1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
|
.help instruments Dec93 noao.imred.ccdred
.ih
NAME
instruments -- Instrument specific data files
.ih
DESCRIPTION
The \fBccdred\fR package has been designed to accommodate many different
instruments, detectors, and observatories. This is done by having
instrument specific data files. Note that by instrument we mean a
combination of detector, instrument, application, and observatory, so
there might be several "instruments" associated with a particular CCD
detector. Creating and maintaining the instrument files is generally
the responsibility of the support staff, though the user may create or
copy and modify his/her own instrument/application specific files. The
task \fBsetinstrument\fR makes this information available to the user
and package easily.
There are three instrument data files, all of which are optional. The
package may be used without the instrument files but much of the
convenience of the package, particularly with respect to using the CCD
image types, will be lost. The three files are an instrument image
header translation file, an initialization task which mainly sets
default task parameters, and a bad pixel file identifying the cosmic
bad pixels in the detector. These files are generally stored in a
system data directory which is a subdirectory of the logical
directory "ccddb$". Each file has a root name which identifies the
instrument.
.sh
1. Instrument Translation File
The instrument translation file translates the parameter names used by
the \fBccdred\fR pacakge into instrument specific parameters and also
supplies instrument specific default values. The package parameter
\fIccdred.instrument\fR specifies this file to the package. The task
\fBsetinstrument\fR sets this parameter, though it can be set
explicitly like any other parameter. For the standard instrument
translation file the root name is the instrument identification and the
extension is "dat" ("*.dat" files are protected from being removed in a
"stripped" system, i.e. when all nonessential files are removed).
Private instrument files may be given any name desired.
The instrument translation proceeds as follows. When a package task needs
a parameter for an image, for example "imagetyp", it looks in the instrument
translation file. If the file is not found or none is specified then the
image header keyword that is requested has the same name. If an
instrument translation file is defined then the requested
parameter is translated to an image header keyword, provided a translation
entry is given. If no translation is given the package name is used. For
example the package parameter "imagetyp" might be translated to "data-typ"
(the old NOAO CCD keyword). If the parameter is not found then the default
value specified in the translation file, if present, is returned. For recording
parameter information in the header, such as processing flags, the
translation is also used. The default value has no meaning in this case.
For example, if the flag specifying that the image has been corrected
by a flat field is to be set then the package parameter name "flatcor"
might be translated to "ff-flag". If no translation is given then the
new image header parameter is entered as "flatcor".
The format of the translation file are lines consisting of the package
parameter name, followed by the image header keyword, followed by the
default value. The first two fields are parameter names. The fields
are separated by whitespace (blanks and tabs). String default values
containing blanks must be quoted. An example is given below.
.nf
# Sample translation file.
exptime itime
darktime itime
imagetyp data-typ
subset f1pos
biassec biassec [411:431,2:573]
datasec datasec [14:385,2:573]
fixpix bp-flag 0
overscan bt-flag 0
zerocor bi-flag 0
darkcor dk-flag 0
flatcor ff-flag 0
fringcor fr-flag 0
.fi
The first comment line is ignored as are blank lines.
The first two lines translate the CCD image type, and the subset parameter
without default values (see \fBccdtypes\fR and \fBsubsets\fR for more
information). The next two lines give the overscan bias strip
section and the data section with default values for the instrument.
Note that these parameters may be overridden in the task \fBccdproc\fR.
The next set of translations requires further discussion. For processing
flags the package assumes that the absence of a keyword means that the
processing has not been done. If processing is always to be done with
the \fBCCDRED\fR package and no processing keywords are recorded in the raw data
then these parameters should be absent (unless you don't like the names
used by the package). However, for compatibility with the original NOAO
CCD images, which may be processed outside of IRAF and which use 0 as the
no processing value, the processing flags are translated and the false values
are indicated by the default values.
If there is more than one translation for the same CCDRED parameter,
for example more than one exptime, then the last one is used.
In addition to the parameter name translations the translation file
contains translations between the value of the image type parameter
and the image types used by the package. These lines
consist of the image header type string as the first field (with quotes
if there are blanks) and the image type as recognized by the package. The
following example will make this clearer.
.nf
'OBJECT (0)' object
'DARK (1)' dark
'PROJECTOR FLAT (2)' flat
'SKY FLAT (3)' other
'COMPARISON LAMP (4)' other
'BIAS (5)' zero
'DOME FLAT (6)' flat
.fi
The values of the image type strings in the header contain blanks so they
are quoted. Also the case of the strings is important. Note that there
are two types of flat field images and three types of object images.
The CCD image types recognized by the package are:
.nf
zero - zero level image such as a bias or preflash
dark - dark count image
flat - flat field image
illum - iillumination image such as a sky image
fringe - fringe correction image
object - object image
.fi
There may be more than one image type that maps to the same package
type. In particular other standard CCD image types, such as comparison
spectra, multiple exposure, standard star, etc., should be mapped to
object or other. There may also be more than one type of flat field,
i.e. dome flat, sky flat, and lamp flat. For more on the CCD image
types see \fBccdtypes\fR.
The complete set of package parameters are given below.
The package parameter names are generally the same as the
standard image header keywords being adopted by NOAO.
.nf
General Image Header and Default Parameters
ccdmean darktime exptime fixfile
imagetyp ncombine biassec subset
title datasec nscanrow
CCDRED Processing Flags
ccdproc darkcor fixpix flatcor
fringcor illumcor overscan trim
zerocor
CCDRED CCD Image Types
dark flat fringe illum
none object unknown zero
.fi
The translation mechanism described here may become more
sophisticated in the future and a general IRAF system facility may be
implemented eventually. For the present the translation mechanism is
quite simple.
.sh
2. Instrument Setup Script
The task \fBsetinstrument\fR translates an instrument ID into a
CL script in the instrument directory. This script is then executed.
Generally this script simply sets the task parameters for an
instrument/application. However, it could do anything else the support
staff desires. Below are the first few lines of a typical instrument setup
script.
.nf
ccdred.instrument = "ccddb$kpno/example.dat"
ccdred.pixeltype = "real"
ccdproc.fixpix = yes
ccdproc.overscan = yes
ccdproc.trim = yes
ccdproc.zerocor = no
ccdproc.darkcor = no
ccdproc.flatcor = yes
ccdproc.biassec = "[411:431,2:573]"
ccdproc.datasec = "[14:385,2:573]"
.fi
The instrument parameter should always be set unless there is no
translation file for the instrument. The \fBccdproc\fR parameters
illustrate setting the appropriate processing flags for the
instrument. The overscan bias and trim data sections show an alternate
method of setting these instrument specific parameters. They may be
set in the setup script in which case they are given explicitly in the
user parameter list for \fBccdproc\fR. If the value is "image" then
the parameters may be determined either through the default value in
the instrument translation file, as illustrated in the previous
section, or from the image header itself.
The instrument setup script for setting default task parameters may be
easily created by the support person as follows. Set the package
parameters using \fBeparam\fR or with CL statements. Setting the
parameters might involve testing. When satisfied with the way the
package is set then the parameters may be dumped to a setup script
using the task \fBdparam\fR. The final step is editing this script to
delete unimportant and query parameters. For example,
.nf
cl> dparam ccdred >> file.cl
cl> dparam ccdproc >> file.cl
cl> dparam combine >> file.cl
...
cl> ed file.cl
.fi
.sh
3. Instrument Bad Pixel File
The bad pixel file describes the bad pixels, columns, and lines in the
detector which are to be replaced by interpolation when processing the
images. This file is clearly detector specific. The file consists of
lines describing rectangular regions of the image.
The regions are specified by four numbers giving the starting and ending
columns followed by the starting and ending lines. The starting and
ending points may be the same to specify a single column or line. The
example below illustrates a bad pixel file.
.nf
# RCA1 CCD untrimmed
25 25 1 512
108 108 1 512
302 302 403 512
1 512 70 70
245 246 312 315
.fi
If there is a comment line in the file containing the word "untrimmed"
then the coordinates of the bad pixel regions apply to the original CCD
detector coordinates.
If the image has been trimmed and the bad pixels are replaced at a later
stage then this word indicates that the trim region be determined from the
image header and the necessary coordinate conversion made to the original
CCD pixel coordinates. Note that if a subraster readout is used the
coordinates must still refer to the original CCD coordinates and
not the raw, untrimmed readout image. If the word
"untrimmed" does not appear then the coordinates are assumed to apply to
the image directly; i.e. the trimmed coordinates if the image has been
trimmed or the original coordinates if the image has not been trimmed.
The standard bad pixel files should always refer to the original, untrimmed
coordinates.
The first two bad pixel regions are complete bad columns (the image
is 512 x 512), the next line is a partial bad column, the next line is
a bad line, and the last line is a small bad region. These files are
easy to create, provided you have a good image to work from and a way
to measure the positions with an image or graphics display.
.ih
SEE ALSO
ccdtypes, subsets, setinstrument
.endhelp
|
