Stacking centered on comet nucleus

[admin]

Hi Jean-Francois,

yes, the tuning is quite important here. To detect the comet, SharpCap needs to separate the image into three parts

1) Background
2) Stars
3) Comet

We need to get rid of the first two and be left with only the comet - once that is done the position can be calculated based on the center of brightness.

Stars are eliminated using a median blur on a copy of the image - you need to pick a size (in pixels) that is significantly bigger than the biggest stars, but significantly smaller than the brighter parts of the comet.

After the stars are removed, SharpCap calculates 100% - (Detection Threshold value) and finds the brightness level corresponding to that centile of the image histogram - that level is the comet detection threshold. Pixels below the threshold are set to black, pixels brighter than the threshold have the threshold subtracted from them. This should leave a frame where everthing is black except for the area around the comet nucleus.

The best bet for setting up is to turn on live stacking, select 'Highlight Stars' and then select 'View : Individual Frames' rather than 'View : Stack' - that way you see the calculated comet position for each frame as you adjust the values.

cheers,

Robin

[DrBobAZ]

Hi Robin

I downloaded the 4.1 beta yet I cannot see the 'align on a comet' menu box under live stacking. I believe I did this after the post showing that option.

Any info would be appreciated!

Bob

[DrBobAZ]

Hi Robin


I kind of answered my own question.

I downloaded the beta 4.1 just now and the comet option now appears.

I also noticed that now the FWHM filter returns ??? for all frames. That didn't occur with the previous beta 4.1 I had.

Bob

[admin]

Hi Bob,

glad you found the latest beta. I would expect the FWHM to give ??? when comet stacking is enabled as in comet mode it doesn't do any star detection, so a measurement of FWHM is not available. In my testing the FWHM values appear when you change back to normal alignment mode.

cheers,

Robin

[DrBobAZ]

Hi Robin

Yes, I was able to generate an image of the comet using this great new feature. Got both dust and ion tail. Very pleased, thanks. The total exposure was 20 min, so there were some star trails. I used the star size feature, but the brightest part of the nucleus was as large as surrounding stars, so to keep that I did have to settle for some trails. Tonight, I'm even going to try a higher gain to see if the ion tail appears with shorter exposures and thus less star trailing.

About the ??? : I was reporting on the ??? being present, even during normal stacking, that is without the comet option engaged. I understand the logic if the comet feature was engaged. Is that something you would have expected?

thanks,

Bob

[abillionnewstars]

I tested it too. It worked. However I was getting a weird effect after 5 stacked subs. The middle of the stars were erased.
It is very promising.

[oopfan]

When do you think you can share some of your photos in SharpCap's Gallery?

[DrBobAZ]

Mine is up there right now

[oopfan]

Saw it. Thank you!

[admin]

@DrBobAZ

no, I would not expect the ??? in normal stacking (well, there are reasons that you might see it like alignment turned off or very poor selection of stars in the field, but not normally). In general it should show up except in those special cases. If it is persistent and none of the above apply then please let me know (and ideally share a few raw frames and your stacking settings so I can test).

Also, the star removal thing was poor naming on my part - it does not attempt to remove the stars from the stack, it attempts to remove them from a copy of the image while working out the comet position (once you have removed the stars and the background, what is left is the comet).

@abillionnewstars - any chance you have sigma clipped stacking turned on? That would interact badly with comet tracking. 'Default' stacking would be the best.

cheers,

Robin