Same color scale

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Same color scale

alain.reymond
Dear all,
I want to compare 2 raster data map where probability values between 0 and 1 are not the same (different mix-max values).
For this I want to apply to both map the same color scale (i.e. 0-0.1 red, 0.1-0.2 blue etc.) in an automatic way without doing it individually for each map.

I have tried many ways :
- reclassifying the probability. But here the problem is if one class do not exist the color will shift and I cannot assign a particular color to a particular class to avoid this.

- rewriting the breaks limits (sequence from 0 to 1 by 0.1) and using the RampPalette but it did not work either because of the max-value of the data (they are not the same and not in the same sequence) and one color will be add in the rampPalette to fill in the difference. It looks like as in the rampPalette there is like a defined min-color and max-color and it will add as many as needed between those 2. Therefore I wrote by hand some color for the sequence and it worked but I had to delet colors to adapt to the max number of really existing sequence. Hence, if I want to do this for a lot of files I can't make it that way.

- I also used the ggplot, but again, there is a min-color and a max-color for the min and max values (i.e. blue-red). If one ggplot has more sequence than the other one, in both case they will have the blue and red as min and max color but then not the same number of color between those 2. And If I do not set a max-color (or any min and max color) it does not work either.

I already thank you if you have any solution to my problem that could be used "easily" without having to write specific color for each map.
Best,

Alain




=
        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: Same color scale

Hadley Wickham-2
> - I also used the ggplot, but again, there is a min-color and a max-color for the min and max values (i.e. blue-red). If one ggplot has more sequence than the other one, in both case they will have the blue and red as min and max color but then not the same number of color between those 2. And If I do not set a max-color (or any min and max color) it does not work either.

Just set the limits - e.g. scale_fill_continuous(limits = c(0, 1))

Hadley


--
Assistant Professor / Dobelman Family Junior Chair
Department of Statistics / Rice University
http://had.co.nz/

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: Same color scale

Oscar Perpiñan
In reply to this post by alain.reymond
Hi,

Perhaps the levelplot method from rasterVis is useful to solve this
problem.

library(rasterVis)

f <- system.file("external/test.grd", package="raster")
r <- raster(f)

s <- stack(r, r-maxValue(r))

levelplot(s, par.settings=RdBuTheme)

If you need separate pages for each map you can use:

trellis.device(pdf, file='test.pdf')
levelplot(s, par.settings=RdBuTheme, layout=c(1, 1))
dev.off()

Best,

Oscar.


El Tue, 22 Nov 2011 14:56:31 +0100
[hidden email] escribió:

> Dear all,
> I want to compare 2 raster data map where probability values between
> 0 and 1 are not the same (different mix-max values). For this I want
> to apply to both map the same color scale (i.e. 0-0.1 red, 0.1-0.2
> blue etc.) in an automatic way without doing it individually for each
> map.
>
> I have tried many ways :
> - reclassifying the probability. But here the problem is if one class
> do not exist the color will shift and I cannot assign a particular
> color to a particular class to avoid this.
>
> - rewriting the breaks limits (sequence from 0 to 1 by 0.1) and using
> the RampPalette but it did not work either because of the max-value
> of the data (they are not the same and not in the same sequence) and
> one color will be add in the rampPalette to fill in the difference.
> It looks like as in the rampPalette there is like a defined min-color
> and max-color and it will add as many as needed between those 2.
> Therefore I wrote by hand some color for the sequence and it worked
> but I had to delet colors to adapt to the max number of really
> existing sequence. Hence, if I want to do this for a lot of files I
> can't make it that way.
>
> - I also used the ggplot, but again, there is a min-color and a
> max-color for the min and max values (i.e. blue-red). If one ggplot
> has more sequence than the other one, in both case they will have the
> blue and red as min and max color but then not the same number of
> color between those 2. And If I do not set a max-color (or any min
> and max color) it does not work either.
>
> I already thank you if you have any solution to my problem that could
> be used "easily" without having to write specific color for each map.
> Best,
>
> Alain
>
>
>
>
> =
> [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: Same color scale

manuel.schneider
In reply to this post by alain.reymond
 <alain.reymond <at> wsl.ch> writes:
>
> Dear all,
> I want to compare 2 raster data map where probability values between 0 and 1
are not the same (different
> mix-max values).
> For this I want to apply to both map the same color scale (i.e. 0-0.1 red,
0.1-0.2 blue etc.) in an automatic
> way without doing it individually for each map.

Dear Alain

not entirely sure whether I really understand your question, but maybe
hsv(prob,1,1) or hsv(round(prob,1),1,1) or rainbow(10)[ceiling(prob*10)] could
also do the job when using image() to plot

Abw

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo