next up previous contents
Next: Getting the Dispersion Solution Up: Geometric Correction Previous: Geometric Correction

Detecting and Identifying Arc Lines

For optimum results, it is important to use a comparison spectrum with as high a signal-to-noise ratio of the lines as possible. It is therefore advisible to flat-field  also the arc spectrum in order to correct for small-scale fluctuations. Another good idea is to filter the frame along the slit using the command FILTER/MEDIAN with a rectangular filter window of one pixel in the dispersion direction and several pixels in the perpendicular direction. For the row-by-row method one could also consider smoothing the spectra along the slit axis which would provide for a stronger coupling between neighbouring rows and thereby yield a solution which is intermediate between the extremes presented by the two pure methods.


next up previous contents
Next: Getting the Dispersion Solution Up: Geometric Correction Previous: Geometric Correction
http://www.eso.org/midas/midas-support.html
1999-06-15