diff options
Diffstat (limited to 'pkg/plot/doc/crtpict.hlp')
| -rw-r--r-- | pkg/plot/doc/crtpict.hlp | 171 |
1 files changed, 171 insertions, 0 deletions
diff --git a/pkg/plot/doc/crtpict.hlp b/pkg/plot/doc/crtpict.hlp new file mode 100644 index 00000000..9ecb3b4e --- /dev/null +++ b/pkg/plot/doc/crtpict.hlp @@ -0,0 +1,171 @@ +.help crtpict Aug87 plot +.ih +NAME +crtpict -- make a hardcopy of an IRAF image +.ih +USAGE +crtpict input +.ih +PARAMETERS +.ls input +Input images to be processed. +.le +.ls device = "dicomed" +The output device. +.le +.ls auto_fill = yes +If set to yes, the image will be scaled to fit the device viewport. +The aspect ratio is always preserved when \fIauto_fill\fR = yes. +.le +.ls xmag = 1.0, ymag = 1.0 +When \fIauto_fill\fR = no, the x and y magnification ratios are specified +by these parameters. +.le +.ls replicate = yes +The image pixels are block replicated to fit the device viewport when +\fIreplicate\fR = yes. Otherwise, the pixels are linearly interpolated +to match the device pixels. +.le +.ls x_block_avg = 1, y_block_avg = 1 +These parameters are used when \fIreplicate\fR = no to decrease the +effective output device resolution, and speed up the interpolation. The +pixels are interpolated to the block averaged output device, then +block replicated to fill the device viewport. +.le +.ls ztrans = "auto" +This parameter specifies how the image intensities are mapped into the +greyscale values of the output device. Intensity z1 maps to black, z2 to white. +The 4 choices for \fIztrans\fR are: +.nf + + "auto" - z1 and z2 centered on median of image + "min_max" - set z1 and z2 to specified intensities + "none" - truncate intensities to fit output range + "user" - user supplies look up table of values +.fi +.le +.ls lutfile = "" +Name of text file containing the look up table when \fIztrans\fR = user. +The table should contain two columns per line; column 1 contains the +intensity, column 2 the desired greyscale output. +.le +.ls contrast = 0.25 +Used when automatically determining z1 and z2. The slope of the transfer +function is divided by \fIcontrast\fR, so negative values of \fIcontrast\fR +result in a negative transfer function. +.le +.ls nsample_lines = 25 +Used when automatically determining z1 and z2, this parameter sets the number +of image lines to be sampled when determining the median. +.le +.ls z1 = 0.0, z2 = 0.0 +These parameters are used when \fIztrans\fR = "min_max", to specify which +pixel values map to black and white. +.le +.ls perimeter = yes +Draw annotated axes around the plot perimeter? +.le +.ls image_fraction = 0.70 +The fraction of the vertical device viewport reserved for the image. +.le +.ls graphics_fraction = 0.20 +The fraction of the vertical device viewport reserved for histogram +plots and id information. +.le +.ls greyscale_fraction = 0.05 +The fraction of the vertical device viewport reserved for the greyscale +step wedge. +.le +.ls output = "" +Output metacode is appended to this file. +By naming an output file, the metacode can be "trapped", and the normal +spooling process intercepted. +.le +.ih +DESCRIPTION +Procedure \fBcrtpict\fR makes a photographic hardcopy plot of IRAF images. + +The image can be automatically scaled to fill the output plotting window, with +the aspect ratio preserved, by setting \fBauto_fill\fR = yes. When +\fBauto_fill\fR = no, magnification factors for the axes are entered as +\fBxmag\fR and \fBymag\fR, where negative values (as well as fractional +values < 1.0), indicate that the image is to be reduced. By default, the +imaged is enlarged by block replication. By setting \fBreplicate\fR = no, +the image will be linearly interpolated to fit the device area. (In this +case, to speed things up, the \fBblock_avg\fR parameters can be set to +reduce the effective output resolution.) In either case, if an image needs +to be reduced in size, it will be decimated. + +Four methods of determining the greyscale transformation are available. +When \fIztrans\fR = "none", no transformation between intensity and +greyscale level occurs, the intensities are simply copied, which will most +likely result in truncation. With this method, the lowest bits of each pixel, +the lowest level variations, are always shown, regardless of the dynamic +range of the image. + +When \fIztrans\fR = "auto", +the greyscale levels are automatically centered on the median of the image +pixels. The window of intensities spanned by the greyscale is controlled +by parameter \fIcontrast\fR, which is divided into the calculated slope of +the transfer function. The larger the absolute value of \fIcontrast\fR, the +higher the contrast in the output image. A subset of the image pixels are +used to determine the median; the number of lines sampled is +\fInsample_lines\fR. + +When \fBztrans\fR = "min_max", intensity \fBz1\fR maps to the minimum +greyscale level (black), \fBz2\fR maps to the maximum greyscale level +(white) and the transfer function is linear in between these two endpoints. +If \fIz1\fR = \fIz2\fR, the image min and max map to black and white, modified +by \fBcontrast\fR. (NOTE: When running \fIcrtpict\fR on an image created with +\fIsnap\fR, \fBztrans\fR should be set to "min_max", with \fBz1\fR = 0 and +\fBz2\fR = 1023, the maximum output value possible from the IIS.) + +When \fBztrans\fR = "user", a look up table of intensity values and their +corresponding greyscale levels is read from the file specified by the +\fBlutfile\fR parameter. From this information, +\fIcrtpict\fR constructs a piecewise linear look up table containing +4096 discrete values. +The text format table contains two columns per line; +column 1 contains the intensity, column 2 the desired greyscale output. +The greyscale values specified by the user must match those available on +the output device. Task \fBshowcap\fR can be used to determine the range +of acceptable greyscale levels. +.ih +EXAMPLES +1. To subsample every 4th pixel of a large image, fill the output area and use +previously determined values of z1 and z2 for the greyscale transformation +the command would be: + + cl> crtpict sunpic[*:4,*:4] ztrans=min z1=0 z2=800 + +2. To process every image with the root name ccdpic, using default values of +all parameters, the command would be: + + cl> crtpict ccdpic* + +3. To process images created with \fBsnap\fR, ztrans and z2 must be changed +from their default values: + + cl> crtpict iis.snap ztrans=min z2=1023 + +4. Image `mypic' is processed using the look up table in file `mylut', + + cl> crtpict mypic ztrans=user lutfile=mylut + +Where file `mylut' contains this information: +.nf + 10 40 + 1500 100 + 2500 100 + 3500 200 + 7500 255 +.fi +.ih +TIMING +For a 512 x 512 real image, \fBcrtpict\fR takes about 40 cpu seconds with +\fBauto_fill\fR and \fBreplicate\fR = yes. When \fBauto_fill\fR = yes +but \fBreplicate\fR = no, \fBcrtpict\fR requires almost 400 cpu seconds. +.ih +SEE ALSO +display, showcap +.endhelp |