How to remove satellite trails?

[MarkS]

I'm interested to know the various approaches people take to dealing with satellite trails. 
Approaches I can think of are:
1) Throw away all subs containing satellite trails
2) Do nothing - just hope the stack makes them "fade away" sufficiently - maybe perform "surgery" on the final image.
3) Perform a sigma stack to do a better job of removal

Up until now I've taken approach 3 but it doesn't always do a complete job so I'm sometimes forced to throw away a few subs with the worst trails.  But throwing away subs is very wasteful.

Ideally I'd like to "mask off" trails in the affected subs so they don't contribute to the final stack - or maybe replace the trail with data from an unaffected sub i.e. to "heal" that part of the data (to use Photoshop-speak.  However I don't have an easy means of doing this apart from "Photoshopping" each affected sub.

So - what do *you* do?

Mark

[JohnP]

Mark - I get about one trail in 100 subs across mine... I knew there had to be one benefit to imaging with a chip that is only 4mm... Normally the stacking fades it to nowt - if not I just omit that one sub... Looking at your m13 though looks like you'd have to ditch tons of your subs - I have never seen so many trails - must be the dark skies coupled with the wide FOV....

Sorry I cant help - John

[Mac]

Just a thought.

if you have three subs, two clean,  A B & S (S) for satelite

If you stack all three (D) you will still have a slight trail as it should be reduced by a factor of 3.

Now if you restack A B & D, you are re adding your two good subs on themselves and to the reduced trail,
this should reduce the trail further, without adding any extra noise,

you can then use these in your final stack.


Actually thinking about it another way,
If you were to stack all your frames in one hit first, (FINAL)
then restack all the non satelite frames and the (Final one)
would this not reduce the trails further?

Might be worth a try.

Mac.

[Fay]

AA5 has a remove lines feature. Does not remove stars in between. You select beginning & end of line

[Rick]

must be the dark skies coupled with the wide FOV....

...and time of year. Less actual night, and mor chance of satellites being illuminated over the pole...