Re: Array manipulation

["Vince Hradil" <hradilv(at)yahoo.com>]

How's this:

IDL> a = indgen(30,30)
IDL> r=reform(a,3,10,3,10)
IDL> print, a[0:2,0:2], r[*,0,*,0]
       0       1       2
      30      31      32
      60      61      62
       0       1       2

      30      31      32

      60      61      62

IDL> r2 = reform(a,5,6,5,6)
IDL> print, r2[*,0,*,0]
       0       1       2       3       4

      30      31      32      33      34

      60      61      62      63      64

      90      91      92      93      94

     120     121     122     123     124



"Leon Majewski" <majewski@cygnus.uwa.edu.au> wrote in message
39ffa4ae.1813674@news.uwa.edu.au">news:39ffa4ae.1813674@news.uwa.edu.au...
> Hello
> I was wondering whether any array minded person could suggest a way of
using array
> indicies to chop up a large array into ordered windows.
> I can't think of a way to do it with reform, translate (though i'm sure
this is my
> limitation not a reform translate limitation)
>
> ---------
> ie given an array of 30*30 elements
> return 100 3*3 elements
> or 36 5*5
> or...
>
> in=
>
> 00 01 02 03 04 05..
> 30 31 32 33 34 35..
> 60 61 62 63 64 65..
>
> out = blocks such as
> 00 01 02
> 30 31 32
> 60 61 62
>
> each block is then processed to one representative number (ie mean or
median....) and
> returned
>
> ------------
> What i've used so far is attached below (it does what i want, just slowly)
>
>
> leon
>
>
>
> FUNCTION QMean2, DATA, $
> WIDTH=WIDTH, $
> MINVAL=MINVAL, $
> MAXVAL=MAXVAL, $
> SILENT=SILENT
>
> ;+
> ;NAME:
> ; QMean2
> ;
> ;PURPOSE:
> ; Calculate the mean of an array, excluding values 0 and NaN.
> ;
> ;CATEGORY:
> ; Array Manipulation
> ; Statistics
> ;
> ;CALLING SEQUENCE:
> ; Result = QMean(DATA)
> ;
> ;ARGUMENTS:
> ; DATA
> ; A 2-dimensional array to be averaged
> ;
> ;KEYWORDS:
> ; MAXVAL
> ; If the data is above this value it is omitted.
> ; MINVAL
> ; Only data greater than this value is included in the mean.
> ; If this keyword is not set, it defaults to 0.
> ; WIDTH
> ; The width of the averaging box. If this keyword is not set then
> ; it defaults to a value of 4.
> ;
> ;OUTPUTS:
> ; An array holding the average of the input data.
> ; This array is of size (Xdim/Width, Ydim/Width).
> ;
> ;SIDE EFFECTS;
> ; If the size of the array holding the data is not a multiple of the
> ; Width of the filter, then the Array is truncated to a size that is a
> ; Multiple of the Width - The program Outputs a warning if this is the
> ; case.
> ;
> ;EXAMPLES:
> ; Result = QMean(findgen(20,20), Width = 5)
> ; Result = QMean(findgen(20,20), Width = 5)
> ; Result = QMean(findgen(20,20), Width = 15, MAXVAL = 210.3)
> ;
> ;MODICIFACTION HISTORY:
> ; Created, Leon Majewski, 3rd August 2000
> ;
> ;-
>
> IF Size(MINVAL, /n_elements) EQ 0 THEN MINVAL = 0
> IF Size(WIDTH, /n_elements) EQ 0 THEN WIDTH = [4,4]
> IF Size(WIDTH, /n_elements) EQ 1 THEN WIDTH = [WIDTH,WIDTH]
> IF NOT KEYWORD_SET(SILENT) THEN SILENT = 1 ELSE SILENT = 0
> ;start_time = SYSTIME(1)
>
> DATA_IN = DATA
> Size_Data = SIZE(DATA_IN)
>
> xdim = Size_Data[1]/Width[0]
> xdim_less1 = xdim-1
> ydim = Size_Data[2]/Width[1]
> ydim_less1 = ydim-1
>
> IF Size_Data[1] NE xdim*width[0] THEN BEGIN
> if silent then begin
> print, 'The x dimension is not a multiple of the specified WIDTH.'
> print, 'Reducing the size of the array from'
> help,  DATA_IN
> endif
>
> DATA_IN = DATA_IN[0:(xdim)*width[0]-1, *]
>
> if silent then help,  DATA_IN
> ENDIF
>
> IF Size_Data[2] NE ydim*width[1] THEN BEGIN
> if silent then begin
> print, 'The y dimension is not a multiple of the specified WIDTH.'
> print, 'Reducing the size of the array from'
> help,  DATA_IN
> endif
>
> DATA_IN = DATA_IN[*,0:(ydim)*width[1]-1]
>
> if silent then help,  DATA_IN
> ENDIF
>
> IF KEYWORD_SET(MAXVAL) THEN $
> IF MAXVAL GT MINVAL THEN $
> DATA_IN = DATA_IN*(DATA_IN lt MAXVAL)
>
> Size_Data = SIZE(DATA_IN)
> n_els = Size_Data[4]
>
> DATA_IN = REFORM(DATA_IN, Width[0], n_els/Width[0], /overwrite)
> DATA_IN = REFORM(DATA_IN, Width[0], xdim, Width[1], ydim, /overwrite)
>
> Average = FLTARR(xdim,ydim)
>
> FOR j = 0, ydim_less1 DO BEGIN
> FOR i = 0, xdim_less1 DO BEGIN
>
> data_sub = REFORM(DATA_IN[*,i,*,j])
> goodpos = WHERE(data_sub GT MINVAL AND $
> FINITE(data_sub) EQ 1, c_goodpos)
>
> IF c_goodpos NE 0 THEN BEGIN
> data_sub = data_sub[goodpos]
> Average[i,j] = TOTAL(data_sub)/N_ELEMENTS(data_sub)
> ENDIF ELSE Average[i,j]=0
>
> ENDFOR
> ENDFOR
>
> ; print, SYSTIME(1) - start_time, 's'
>
> RETURN, Average
> END
> -------------------------
> Leon Majewski
>
> Remote Sensing & Satellite Research Group
> Curtin University of Technology, Perth, Australia
>
> email: majewski@ses.curtin.edu.au