SHG (spectroheliograph) reconstruction software v4.3

The place for all your solar software; capture, stacking, post processing, ancillaries etc. It's all up for discussion here!
Post Reply
thesmiths
Almost There...
Almost There...
Posts: 1114
Joined: Mon Jun 04, 2012 8:13 pm
Location: London, England
Has thanked: 62 times
Been thanked: 1626 times
Contact:

SHG (spectroheliograph) reconstruction software v4.3

Post by thesmiths »

The Windows executable file can be found here: https://github.com/thelondonsmiths/Sole ... .3_Windows

The Windows executable will run on both Windows 10 and 11. It does not need Python to be installed on the computer. It should run the same as the Python source code but without the ability to make any changes to the code. For most people, this Windows executable version will be more convenient to use.

Compared to v4.2, this version has the following new features:
  • Save protus only (similar to Save clahe only)
  • Stubborn transversalium (to remove very strong transversalium defects)
  • De-vignette (vignetting compensation for uneven illumination along the slit)
  • Ellipse fit shift (change the shift from the default of 10 pixels if the dispersion is much larger than normal; this can improve the edge detection used in the ellipse fit)
The v4.3 user interface is shown below:

SHG v4.3.JPG
SHG v4.3.JPG (80.48 KiB) Viewed 1099 times


thesmiths
Almost There...
Almost There...
Posts: 1114
Joined: Mon Jun 04, 2012 8:13 pm
Location: London, England
Has thanked: 62 times
Been thanked: 1626 times
Contact:

Re: SHG (spectroheliograph) reconstruction software v4.3

Post by thesmiths »

The function called "Pixel Offset Live" has been renamed "Spectral Analyser".

The "Save protus" feature was added since the Protus image looks better with the transversalium turned off, while the Clahe image looks better with it turned on.

"Stubborn transversalium" can be used for very intense line defects. Since it can also cause artefacts, this function should be used sparingly (for example, to rescue data that otherwise would be spoiled). Ideally, it is better to try to clean dust off the slit so that a low setting of the normal transversalium correction is sufficient.

"Ellipse fit shift" does not normally need to be changed from the default value of 10 pixels. But if the dispersion is very high (for example, if using a higher power diffraction grating like 3600 l/mm) then a larger pixel shift is required to move the same wavelength distance. A value of 20 then will help the program with its edge detection, which is necessary to do ellipse fit.

The "de-vignette" function is a very useful new feature that attempts to equalise the brightness along the slit. For a number of reasons (typically optical misalignment or an obstructed light path), there can be either unequal darkening on the ends of the slit or there can be excessive darkening on both sides. This function tries to match the limb darkening along the slit with that perpendicular to the slit.

Below are two examples of "de-vignette". The first shows a mild darkening on the right side (the scan in vertical). The second shows a very strong vignetting on both ends of the slit, with the right side much darker. The "de-vignette" does an acceptable job of correcting even the very strong case, although ideally the source of the vignetting should be addressed.

"De-vignette" can potentially lead to excessive brightness at the edges of the sun and possibly also some slight brightness banding.

Example of a mild vignetting. Left: uncorrected. Right: with de-vignette.
Example of a mild vignetting. Left: uncorrected. Right: with de-vignette.
092940_H-alpha_de-vignette.jpg (267.26 KiB) Viewed 1096 times
Example of very strong vignetting. Left: uncorrected -- data almost unusable. Right: with de-vignette -- image passable.
Example of very strong vignetting. Left: uncorrected -- data almost unusable. Right: with de-vignette -- image passable.
075945_H-alpha_de-vignette.jpg (361.9 KiB) Viewed 1096 times


Post Reply