What could have caused this?

[ice9481]

Details
Location: From Melbourne inner suburbs (so lots of city lights)
Camera: Canon A80 15s exposure stack of 30 pics using software Rot'n'Stack
Time: 12 May about 7:45 pm
The stacking software has an output of minimum, mean, and maximum images. The dark dotted streak is seen only on the "minimum" image.
Any one have any idea what this could be caused by? I'm guessing a slow moving satellite?

[Kal (Andrew)]

It could be a couple of 'dark' pixels in the camera?

[ice9481]

I think you're right, kal. A stack of pics taken much later from the same sequence shows this moving down-leftwards, as might be expected from a fixed point on the camera against the rightwards rotation of the sky.
Thanks, I hadn't thought of that.

[[1ponders] (Paul)]

Definately a pixel defect. I count at least 8 tracks like that when adjusted in PS

[rogerg (Roger)]

It might not necessarily be a camera defect. You can get this kind of thing when you do dark frame subtraction and the dark frames aren't of good enough quality - they differ too much from the actual dark frames required for a perfect match. Then you end up subtracting pixel values where you shouldn't be so you end up with black pixels where there shouldn't be.

If this is the case, averaging dark frames and then subtracting that resulting averaged image helps.

Roger.

[[1ponders] (Paul)]

You're right Roger, I should have been more specific. I wasn't necessarily referring to the camera itself, however the straight lines and that they align with the direction of the trailing stars, do tend to indicate that there are a number of regular misbehaving pixels camera pixels.

[[1ponders] (Paul)]

ice9481, were you tracking to capture the collection of images or are they from a non-tracking tripod?

[ving (David)]

i'd have picked a processing defect.

[Terry B]

I have had the same occur when I am slack and don't do enough darks. THe other problem is occasionally I will get some negative values in my darks after subtracting an offset. This also stuffs up the final image. To get around it you need to make sure that there are no negative pixels in the darks. I use Iris to do this processing and the command is just "min"

[ice9481]

The camera was set up on a fixed tripod. Images were stacked with no dark frame subtraction. Although it looks as if this is an artefact that can arise in a number of ways, in this instance it looks like the "dark pixel" suggested by Kal and 1ponders is the most likely explanation. The median and maximum (attached) output images from Rot'n'Stack are fine and do not show this.