New Feature : Planetary Live Stacking

[admin]

Hi Peter,

looks like a bug that has always been there - fortunately looks simple to fix.

cheers,

Robin

[PeterC65]

Hi Robin,

Another thing I've just noticed from last night ...

The filenames of snapshots and time lapse videos are missing some information. I use the following filename templates:
Single File (Video): {TargetName}\Captures\{TargetName} {Filter} {Exposure} x{Gain} D{Date} T{Time}
Sequence (Stills): {TargetName}\Snapshots\{TargetName} {Filter} {Exposure} x{Gain} D{Date} T{Time} I{Index:2}

Last night the snapshots and AVI timelapse files were missing the exposure and gain values. They were correct in the SER capture files though. I've noticed this happening during previous sessions too.

Peter

[johnpd]

Fantastic. Is it possible to change the color a little to a more yellow hue.

JohnD

Sun_Disk.jpg (551.91 KiB) Viewed 90303 times

Sun_Spots.jpg (727.72 KiB) Viewed 90303 times

[admin]

Hi Peter,

you are correct, and actually I noticed that separate snapshots from the same stack end up in separate folders, which is probably not right either. I need to go back and revisit how saved files are named when planetary stacking to bring some consistency.

cheers,

Robin

[admin]

Fantastic. Is it possible to change the color a little to a more yellow hue.

JohnD

Sun_Disk.jpg

Sun_Spots.jpg

Hi John,

just went back and checked the colour transformation being used for solar colorization. It was supposed to roughly match the ones shown here - https://www.photographingspace.com/colo ... ar-images/ , and it did... roughly...

I have re-done the curves being used to make them approximate more closely the graphs shown on that page and it does make the image less orange and more yellow (see below - old on the left, new on the right). This will be in the next SharpCap update

cheers, and nice images!

Robin

Solar.JPG (42.66 KiB) Viewed 90302 times

[johnpd]

Thanks Robin,

This entire live planetary stacking is a fantastic update.

JohnD

[lumenmikie]

First, thanks for a great addition to a great app. I’m doing H-alpha imaging on a solar tracking, alt-az mount. I tried the latest version, using surface alignment and it seems to work very well. I also like the ability to stack only the best x% frames. I’ve found an issue with the time lapse capability that my particular use case exposes. Since there is image rotation with an alt-az mount, accumulating frames over an extended period results in the stack becoming blurred. If I manually reset the stack every minute or so, I can avoid this problem.

My time lapse setup is using a 100 frame stack, using the best 20% of frames, with one image every 30 sec. This works well with the local seeing and my camera frame rate (ASI 174MM). In this case automatically resetting the stack for each frame (or each N frames) would be a good solution.

Thanks again for the great program.
Mike

[admin]

Hi Mike,

glad that you are enjoying the new feature.

To deal with the frame rotation in the software as it stands, the best bet is to either drop the number of frames being stacked or increase the percentage being used. To work out the right values, you need to think about the 'persistence time' of the stack - how much time does it take before old data is basically pushed out of the stack and new data replaces it.

Roughly speaking the persistance time of the stack is 3 * stack length (frames) * 100 / frame rate / percent of frames used.

Suppose you are imaging at 20fps, with your 100 frame stack length and 20%, you get about 75s. (If you are only at 10fps then its 150s, at 30fps 50s)

If you increase the filter to allow 40% of frames through, you halve the persistence time straight away, bringing the 20fps calculation down to 37.5s, which probably means no need to reset at all as the old data will clear before rotation is an issue. You can achieve the same effect by reducing the number of frames in the stack, but I suspect that will give worse results due to increased noise.

One of the things on my to-do list is an improved stacking algorithm which would give a much harder cut-off of the data included in the stack - basically the factor of 3 in the above calculation would get removed. On the flip side, this change will use more memory and add some additional calculation work to the stacking. I think that would help you a lot when it happens.

cheers,

Robin

[admin]

Just to note that there is a new update available today. Only a few changes relating to Planetary stacking this time though:

* Renamed in the menu to 'Live Solar/Lunar/Planetary Stacking and Enhancement' - the old name was inaccurate since the addition of surface alignment
* Improvement of the solar colorization color palette (see image about 3 posts up)
* Addition of an 'Invert' option for solar colorization in planetary live stacking
* some bug fixes

cheers,

Robin

[lumenmikie]

Hello Robin,
Thank you for the response. In my case:
stack size: 100
camera speed: 60fps
good frame %: 20
persistence: 25 sec
I'm saving a frame every 30 sec so I should usually be able to collect the specified amount of data.

When I looked at the frames again, I believe the source of the problem is the mount, a Sky Watcher Solar Quest alt-az tracker. It keeps the sun centered, but not exactly. Occasionally "large" movements are seen. After frame 27, it looks like movement of ~5% of the solar disk occured, maybe multiple times. I noticed what was happening after 4 minutes and reset the stack (frame 34). After that, I manually reset the stack every minute so the problem wouldn't propagate, if it were to reoccur.

2024-01-01_19_35_49Z_00027.jpg (68.52 KiB) Viewed 89891 times

2024-01-01_19_35_49Z_00028.jpg (63.62 KiB) Viewed 89891 times

2024-01-01_19_35_49Z_00030.jpg (63.21 KiB) Viewed 89891 times

2024-01-01_19_35_49Z_00031.jpg (64.89 KiB) Viewed 89891 times

2024-01-01_19_35_49Z_00032.jpg (59.03 KiB) Viewed 89891 times

2024-01-01_19_35_49Z_00033.jpg (56.38 KiB) Viewed 89891 times

2024-01-01_19_35_49Z_00034.jpg (74.22 KiB) Viewed 89891 times

As I said, I believe this is a mount problem, but am including the images in the event they suggest something to you. Resetting the stack after each frame is a brute force solution to this issue, but might obscure a more subtle problem, I don't know.

I believe someone else pointed out some issues on file naming in this thread so it may be already on your radar. Enabling the ability to get the time of each frame in the still sequence would be wonderful (I see a checkbox but it won't check!).

Thank you again for all your time and effort.
Cordially,
Mike