Dark frames

I've been quite impressed by the results of simply pointing my Olympus
E10 at the sky and opening the shutter for eight seconds; it's got a
good f/2.4 lens, and I get images down to about magnitude 8.5 through
dire suburban skies.

On the other hand, it seems to have hot pixels. I thought hot pixels
were essentially a *manufacturing* fault in the CCD, and so a single
dark frame (leave aperture and speed settings fixed, just also leave
the lens-cap on) would characterise them adequately for all time. But
in a set of 100 photos taken last Tuesday, I'm still seeing hot pixels
after subtracting a dark-frame taken two weeks ago.

My temptation is to compute a minimum for each pixel position across
all the photos and use that as a dark frame; the photometry is
dreadful in any case, even after I've summed across 2x2 pixel groups -
not sure if that's a matter of the CCD response and the V filter being
vastly different, I should probably plot brightness-residual against
spectral type and see if there's a correlation. Is this a sensible
way to proceed, given that most of the photos are star fields?

Should I in fact be taking a dark frame for each observing session,
and if so does this indicate there's something wrong with the camera?

Tom

On 20 Sep 2004 22:57:47 +0100 (BST)
Thomas Womack wrote:

Should I in fact be taking a dark frame for each observing
session, and if so does this indicate there's something wrong
with the camera?

I Was under the impression you should take a dark frame for
almost every photo, as things can vary, depending on
temperature, battery charge, subject being photographed etc...
But I'm not photo expert.

It's certianly not a problem with the camera - It's not a
manufacturing fault, it's just something that happens with CCDs
due to current leakage that can't really be avoided.

I Did a google on "hot pixels digital camera" and found a lot of
links to software that removes hot pixels from images... How
good it is I don't know, but if it's free, can't hurt to try.

Also, I was under the impression that most newer cameras
automatically take a dark frame for long exposure shots. I'm
pretty sure the Canon A85 I'm looking at does it for shots over
2 seconds (up to 30 seconds), and that's just a budget, consumer
level camera.

-Chris D

Im sorry, but quite frankly, sometimes I wonder if this is all worth it,
just to avoid having to develop film and getting an instant image. The
trad off is less than impressive at times, to say the least!





Thomas Womack wrote:

I've been quite impressed by the results of simply pointing my Olympus
E10 at the sky and opening the shutter for eight seconds; it's got a
good f/2.4 lens, and I get images down to about magnitude 8.5 through
dire suburban skies.

On the other hand, it seems to have hot pixels. I thought hot pixels
were essentially a *manufacturing* fault in the CCD, and so a single
dark frame (leave aperture and speed settings fixed, just also leave
the lens-cap on) would characterise them adequately for all time. But
in a set of 100 photos taken last Tuesday, I'm still seeing hot pixels
after subtracting a dark-frame taken two weeks ago.

My temptation is to compute a minimum for each pixel position across
all the photos and use that as a dark frame; the photometry is
dreadful in any case, even after I've summed across 2x2 pixel groups -
not sure if that's a matter of the CCD response and the V filter being
vastly different, I should probably plot brightness-residual against
spectral type and see if there's a correlation. Is this a sensible
way to proceed, given that most of the photos are star fields?

Should I in fact be taking a dark frame for each observing session,
and if so does this indicate there's something wrong with the camera?

Tom

I recently found that the hot pixels seen in a frame with the lens
completely covered were different from the ones which appeared in
a picture of the open sky; there were far fewer in the completely
dark case. By taking a few frames of the sky and not tracking, you
can see the stars move while other dots stay still. The fainter dots
tend not to show up in the usual dark frame and hence remain in
stacked images. I'm guessing that pixels have different "turn on"
thresholds. Registax has a way of making dark frames from multiple
images (it just stacks them with no alignment) so maybe the thing to
do is to take lots of pictures with no tracking so that all the
stars average out.

Albert

Thomas Womack ) wrote:
: I've been quite impressed by the results of simply pointing my Olympus
: E10 at the sky and opening the shutter for eight seconds; it's got a
: good f/2.4 lens, and I get images down to about magnitude 8.5 through
: dire suburban skies.

: On the other hand, it seems to have hot pixels. I thought hot pixels
: were essentially a *manufacturing* fault in the CCD, and so a single
: dark frame (leave aperture and speed settings fixed, just also leave
: the lens-cap on) would characterise them adequately for all time. But
: in a set of 100 photos taken last Tuesday, I'm still seeing hot pixels
: after subtracting a dark-frame taken two weeks ago.

: My temptation is to compute a minimum for each pixel position across
: all the photos and use that as a dark frame; the photometry is
: dreadful in any case, even after I've summed across 2x2 pixel groups -
: not sure if that's a matter of the CCD response and the V filter being
: vastly different, I should probably plot brightness-residual against
: spectral type and see if there's a correlation. Is this a sensible
: way to proceed, given that most of the photos are star fields?

: Should I in fact be taking a dark frame for each observing session,
: and if so does this indicate there's something wrong with the camera?

: Tom

Should I in fact be taking a dark frame for each observing session,
and if so does this indicate there's something wrong with the camera?

Tom


I think you need to take frequent dark frames. Probably more often that just
once every observing session.

The dark current is a function of the temperature of the chip and the
temperature of the chip is not stable. The dark frame used should be taken at
the same temperature as the image, the best way to do this is to use a dark
frame that has been taken within moments of the actual image so the
temperatures of the two are as close as possible.

Also, I believe it is necessary to take individual dark frames for each length
exposure. I know that some more sophisticated cameras with thermostatically
cooled CCD chips use multiple dark frames and average them.

Also, make sure you are saving both the original image and the dark frame in an
Uncompressed Format, either some sort of RAW file or as a TIFF. If you save it
as a JPG, the "lossy" compression will not retain the accurate pixel by pixel
infomation necessary for the dark frame subtraction to be effective.

Best wishes, clear skies...

jon

Hi, I have a related question. I'm using RegiStar to align/derotate frames
taken in alt/az mode. It does a really nice job, but it has no provision to
use dark/flat frames. Can someone point me to a RegiStax group that will
let me know how to use RegiStax to subtract my dark/flat frames and save
each processed frame without stacking them so I can align them in RegiStar?

--

Thanks,

Chuck

"Jon Isaacs" wrote in message
...
Should I in fact be taking a dark frame for each observing session,
and if so does this indicate there's something wrong with the camera?

Tom


I think you need to take frequent dark frames. Probably more often that
just
once every observing session.

The dark current is a function of the temperature of the chip and the
temperature of the chip is not stable. The dark frame used should be
taken at
the same temperature as the image, the best way to do this is to use a
dark
frame that has been taken within moments of the actual image so the
temperatures of the two are as close as possible.

Also, I believe it is necessary to take individual dark frames for each
length
exposure. I know that some more sophisticated cameras with
thermostatically
cooled CCD chips use multiple dark frames and average them.

Also, make sure you are saving both the original image and the dark frame
in an
Uncompressed Format, either some sort of RAW file or as a TIFF. If you
save it
as a JPG, the "lossy" compression will not retain the accurate pixel by
pixel
infomation necessary for the dark frame subtraction to be effective.

Best wishes, clear skies...

jon

In article ,
Jon Isaacs wrote:
Should I in fact be taking a dark frame for each observing session,
and if so does this indicate there's something wrong with the camera?

Tom


I think you need to take frequent dark frames. Probably more often that just
once every observing session.

The dark current is a function of the temperature of the chip and the
temperature of the chip is not stable. The dark frame used should be taken at
the same temperature as the image, the best way to do this is to use a dark
frame that has been taken within moments of the actual image so the
temperatures of the two are as close as possible.

Also, I believe it is necessary to take individual dark frames for each length
exposure. I know that some more sophisticated cameras with thermostatically
cooled CCD chips use multiple dark frames and average them.

I'm doing all the averaging and suchlike in software; a gigabyte
CompactFlash card stores 150 frames uncompressed (2256x1684, 2-byte
values of which only the top ten bits are used). I'm using a constant exposure,
largest that the camera can do (8 seconds). This is why I was confused that the
dark frames were so different; temperature on clear evenings a week apart in the
same place in autumn aren't going to be _that_ different.

Tom

There are other factors other than the temperature that can affect the
output from the dark frame, and I certainly agree with taking dark frames
frequently. The best policy is to do it after or before each exposure you
do, but this takes up time and memory space.

Unless the cooling of the CCD is regulated to a certain temperature, and not
just to some temperature below ambient, even slight differences in
temperature might translate to differences in the content of dark frames.
And even then, differences in the temperature of the readout electronics
induce something into the signal, and these are not regulated, although the
effect is less than that from the CCD itself.

In short, as you experienced yourself, dark frames need to be taken near the
time the image is taken to insure a good match.
--
Yours Truly,
--- Dave

----------------------------------------------------------------------
'raid if you're afraid you'll have to overlook it.
Besides, you knew the job was dangerous when you took it.
----------------------------------------------------------------------
"Thomas Womack" wrote in message
...
In article ,
Jon Isaacs wrote:
Should I in fact be taking a dark frame for each observing session,
and if so does this indicate there's something wrong with the camera?

Tom


I think you need to take frequent dark frames. Probably more often that
just
once every observing session.

The dark current is a function of the temperature of the chip and the
temperature of the chip is not stable. The dark frame used should be
taken at
the same temperature as the image, the best way to do this is to use a
dark
frame that has been taken within moments of the actual image so the
temperatures of the two are as close as possible.

Also, I believe it is necessary to take individual dark frames for each
length
exposure. I know that some more sophisticated cameras with
thermostatically
cooled CCD chips use multiple dark frames and average them.

I'm doing all the averaging and suchlike in software; a gigabyte
CompactFlash card stores 150 frames uncompressed (2256x1684, 2-byte
values of which only the top ten bits are used). I'm using a constant
exposure,
largest that the camera can do (8 seconds). This is why I was confused
that the
dark frames were so different; temperature on clear evenings a week apart
in the
same place in autumn aren't going to be _that_ different.

Tom