Username:

Password:

Altair AstroDIO DehumidifiersAtik CamerasModern AstronomyDavid HindsNe3 Filters
Solar Gravity 2018

Author Topic: Pixinsight - Using Mosaic By CoOrdinates (Large Image Warning!)  (Read 938 times)

0 Members and 1 Guest are viewing this topic.

Offline chris.bailey

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 9672
  • If at last you do succeed, don't try again
    • Rosemount Observatory
There has been a bit of discussion on mosaics recently so I thought it would be worth posting some screenshots of a method I have used on my own images as well as some I have been sent. Note that although the method is pretty much automated, it is processor intensive and some steps can take an hour or more to complete.

The beauty of this method is that it does not rely on detection and matching of stars in overlap areas. Mosaics could (should you wish) be created with no overlaps whatsoever and it would still work (though you would have gaps in the final mosaic). Instead, each pane of the mosaic is plate solved so as to add WCS (World CoOrdinate System) data entries into the fits header so that pane is then in a known position. the Mosaic By CoOrdinates script then re-projects each pane onto a chosen projection to a size equal to the extent of the full mosaic. Finally the individual pane are merged Using Gradient Merge Mosaic to disguise any seems.

Before starting prepare each panel as if it were a standalone image. If the images have significant gradients it is worth removing these with DBE and I also carry out a round of linear noise reduction with Mure De-Noise being my current weapon of choice. It is also worth cropping a little off the borders, trying to avoid leaving a bright star on or near any crop.

Scripts>Image Analysis>Image Solver.

This can be a little finicky and it is important to get the image scale right and to seed the script with RA and Dec co-ordinates that are close to the panels to be solved. The search section normally works very well. I sometimes find that one or two images from a set will fail the first time around but will solve if retried. Setting the Noise threshold to 1 or 2 can help as can increasing the limiting magnitude from the default. Turn on distortion correction. The Mercator projection works well with most images though others will work better with very wide field images. Use the same projection here as used in the latter steps. The output will be an image set with WCS appended to the file name.
LX200|ZS70|FSQ85|FLT110|Altair DF250RC|EQ6 Pro(Rowan Belt Mod)|ParamountMX
ATIK383L+/EFW2/OAG|Lodestar|Baader 36mm LRGBHaSIIOIII
Starlight Express SXVRH16/ONAG/FW|Lodestar X2|Baader 2" Filters
Starlight Express SXVRH814/ONAG/FW|LodestarX2|Baader 2" Filters
Lunt LS60PTBF1200|DMK41|Quark Chromo
Samyang 135mm f1.8

Offline chris.bailey

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 9672
  • If at last you do succeed, don't try again
    • Rosemount Observatory
Re: Pixinsight - Using Mosaic By CoOrdinates (Large Image Warning!)
« Reply #1 on: 13:36:53, 01 February, 2016 »
Scripts>Utilities>Mosaic By CoOrdinates

Pick the newly created WCS images as inputs and the same projection as used in plate solving. Click on 'Calculate' and the script will calculate all the other settings needed. Under Quality choose High and I find it helps to reduce the star clamping threshold a little from the default. When you click on 'OK' there may well be a warning about creating a large image but click through this and go and make a coffee (or two). The output of the script will be a set of imaged with a _registered suffix.
LX200|ZS70|FSQ85|FLT110|Altair DF250RC|EQ6 Pro(Rowan Belt Mod)|ParamountMX
ATIK383L+/EFW2/OAG|Lodestar|Baader 36mm LRGBHaSIIOIII
Starlight Express SXVRH16/ONAG/FW|Lodestar X2|Baader 2" Filters
Starlight Express SXVRH814/ONAG/FW|LodestarX2|Baader 2" Filters
Lunt LS60PTBF1200|DMK41|Quark Chromo
Samyang 135mm f1.8

Offline chris.bailey

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 9672
  • If at last you do succeed, don't try again
    • Rosemount Observatory
Re: Pixinsight - Using Mosaic By CoOrdinates (Large Image Warning!)
« Reply #2 on: 13:40:50, 01 February, 2016 »
Process>Gradient Domain>Gradient Merge Mosaic

Use the newly created _registered images as input. I find it useful to increase both the Shrink and Feather values from the default to avoid artefacts around external corner margins and bright stars near to overlaps.

If all goes well you will have a mosaic with near invisible borders ready for cropping and subsequent processing.

LX200|ZS70|FSQ85|FLT110|Altair DF250RC|EQ6 Pro(Rowan Belt Mod)|ParamountMX
ATIK383L+/EFW2/OAG|Lodestar|Baader 36mm LRGBHaSIIOIII
Starlight Express SXVRH16/ONAG/FW|Lodestar X2|Baader 2" Filters
Starlight Express SXVRH814/ONAG/FW|LodestarX2|Baader 2" Filters
Lunt LS60PTBF1200|DMK41|Quark Chromo
Samyang 135mm f1.8

Offline Roger B

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 8997
    • Capricorn1 observatory
Re: Pixinsight - Using Mosaic By CoOrdinates (Large Image Warning!)
« Reply #3 on: 14:21:43, 01 February, 2016 »
This is the way you showed me Chris,  As you said the plate solving can be a bit funny,  I had to try quite a few times before I could get all 3 panes to solve.    To me that was the real PITA as each panel had different coordinates even though its on the same part of the sky.   

anyway it worked in the end and I was able to get my first mosaic..

Roger 
Scopes - TS65q, W/O 110 Triplet
Mount - EQ6 PRO via  Eqmod,  Rowan belt modded
Cameras - SX694, QSI583WSG,Lodestar Guidecam, GP-Cam, QHY5II.
TS 9mm OAG
Quantum 7x1.25 EFW, Baader LRGB,Ha,OIII,SII Filters 
Quark chromo

SGPro, Pixinsight.

************AND NO TIME TO USE IT************

 

ukbuysellRemote Imaging from AustraliaSharpSkyblank APTUKAI on Facebook
Powered by SMF 2.0.15 | SMF © 2006, Simple Machines LLC
DarkBreak by DzinerStudio. Theme modified by The UKAI Team

Page created in 0.268 seconds with 38 queries.
TinyPortal © 2005-2012