Welcome to iraf.net Monday, May 20 2024 @ 12:31 AM GMT
Anonymous: Guest |
07/07/1998 07:59PM (Read 341 times)
|
|
|
|
Dear IRAF -I am trying to use the Legendre coefficients generated by the identify task
to do other things with the spectra, but I find that when I try to
calculate the wavelengths from the pixel numbers and the coefficients, the
results are off by typically 5 Angstroms. I see the same effect if I try
to do the calculation for the lines that were used by identify, with the
same pixel numbers -- the results do not match those in the database file
(included below). We are running IRAF 2.10.2 on a Sun. The spectra are 512
pixels long. My guess is that there is something wrong with the way that I
am calculating the argument for the Legendre polynomials. I am using x = (2*pixel/512)-1where the pixel values run from 1 to 512. I'd be grateful for any insight
that you can provide.Bob Cadmus
Department of Physics
Grinnell College# Sun 17:08:22 21-Jun-98
begin identify ngc7469_123197s - Ap 1
id ngc7469_123197s
task identify
image ngc7469_123197s - Ap 1
aperture 1
aplow INDEF
aphigh INDEF
features 11
7.01 4669.4676 4669. 4.0 1 1 Na I
47.66 4980.988 4983. 4.0 1 1 Na I
69.92 5150.929 5149. 4.0 1 1 Na I
110.71 5460.4168 5461. 4.0 1 1 Hg I
126.10 5576.3537 5577. 4.0 1 1 O I
141.03 5688.3231 5688. 4.0 1 1 Na I
152.04 5770.4938 5770. 4.0 1 1 Hg I
154.91 5791.9059 5791. 4.0 1 1 Hg I
204.75 6159.1132 6161. 4.0 1 1 Na I
224.27 6300.6792 6300. 4.0 1 1 O I
232.98 6363.3297 6363. 4.0 1 1 O I
function legendre
order 4
sample *
naverage 1
niterate 0
low_reject 3.
high_reject 3.
grow 0.
coefficients 8
2.
4.
1.
512.
6483.323076353774
1782.369922983237
-95.47952349800706
-17.87575427121301
|
|
|
|
Anonymous: Guest |
07/07/1998 07:59PM
|
|
|
|
Hello Bob,The IDENTIFY help page can fill in all the details on this, but briefly
yes there is something wrong with your calculation for the argument
of the polynomials. You want to map the range (1,512) to the range (-1,1),
but yours actually maps to (-0.9961,1). Another way to look at it is that
you're mapping from (0,512) instead of (1,512). You want: (2 * pixel - ( 512 + 1 ) / (512 - 1)which can be written as: (2 x (pixel-1)/511) - 1Let me know if you have any other questions.Dave
|
|
|
|
| |
|