Color noise "stripes" in integrated light

[Diane Miller]

I have gotten to an integrated light and have done a dynamic crop.  I haven't done any denoise as I don't yet know how.  I did ABE with subtraction and then repeated it with division.  I got a result that seems a bit high in contrast, but this exposure was too long for the brightest areas.  (It will be redone in a few months with two or three exposures.  This is just early in the learning process.)  I'll try DBE for comparison but in the meantime I'm wondering about the color noise in the BG, which follows a stripe pattern, which is going in the same direction as the drift between frames.

This is happening in the Star Alignment step.  I assume it's another parameter I have wrong.

Screenshots are here:
https://dianemiller.smugmug.com/Photography/Files-for-PI-forum-2/n-BBs62B/i-ZMMXjhZ

Flats were calibrated with master-bias and integrated.  They are gray and show some noise but not in this pattern.
Lights were calibrated with master-flat.
Did CC, debayered, SubframeSelector, StarAlignment and integrated.
Lights debayered and registered files have a lot of color noise but not in this pattern.

I was using the Canon 7D2 with a LP filter, ISO 1600, 2 min subs at f/4.  Astrotrac.  About 26 "approved" lights used.  No darks taken this session. 

[msmythers]

Diane

Look at these postings.
http://pixinsight.com/forum/index.php?topic=7055.msg

http://pixinsight.com/forum/index.php?topic=6370.msg

http://pixinsight.com.ar/en/info/processing-examples/19/crux-southern-cross.html

Another reason for full calibration using darks.


Mike

[Diane Miller]

Thanks guys.  I'll toss out this set and practice on some that have darks.  I was frightened away from using darks by some things I read earlier, but at least I should take them and see how it goes with and without.  It looks like using them is mandatory, though.  I can almost always manage to get a set within a few degrees of the same ambient temp.

Am getting more precise with PA too, with a new scope, so drift is usually small.  If I can master a few more processing steps I need to get something like the Orion Sirius.

[msmythers]

Remember you can always collect darks at a later date and reprocess this set of data.    Still worth having the data to practice on.


Mike

[Diane Miller]

Yes, but unfortunately in this case I didn't record the temperature in my notes on the session.  Now have developed better habits, and certain I'll add more as I go along.

[pfile]

the 7d2 might add a temperature keyword to the CR2 exif - most canon cameras do but apparently some newer ones do not.

the temperature is for some random chip somewhere in the body, but it will correlate with the sensor temperature.

rob

[msmythers]

Not the best practice but if you check the weather history for your area you can find what the temperature was outside on an hour by hour basis for the night you shot. I use weather underground and local weather stations in the US. It gets you in the ball park. Then just go outside on a night close to that and take your darks. Again remember this is for practice so the results may not be perfect but you can see the difference.

I shoot with a camera, an original Sony Nex-5 that doesn't record temp. I also live on the coast in Florida and 80 percent of the year I can shoot darks inside my house during the day. It's the same temperature as outside at night within a few degrees. The other 20 percent it's within 10 degrees. Good for banana trees bad for astrophotography.


Mike

[Diane Miller]

Thanks Mike -- I didn't know about weather history.  Unfortunately we have a lot of microclimates here (in the hills NW of San Francisco) and the local dark site where I prefer to shoot is in the hills and colder than the local places near sea level that would report temps.  Sometimes the temp drops 10 degrees or more over an hour and a half.  But by the time I get home it's usually comparable there so I can leave the camera outside in a sheltered location and get the darks there.

If I open an image in PS and look at File Info, I've never seen a temperature value (except the color temp).

[pfile]

you'll have to try exiftool, i doubt PS would show you all the junk in the exif header.

http://www.sno.phy.queensu.ca/~phil/exiftool/install.html

rob

[ChoJin]

I'm pretty sure the 7Dm2 has a temperature sensor, exiftool is the way

[Diane Miller]

Thanks -- I'll check it out for the 7D2 and for the 1DX Mk 2 and report back in a few days when I have time.