[STATSGRASS] spatstat: trend to spatialgriddataframe - help needed

[Jarekj Jasiewicz]

Roger Bivand napisał(a):
> On Thu, 10 May 2007, Jarekj Jasiewicz wrote:
>
>   
>> Hi
>>
>> It seems very simple but I cannot pass throuh one step:
>>
>> I computed trend (ppm) in spatstat package:
>>
>>     ppm_2=ppm(unmark(pp), ~polynom(x,y,2))
>>
>> next I found, that I can convert that  object to  "something almost 
>> like  <<im>> object" :
>>
>>     k=plot(ppm_2, ngrid=c(gmeta6()$cols, gmeta6()$rows), 
>> superimpose=FALSE, how="image", pause=FALSE, plot.it=FALSE)
>>
>> but: it is not im object. It is object which only contains im inside. 
>> Here is its structure:
>>
>> str(k)
>> List of 1
>>  $ trend:List of 1
>>   ..$ :List of 11     ##here is something like "unnamed atribute" I 
>> cannot pass this
>>     
>
> Try str(k$trend[[1]]$v) - if this accesses the matrix with the [[ operator
> (first unnamed component of trend), then the other bits ought to follow.
> Please note that polynom() does not protect the powered coordinates from
> numerical overflow, so I'd be cautious unless you know that x and y are
> only small numbers.
>
> Roger
>
>   
>>   .. ..$ v     : num [1:986, 1:1051] 1.06e-09 1.07e-09 1.08e-09 1.10e-09 
>> 1.11e-09 ...
>>   .. ..$ dim   : int [1:2] 986 1051
>>   .. ..$ xrange: num [1:2] 3684409 3713995
>>   .. ..$ yrange: num [1:2] 5649780 5681325
>>   .. ..$ xstep : num 28.2
>>   .. ..$ ystep : num 32
>>   .. ..$ xcol  : num [1:1051] 3684423 3684451 3684479 3684507 3684535 ...
>>   .. ..$ yrow  : num [1:986] 5649796 5649828 5649860 5649892 5649924 ...
>>   .. ..$ lev   : NULL
>>   .. ..$ type  : chr "real"
>>   .. ..$ units :List of 3
>>   .. .. ..$ singular  : chr "unit"
>>   .. .. ..$ plural    : chr "units"
>>   .. .. ..$ multiplier: num 1
>>   .. .. ..- attr(*, "class")= chr "units"
>>   .. ..- attr(*, "class")= chr "im"
>>  - attr(*, "class")= chr "plotppm"
>>  - attr(*, "mrkvals")= num 1
>>
>>
>> I simply don't know how extract that part which is im object (to convert 
>> it next to spatialgriddataframe) or even only v (v I can insert to any 
>> im object and next to convert it)
>>
>> I hope that problem is trivial to more advanced R users
>>
>> Jarek
>>
>>
>>
>>
>> _______________________________________________
>> statsgrass mailing list
>> statsgrass at grass.itc.it
>> http://grass.itc.it/mailman/listinfo/statsgrass
>>
>>     
>
>   
thanks, it works. As I thought It was trivial :)

thanks also for informations of overflow. as you see x and y are rather 
huge, I tested trend up to fourth polynominal, without overflow, the 
numbers rather are in double number range


Jarek