Win some , lose some (dark bands across image)

[rubecula]

After last night's session I finally thought that I was getting somewhere with imaging. Managed to get 21 x 300 sec dithered images with no tracking or camera issues. ( latest problem with camera disconnecting cured by new USB cable).  However when I developed the image in DSS, StarTools and Lightroom I found that the image had a series of horizontal dark bands.  I've tweaked the image below in Lightroom to make the bands more obvious.

Camera is a modified Canon 550D.  I'm sure that I have seen something about dark bands somewhere here but a search found nothing useful.  Does anyone know the cause of this and potential cure?

Thanks,

[MARS1960]

I'm new to imaging but from what i have read the consensus seems to be cooling issues, maybe try leaving a longer interval between exposures.

[Oddsocks]

On 06/10/2016 at 17:29, rubecula said:

Does anyone know the cause of this and potential cure?

Horizontal , diagonal and vertical banding can be rather difficult to pin down, often due to hum loops caused when camera and PC are running off individual external power supplies and connected via USB cable thus creating a source of interference within the camera or a current leakage from the PC to ground via the camera, telescope and the mount/tripod etc. Easy enough to prove if this is the case by running the laptop (if used) or the camera off battery power only for a while and see if the banding has disappeared.

Probably easiest is to deal with effectively in post processing, if using Photoshop consider the very reasonably priced Astronomy Tools V1.6 Photoshop Plug-In from ProDigital Software, currently $21.95. You won't find a better package of useful tools for the money, running two iterations of the "Horizontal Banding Noise Reduction" tool on your posted image results in the cleaned up image attached below, this was on the rather low quality JPG, it works better on the uncompressed source image but at least you get the idea.

This is a tool set I use a lot for DLSR images in Photoshop.

http://www.prodigitalsoftware.com/Astronomy_Tools_For_Full_Version.html

William.

 

[geordie85]

I'm curious to know if you took flats using a laptop or something similar? 

This can be caused by having the exposure time for your flats too low and your catching the flicker of the screen. 

Have you ever pointed a camera at a TV screen and noticed the rolling bands that are very similar to the ones in your photo?

If this is what you have done then try putting something plain and white over your scope to increase  your exposure time.

Just a thought 

[alacant]

Hi. How did the StarTools Band module do with this? I remember something about being applied before Decon. Maybe try both? There's not much it can't shift. HTH.

[Thalestris24]

Hi

I've recently been getting horizontal noise bands with my cooled mono 550d. It didn't seem to be a usb problem but after a long stint last night I kept getting loss of communication and also the noise suddenly got worse. Unfortunately the camera failed in the end (don't want to worry you!). As things stand I'm not sure if it's an electronic component failure or something else. As a precaution I've put the camera in with some silica drying beads in case it was a build up of condensation internally. The noise appeared on single subs so not any kind of processing issue. Investigation will continue...

Louise

[Oddsocks]

On 06/10/2016 at 17:29, rubecula said:

Does anyone know the cause of this and potential cure?

 

On 06/10/2016 at 21:31, alacant said:

 

Hi. How did the StarTools Band module do with this?

 

For the posted comparison image above I did this in Photoshop using the Astronomy Tools v1.6 actions set plug-in, not Startools, but you jogged my memory and you can also do this in Startools, you need to select the band orientation mode at the bottom of the Startools desktop first then apply the "Band" tool, several times may be required.

In the image posted below this is Robin's original image taken back into Startools, "Band" tool applied four times, no other processing done.

 

And below is the same image processed in Photoshop using two iterations of the Astronomy Tools v1.6 actions set plug-in "Horizontal Band Noise Reduction"

 

Of the two tools I think the Photoshop action-set plug-in does a better job with Robin's compressed JPG image, the Startools version leaves some faint banding at the bottom of the picture but in practice working with an uncompressed TIF the difference between the two tools might be closer.

Just to add to the possibilities, below is Robin's original image processed by running a single operation of the CanonBandingScript in PixInsight but this seems to introduce a slight colour cast in the bottom part of the image so with the compressed jpg my favourite is still the Photoshop + Astronomy Tools v1.6 action set plug-in

 

HTH

William.

[rubecula]

Thanks all for your help, particularly William.  You've given me lots of ideas to sort this out. 

The camera and laptop are powered by a single supply from the house to the observatory so different supplies might not be the cause.  Nevertheless, next time I'm out I will take some images with the camera on mains and the laptop on battery power.

I tried using the StarTools band function (it has options for both vertical and horizontal banding) just once and it seemed to have no effect.  I've just tried 5 passes of the Startools band function on the tiff file that I got the above jpeg from.  Again it didn't make much difference.  I'll try again starting from the original DSS output and see what happens.

I've been struggling a bit with StarTools, it all seems so hit and miss, especially on the noise reduction.  I haven't a clue what the instructions mean.  What is a "noise floor" what does it look like?  How can you see "read noise"?

I have a copy of Photoshop CS5 and I've wondered about using that instead of StarTools. I used StarTools as I thought it might be a shorter learning process.  I am going to give Astronomy Tools a go and see how I get on.

In all of this I'm happy that it's nothing to do with the dithering.

[Oddsocks]

1 hour ago, rubecula said:

I tried using the StarTools band function (it has options for both vertical and horizontal banding) just once and it seemed to have no effect.  I've just tried 5 passes of the Startools band function on the tiff file that I got the above jpeg from.  Again it didn't make much difference.  I'll try again starting from the original DSS output and see what happens.

I've been struggling a bit with StarTools, it all seems so hit and miss, especially on the noise reduction.  I haven't a clue what the instructions mean.  What is a "noise floor" what does it look like?  How can you see "read noise"?

I have a copy of Photoshop CS5 and I've wondered about using that instead of StarTools. I used StarTools as I thought it might be a shorter learning process.  I am going to give Astronomy Tools a go and see how I get on.

In all of this I'm happy that it's nothing to do with the dithering.

I'm glad you replied Robin, I had missed the fact that in Startools you can select the "Band" tool orientation by toggling the command bar at the bottom of the desktop. I have been back to my post and corrected it for anyone else that picks it up.

I don't (as you can tell) use Startools very much, had it a couple of years but find it difficult to control. Every time I start with the same image and run through with Startools I end up somewhere different and I just can not pin down what was the difference between the two run-throughs, I always feel that the program is in control of me and not vice-versa.

PixInsight and Photoshop (with plug-ins) does everything I need and you are always in control to the extent that you can always go back to the beginning, start again and if you follow the same steps you end up in the same place so it gives you the confidence to try small changes, see where they they lead, and know that change always results in this repeatable end result...

Banding in DSLR's must be fairly common since there are quite a few "special" tools to deal with it, definitely not related to dithering or guiding, seems quite a few users report it is temperature related. My Nikon doesn't produce bands but it just eats stars instead and since it was primary for conventional use I don't want to attempt the Nikon firmware "StarEater" hack and risk breaking it!

William.

[geordie85]

1 hour ago, rubecula said:

Thanks all for your help, particularly William.  You've given me lots of ideas to sort this out. 

The camera and laptop are powered by a single supply from the house to the observatory so different supplies might not be the cause.  Nevertheless, next time I'm out I will take some images with the camera on mains and the laptop on battery  power.

I saw a video on YouTube a while ago from astronomyshed and he was having problems with his qhy8l images. He was running both ccd and laptop from the same supply and getting a power loop. Once he ran them on separate currents his images improved dramatically. 

This could also be causing your banding.

[rubecula]

16 hours ago, Oddsocks said:

 Every time I start with the same image and run through with Startools I end up somewhere different and I just can not pin down what was the difference between the two run-throughs, I always feel that the program is in control of me and not vice-versa.

William, that's a very good description of my StarTools experience as well.

geordie, thanks for that. As I said above next time we have a clear sky I'll try with one on mains and the other on battery power.  I'm not sure how I could get them on seperate currents so may have to settle for removing them during post-processing.

Thanks

[geordie85]

41 minutes ago, rubecula said:

William, that's a very good description of my StarTools experience as well.

geordie, thanks for that. As I said above next time we have a clear sky I'll try with one on mains and the other on battery power.  I'm not sure how I could get them on seperate currents so may have to settle for removing them during post-processing.

Thanks

Sorry I didn't mean currents as in AC and DC, I mean using 2 different outlets. So maybe try a different extension for each. 1 for laptop and 1 for your camera.

[alacant]

3 hours ago, rubecula said:

StarTools experience

Hi. One of the things which attracted me to StarTools was that if there is a problem, you have direct contact with the author. He's also open to requests for features and improvements. HTH.

[Stub Mandrel]

If you have 'Noel's Actions' for Photoshop, contact him and ask if you can have the 'horizontal large band reduction' action he put together for Neil Wyatt.

 

[rubecula]

Thanks for that Neil, I just bought them yesterday. Not had chance to try them but will contact him.