster)* 5)

topic Re: Kernel Density and Reclassified Integers in ArcGIS Spatial Analyst Questions

Kernel Density and Reclassified Integers

Re: Kernel Density and Reclassified Integers