Transform_Process_Set Plots

% cd /home/lbto/data/20200108/lfbg/plot   (or just % cd plot)
% ./lfbg.csh

Three plots will be displayed:
  • Transformation Errors -vs- LUCI Image Number
  • Transformation Errors -vs- AGw Coordinates
  • Source Postion on LUCI Images

Transformation Errors -vs- LUCI Image Number

This plot shows the error in the new AGw transformation relative to the needed AGw coordinate required to place the source measured in each LUCI image at the on-axis position. In this plot, if a point in either X or Y indicates that the transformation error for a specific datum is larger than expected (larger than 3 or even 2 sigma), the LUCI image should be viewed to see if there are problmems. Problems could include poor image quality or telescope vibrations.

If it has been determined that a datum should be removed then the steps to recalculate the new transformation is:
  • edit the file lfbg.log and remove the entry line with the datum you which to remove. Note: the line can be move to the top of the file and a comment character (#) can be placed in the first column. Also, a comment as to the reason for removal can be added. Be sure to save the file.
  • execute the command IDL> transform_process_set, 'lfbg.log', /pt. Note, the /display flag can be added to the IDL command to see the LUCI data displayed. This execution of the transform_process_set is much faster because the reduction of the LUCI images to find the source postion and the pointing mask postion because this information is saved in IDL .sav files. However, the full tranformation analysis is performed with the new list of images. Re-running the analysis will overwrite all the previous results files. If you wish to old results files for later comparison you should copy them to new file name.
  • execute the shell script ./lfbg.csh to create the plots showing the results with the modified input image list

To view a LUCI image that has been background subtracted:
% ds9 ../reduced/luci1.20200108.0025,fits &
Note the name of the LUCI image is (luci1/luci2).(date).(image number).fits. The date is displayed in the plot subtitle (yyyymmdd) and the x-axis of the plot is the LUCI image number (note the image number has got be 4 digits long so 0's may need to be added to the beginning of the image number.

The Y range of this plot has been set to [-0.35,0.35] so that plots from results from different transformation analysis can be directly compared. These default limits can be changed in the IDL routines that create the XMGRACE (a plotting package) commands which makes this plot.

In the plot below the datum from luci1.20200108.0025.fits has an error in X of about ~0.21, which is a fairly high error, could be removed and the transform analysis rerun. Note, since there are several errors of around 0.19 I would sugggest that removal of this single point would not change the transformation and the RMS of the error would not decrease much.

  • Transformation Errors -vs- LUCI Image Number:
    lfbg_sigma.jpg

Transformation Errors -vs- AGw Coordinates

Like the plot above this plot shows the error in the new AGw transformation but as a function of the postion of the the guide star in the focal plane. These plots may show if the new transformation fit errors have a dependence on either the X or Y position in the focal plane. The cause of such dependence may be from a few outlying datum. The remedy may not be straight forward.

If it is determined that a datum should be removed then follow the itemized list above to recalculate the transform

  • Transformation Errors -vs- AGw Coordinates:
    lfbg_errors.jpg

Source Postion on LUCI Images

This plot shows the postion of the source in the LUCI image. Note, the correction for the pointing mask position has been included. The size of points is propostional to the measured FWHM of the source. This plot can show if the IQ of one or more of the images is poor enough to require the image removed from the transformation analysis.

This plot can also include data from another tranformation data set. In order to add another set, add the date (yyymmdd) as an argument of the script ./lfbg.csh
% ./lfbg.csh 20200109
Note: the other data set must have been analysed in order for the need data to be available.

  • Source Postion on LUCI Images:
    lfbg_measured.jpg

-- DougMiller - 24 Jan 2020
I Attachment Action Size Date Who Comment
lfbg_errors.jpgjpg lfbg_errors.jpg manage 157 K 24 Jan 2020 - 22:42 DougMiller  
lfbg_measured.jpgjpg lfbg_measured.jpg manage 135 K 30 Jan 2020 - 18:00 DougMiller  
lfbg_sigma.jpgjpg lfbg_sigma.jpg manage 200 K 24 Jan 2020 - 22:42 DougMiller  
lfbg_sigma.pdfpdf lfbg_sigma.pdf manage 4 K 24 Jan 2020 - 22:39 DougMiller  
Topic revision: r5 - 30 Jan 2020, DougMiller
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback