Michael,
I had to copy the image and blow it up to see the trails. By the way, the hot
pixels actually do stay at the same location, but if the position of the stars
is dithered between frames (that is, the camera is dithered) then stacking the
subs will line up the stars and cause the hot pixels to move. It looks like
your camera dithered in a more or less straight line.
It's interesting that each trail consists of one color of dots? I have seen
this effect before, but .I'd get multi-color dots in each trail because I
dithered my monochrome camera between each color filter exposure. Now I always
attack the hot pixels themselves using Cosmetic Correction during
pre-processing. Sigma clipping is still used in stacking, but mostly to
eliminate satellite trails.
Paul
----- Original Message -----
From: Michael McDonald <mikemac@xxxxxxxxxxx>
To: az-observing@xxxxxxxxxxxxx
Sent: Sat, 13 Oct 2018 14:15:57 -0400 (EDT)
Subject: [AZ-Observing] Re: Hot Pixel tracks?
From everything I’ve read, I was under the impression that the hot pixels were
a “defect” in the imager. As such they should be the same pixel(s) in every
frame, including your dark master. Which allows the master dark to subtract out
the hot pixels. I’ve never run across anything suggesting that hot pixels can
move frame to frame.
I used the “Average Sigma Clip” rejection algorithm in PixInsight’s
ImageIntegration tool. Somewhere I read that that was good for < than 10
subframes, “Winsorized Sigma Clip” was the best for 10 to 20 subframes and
“Linear Fit Clip” was best for more than 20 subframes.
I’ll try some of the other rejection algorithms to see if they do any better.
Mike McDonald
mikemac@xxxxxxxxxxx
On Oct 13, 2018, at 11:03 AM, Bernard Miller <bgmiller011@xxxxxxx> wrote:
Mike,
Every camera has hot and/or cold pixels. This is usually removed during
processing by using a sigma rejection algorithm while stacking. This
algorithm looks at the mean of all the pixels and rejects any pixel that is a
certain sigma above the mean (usually 2-3 sigma). With only 7 frames, it is
sometimes better to use a Poisson sigma rejection algorithm for stacking. I
am surprised Pixinsight did not remove these pixels since I know it has these
or similar algorithms.
Bernard
-----Original Message-----
From: az-observing-bounce@xxxxxxxxxxxxx <az-observing-bounce@xxxxxxxxxxxxx>
On Behalf Of Michael McDonald
Sent: Saturday, October 13, 2018 10:51 AM
To: az-observing@xxxxxxxxxxxxx
Subject: [AZ-Observing] Hot Pixel tracks?
Here’s my one attempt at taking an image at last week’s AASP. It consists of
seven 3 minute subframes taken with my Canon SL1 with my Canon EF200f2.8L
lens on my Star Adventurer. The subframes were then stacked in PixInsight.
http://www.mikemac.com/Astrophotos/Pleiades-20181005-stacked.jpg
At first glance, it’s not too bad. (For me!) But if you look closer, there
are hot pixel tracks all over the image! All of the tracks are 7 pixels long,
which happens to corresponds to the number of subframes. So I went back and
looked at the original images and low and behold there is a hot pixel in each
original subframe. And it’s not the same pixel that’s hot, but moves slightly
frame to frame.
Has anyone seen anything similar before and know what the root cause is?
Mike McDonald
mikemac@xxxxxxxxxxx
--
See message header for info on list archives or unsubscribing, and please
send personal replies to the author, not the list.
--
See message header for info on list archives or unsubscribing, and please
send personal replies to the author, not the list.
