| Re: log scale of data coloring of IDLgrVolume object, not the axes [message #29021] |
Wed, 30 January 2002 08:38  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Todd Bowers (tbowers0@yahoo.com) writes:
> Ok, I think I've figured out the whole log axis thing to get an
> IDLgrSurface axis to show log scale, its all about scaling the
> xyzCoord_Conv property of the surface with the logged bounds of your
> data.
>
> But, I'm having trouble carrying this over using log scale for the
> coloring of an IDLgrVolume object and the associated Hcolorbar from
> David F. (keeping the xyz axes linear, just logging the color ramp).
> Could anyone help?
HColorbar isn't designed for log scales, of course,
but I think it is just a matter of passing it a
color palette of values that have been scaled by
some kind of log function. For example:
r = BytScl(Findgen(256)^10e-2)
g = r
b = r
thePalette = Obj_New('IDLgrPalette', r, g, b)
theColorbar = Obj_New('HColorbar', Palette=thePalette)
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: log scale of data coloring of IDLgrVolume object, not the axes [message #29249 is a reply to message #29021] |
Fri, 08 February 2002 08:49  |
1
Messages: 1
Registered: February 2002
|
Junior Member |
|
|
David Fanning <david@dfanning.com> wrote in message
>
> HColorbar isn't designed for log scales, of course,
> but I think it is just a matter of passing it a
> color palette of values that have been scaled by
> some kind of log function. For example:
>
> r = BytScl(Findgen(256)^10e-2)
> g = r
> b = r
> thePalette = Obj_New('IDLgrPalette', r, g, b)
> theColorbar = Obj_New('HColorbar', Palette=thePalette)
>
> r = BytScl(Findgen(256)^10e-2)
Why not alog10(Findgen(256)) ??
Thanks David. I had already tried this without success. The colors don't
really ramp properly for a general color table (like Rainbow, colortable
13). Even if it did, the colors would be squished so much on one end that
it'd be difficult to see what colors = what value.
Playing around, I did find that I get good coloring if I log the data as I
bytscl it for the volume object.
loadct, 34 ; rainbow color table
tvLCT, r,g,b, /GET
colorTable = [[r],[g],[b]] ; [256,3] color table for vol coloring
;data of float values in fData. IDLgrVolume's need data bytscl'd
bData = bytscl(alog10(fData), /NaN)
oVol = OBJ_NEW('IDLgrVolume', DATA0=bData, RGB_TABLE0=colorTable)
Now the data looks good, but my Hcolorbar still has text annotations that
range linearly from min(fData) to max(fData). I noticed that David's
hcolorbar is just a little colored IDLgrImage with IDLgrAxis. So, I thought
I'd just set the LOG keyword for the textAxis object in his code, maybe
this'll work:
;//in David's HColorBar code
textAxis = Obj_New("IDLgrAxis", 0, Color=self.color, Ticklen=0.025, $
Major=self.major, Minor=self.minor, Title=thisTitle, Range=self.range,
/Exact, $
XCoord_Conv=longScale, Location=[1000, self.position(1), 0.001], $
/LOG)
but this squooshes this axis all up to the left of where it should be.
Any suggestions?
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
