| Re: Printing Images [message #2632] |
Mon, 08 August 1994 09:53 |
sbarrkum
Messages: 5
Registered: August 1994
|
Junior Member |
|
|
In article <31qsi4$33v@reznor.larc.nasa.gov> tai@rockola.larc.nasa.gov (Alan Tai) writes:
> I'm having some problems printing an image to a Postscript device. By
> 12000 byte array. Does anyone have any suggestions? Thanks.
I use a little routine to rescale the pixels according to the device.
PS devices are capbale of high pixel density but the pixels can be rescaled.
Hence can makedo with less pixels.
pro mak_plot, x1,y1,x2,y2
; x1, y1 Bottom left Normalised coordinates
; x2, y2 Top right Normalised coordinates
data=fltarr(26,14) ; array to hold the image
;Plot image
arr_size=convert_coord(x2-x1+(x2-x1)*2/24,y2-y1+(y2-y1)*2/12 ,/Normal, /To_device)
if arr_size(0) gt 1000.0 then arr_size(0)=26.0 * 50.0 ; If array to large when
if arr_size(1) gt 700.0 then arr_size(1)=14.0 * 20.0 ; using PS device use smaller dimnsns
img=congrid(data,arr_size(0),arr_size(1))
if arr_size(0) ge arr_size(1) then smooth_width =0.1 * arr_size(1)
if arr_size(0) le arr_size(1) then smooth_width =0.1 * arr_size(0)
img=smooth(img, smooth_width)
tv, bytscl(img, tmin, tmax, top=255),x1, y1, /Normal, $
xsize = x2-x1, ysize = y2-y1
;Plot lat long grid and Manus coast line
plot,[140.5, 164.5], [-5.5,6.5], /Nodata, Xrange=[140.5, 164.5],YRange = [-5.5,6.5], $
/NOERASE, Linestyle=0, thick =3, XStyle=1 , Ystyle=1, color=backcolor, $
position=[x1,y1,x2,y2], charsize=0.6
end
===============
Hope this helps
Barr-Kumarakulasinghe
|
|
|
|
| Re: Printing Images [message #2644 is a reply to message #2632] |
Thu, 04 August 1994 09:50  |
sjt
Messages: 72
Registered: November 1993
|
Member |
|
|
Alan Tai (tai@rockola.larc.nasa.gov) wrote:
: I'm having some problems printing an image to a Postscript device. By
: using the same dimensions for the image as the X display (240 x 260),
: the plot comes out larger than it should be (weird--I kind of expected
: it to be a lot smaller). I tried re-scaling the image for the
: postscript device but ended up trying in vain to congrid a 12000 x
: 12000 byte array. Does anyone have any suggestions? Thanks.
: Alan
Let's suppose for the sake of argument you want your image to come out
10cm square on the print device and that it is in the array image, then a
suitable piece of code might be:
if ((!d.flags and 1) eq 1) then tv, image, xsize=10., ysize=10, /centi $
else tv, image
This checks if the device has scalable pixels (I think PS is currently
the only one that does, but it's as well to be general) [N.B. be careful
about the parentheses when working with !d.flags in this case it's O.K.
but for others missing them out will mess up precedences], and if the
pixels are scalable then it plots the image 10cm square, otherwise it
just plots it normally. The problem you were having is that for scalable
devices an image is by default plotted as big as possible.
--
James Tappin, School of Physics & Space Research
University of Birmingham
sjt@xun8.sr.bham.ac.uk
"If all else fails--read the instructions!"
O__
-- \/`
|
|
|
|
| Re: Printing Images [message #2646 is a reply to message #2644] |
Thu, 04 August 1994 09:24 |
n9140397
Messages: 13
Registered: July 1994
|
Junior Member |
|
|
In article <31qsi4$33v@reznor.larc.nasa.gov> tai@rockola.larc.nasa.gov (Alan Tai) writes:
> I'm having some problems printing an image to a Postscript device. By
> using the same dimensions for the image as the X display (240 x 260),
> the plot comes out larger than it should be (weird--I kind of expected
If you are saying device,xsize=240,ysize=260, then, yeah. I believe
that the default units for PS are cms!
That in itself should be useful, but, to fill up a page, try
set_plot,'PS'
device,/inches,xoffset=1, yoffset=1
device,/inches,xsize=6.5, ysize=9
This sets up a standard 8.5"x11" paper with one inch margins. You
can figure out the sizing you need for your case. Note that I
believe specifying landscape will exchange the x and y axes.
Hope that helps.
Mike H.
UCSD SIO CRD
|
|
|
|
| Re: Printing Images [message #2649 is a reply to message #2646] |
Thu, 04 August 1994 09:30 |
zawodny
Messages: 121
Registered: August 1992
|
Senior Member |
|
|
In article <31qsi4$33v@reznor.larc.nasa.gov> tai@rockola.larc.nasa.gov (Alan Tai) writes:
> I'm having some problems printing an image to a Postscript device. By
> using the same dimensions for the image as the X display (240 x 260),
> the plot comes out larger than it should be (weird--I kind of expected
> it to be a lot smaller). I tried re-scaling the image for the
> postscript device but ended up trying in vain to congrid a 12000 x
> 12000 byte array. Does anyone have any suggestions? Thanks.
>
>
> Alan
Congrid the image to the size that is appropriate for the amount of detail
you wish to retain in the image (240x260 sounds like a good starting point).
Then use the fact that PostScript has scaleable pixels and TV the image using
the xsize, ysize, & /NORMAL keywords to scale the image correctly. This will
probably entail some extra code (mostly CONVERT_COORD calls) to calculate
the correct values of xsize and ysize (as well as the x and y parameters to the
TV call). I once tried to use the /DATA key to TV but this did not work as
expected (maybe it does now). It also did not work when I was using log axies.
--
Joseph M. Zawodny (KO4LW) NASA Langley Research Center
Internet: zawodny@arbd0.larc.nasa.gov MS-475, Hampton VA, 23681-0001
TCP/IP: ko4lw@ko4lw.ampr.org Packet: ko4lw@n4hog.va.usa
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
