How are people doing longer exposures (5+ minutes), when the satellites ruin everything?

[Trippelforge]

Every time I push my exposures time up I tend to lose a lot of data. Mostly due to streaks from satellites, which has caused me to stick to around 45 seconds. How are people doing much longer exposures with so many of the damn things up there?

[Vroobel]

What software do you use for processing your data?

[The Lazy Astronomer]

Rejection algorithms will deal with satellite trials when stacking. It's like camera amp glow - easily removed and a complete non-issue.

[a6400]

I guess you also need more frames, or higher rejection sigma setting.

[ONIKKINEN]

Rejection when stacking will remove them easily. There could be a hundred trails in total and none will be visible in the stack at the end.

DSS has kappa-sigma clipping which works well with a modest number of images (a few dozen at least to get the best result). Siril has winsorized sigma clipping that does the same thing. Siril also has the Generalized extreme studentized deviate test (awful name) method, which is very good with a large number of frames, think close to or more than a hundred. At the end of the day satellite trails are meaningless in subexposures and get rejected.

[Trippelforge]

3 hours ago, Vroobel said:

What software do you use for processing your data?

PixInsight 

11 minutes ago, ONIKKINEN said:

Rejection when stacking will remove them easily. There could be a hundred trails in total and none will be visible in the stack at the end.

DSS has kappa-sigma clipping which works well with a modest number of images (a few dozen at least to get the best result). Siril has winsorized sigma clipping that does the same thing. Siril also has the Generalized extreme studentized deviate test (awful name) method, which is very good with a large number of frames, think close to or more than a hundred. At the end of the day satellite trails are meaningless in subexposures and get rejected.

For every rejected frame it's "X" amount of data loss, that's what I am confused about. But perhaps I am not understanding things correctly. Say I take for example an hours worth of images at 5 minutes each, that's a total of 12 images (obviously). But if due to the longer exposures it caused me to lose say 4-5 images I just lost 1/3 of my data. 

 

 

 

[ONIKKINEN]

2 minutes ago, Trippelforge said:

PixInsight 

For every rejected frame it's "X" amount of data loss, that's what I am confused about. But perhaps I am not understanding things correctly. Say I take for example an hours worth of images at 5 minutes each, that's a total of 12 images (obviously). But if due to the longer exposures it caused me to lose say 4-5 images I just lost 1/3 of my data. 

 

 

 

You lose only the pixels that are rejected, not the whole image. Usually rejection rates are below 1% of all pixels so not an issue at all. The rejected pixels are replaced by the median value of that pixel from the stack so there will be no evidence left that there ever was a satellite trail in the first place.

[ollypenrice]

It's a total non-issue if you have a decent number of subs and a good outlier-rejection algorithm, as others have said. Now that I'm working with CMOS in 3 minute subs, rather than CCD with up to 30 minute subs, sat trails no longer exist, despite the increase in the wretched things.

Olly

[tomato]

I use 2 minute exposures these days, satellite trails seem to depend on where you are imaging in the sky. Some nights I get 3 or 4 in a 4 hour stint, on other sessions every sub has at least one trail.

But as has already been stated, they are just a curiosity these days, the stacking software (APP in my case) deals with them efficiently every time.

[Adam J]

2 hours ago, Trippelforge said:

PixInsight 

For every rejected frame it's "X" amount of data loss, that's what I am confused about. But perhaps I am not understanding things correctly. Say I take for example an hours worth of images at 5 minutes each, that's a total of 12 images (obviously). But if due to the longer exposures it caused me to lose say 4-5 images I just lost 1/3 of my data. 

 

 

 

Because its not the whole frame that is rejected just the effected pixels.