Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

walking noise help

  • Please log in to reply
24 replies to this topic

#1 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 21 November 2020 - 08:30 PM

I know that dithering would correct this issue, but I would like to know as to why I feel like its getting worse in my photos.  

 

I have a skyguider pro that I use my dslr on it with my camera lenses. here is the latest disappointing image of the heart and soul nebulas, I wasnt looking for perfect I was actually testing out my CLS clip in filter and how to process it.  but lately the walking noise just overwhelms my photos and I dont know why that in the past few sessions its gotten worse.  Any ideas, my thought would be that its because my polar alignment is spot on and there is not that much movement of the target to get rid of the hot pixels, I take about 30 darks right after the lights, I take 30 flats and 30 bias, I used to take dark flats but I found that it causes more problems then helped.

 

I use DSS and APP, both with as much walking noise, and PS to do my stretching.  please any advice would be greatly appreciated.  thank you all in advance.

 

Heart and soul.jpg



#2 Jerry Lodriguss

Jerry Lodriguss

    Vendor

  • ****-
  • Vendors
  • Posts: 7,792
  • Joined: 19 Jul 2008
  • Loc: Voorhees, NJ

Posted 21 November 2020 - 08:45 PM

No ambient temps listed. Maybe it got worse because the temp increased.

Something is wrong with your flat fielding.

The temps of your lights and darks don't match.

Does the walking noise align north-south? If so, you are not polar aligned as well as you think.

If its east-west, you may have a huge amounting periodic error.

It hard to say because we need more info.

Problems with flat darks indicates something is wrong somewhere. Probably deeper.

Jerry

Edited by Jerry Lodriguss, 21 November 2020 - 09:03 PM.


#3 AZ Maverick

AZ Maverick

    Viking 1

  • *****
  • Posts: 574
  • Joined: 02 Nov 2017
  • Loc: Prescott, Arizona

Posted 21 November 2020 - 08:51 PM

Yes, dithering will help.

I have found that if your flats aren't exactly correct that they will exacerbate the problem.

If you stack with just your lights and darks and with no flats is the pattern better?  If so then you have something wrong with your flats.


Edited by AZ Maverick, 21 November 2020 - 08:52 PM.


#4 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 7,040
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 21 November 2020 - 09:29 PM

How do you know your PA is perfect...are you using a PoleMaster or Sharpcap to nail your PA ?
 


Edited by 17.5Dob, 21 November 2020 - 09:31 PM.


#5 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 21 November 2020 - 09:31 PM

I'll look into doing just lights and darks and seeing how that comes out and will let you guys know after. It's an unmodded dslr so there would be no temp readings. They were 2 min subs. And as far as tracking there was little to no movement in the frames.  But no I am just using an app on my phone to tell me where polaris should be through my skyguider pro polar scope


Edited by Tragusa113, 21 November 2020 - 09:36 PM.


#6 GIFTED1570

GIFTED1570

    Vostok 1

  • *****
  • Posts: 157
  • Joined: 21 Feb 2018

Posted 21 November 2020 - 09:35 PM

the field, tracking... is good enough, do not use bias, the darks are not matching the temperature or other parameters so you have to either take the darks on the same site you take the subs or monitor the temps and try to reproduce by a cold source, some PC dslr capture software can see the current temps, ALWAYS dither and play with different levels of rejection high and clipping algorithms 



#7 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 21 November 2020 - 09:38 PM

I guess to maybe test the PA theory I should shorten the exposure length and up the ISO.  Ill try that next clear night, same setup.  As far as my darks, as soon as I am done with my lights I put the cap on the lens and take 30 exposures 


Edited by Tragusa113, 21 November 2020 - 09:39 PM.


#8 Jerry Lodriguss

Jerry Lodriguss

    Vendor

  • ****-
  • Vendors
  • Posts: 7,792
  • Joined: 19 Jul 2008
  • Loc: Voorhees, NJ

Posted 21 November 2020 - 10:03 PM

It's an unmodded dslr so there would be no temp readings.


Modding has nothing to do with temps being recorded in the exif.

Canons records the temps. Nikon doesn't.

Jerry

#9 Jerry Lodriguss

Jerry Lodriguss

    Vendor

  • ****-
  • Vendors
  • Posts: 7,792
  • Joined: 19 Jul 2008
  • Loc: Voorhees, NJ

Posted 21 November 2020 - 10:05 PM

As far as my darks, as soon as I am done with my lights I put the cap on the lens and take 30 exposures


Yes, and the temps don't match.

Jerry

#10 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 21 November 2020 - 10:23 PM

Yes, and the temps don't match.

Jerry

How can I correct this issue then?



#11 klaussius

klaussius

    Viking 1

  • *****
  • Posts: 834
  • Joined: 02 Apr 2019
  • Loc: Buenos Aires

Posted 21 November 2020 - 11:01 PM

A CLS filter will make walking noise morw apparent.

Walking noise is caused by sensor PRNU and hot or warm pixels. With a CLS you'll be able to take longer subs that will show higher dark current kind of noise, which adds as walking noise in the final integration.

Better darks will help, but an uncooled camera can't fully get rid of it without dithering.

So dither.

Also, get proper darks. Try to match temperature as best as you can. If your camera records temperature in the raw file (many do), you can sort lights and darks by temperature and calibrate each light with similar darks.

Flats can also be noisy. If you've got noisy flats, the noise in the flats can cause a kind of walking noise. Inspect the noise level on your master flat, and maybe try applying some smoothing to it to remove some noise. The best bet is to capture enough flats with settings that minimize noise. In my canon t4i I use 200iso because it doesn't cause troubles but ymmv.
  • F.Meiresonne likes this

#12 F.Meiresonne

F.Meiresonne

    Fly Me to the Moon

  • *****
  • Posts: 6,479
  • Joined: 22 Dec 2003
  • Loc: Eeklo,Belgium

Posted 22 November 2020 - 04:49 AM

A CLS filter will make walking noise morw apparent.

Walking noise is caused by sensor PRNU and hot or warm pixels. With a CLS you'll be able to take longer subs that will show higher dark current kind of noise, which adds as walking noise in the final integration.

Better darks will help, but an uncooled camera can't fully get rid of it without dithering.

So dither.

Also, get proper darks. Try to match temperature as best as you can. If your camera records temperature in the raw file (many do), you can sort lights and darks by temperature and calibrate each light with similar darks.

Flats can also be noisy. If you've got noisy flats, the noise in the flats can cause a kind of walking noise. Inspect the noise level on your master flat, and maybe try applying some smoothing to it to remove some noise. The best bet is to capture enough flats with settings that minimize noise. In my canon t4i I use 200iso because it doesn't cause troubles but ymmv.

That is interesting. I have the issue also...never had it much in all my former stacks, but the latest was really horrible. I used a rather aggressive Omega Optical NPB filter.

I did dither on that occassion but my settings were bad, meaning the dithering distance was far from adequate.

But now i understand why i saw it on that occassion so heavily...



#13 sharkmelley

sharkmelley

    Soyuz

  • *****
  • Posts: 3,912
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 22 November 2020 - 05:20 AM

You haven't told us which camera you are using.  If you are not using a Nikon or Sony then dark scaling/optimisation is highly recommended (darks are much less effective on Nikon/Sony because of the raw data filtering).  Dithering is absolutely essential with Nikon/Sony.

 

If the background sky brightness changes during your imaging session then the pixel rejection during stacking is much less effective and more walking noise is left behind.  This (and not just temperature variations) is often the reason that walking noise is more apparent on some nights, compared with others. Normalisation of the subs is then required as part of the processing sequence.

 

You could try uploading a raw light, dark, flat and bias because there may be other concrete advice we can give.

 

Mark


Edited by sharkmelley, 22 November 2020 - 06:11 AM.


#14 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 22 November 2020 - 06:23 AM

I am using an unmodded canon 70d. CLS clip in filter. 50mm ef lens. On a skyguider pro.

Between the first and last sub it did appear to be diffrent exposures. I'll upload a few samples of all my frames in a bit.

#15 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 22 November 2020 - 06:53 AM

https://drive.google...iew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...ew?usp=sharing, https://drive.google...iew?usp=sharing

 

I gave 4 photos of each, first two and last two of the sets.  thank you everyone for their help.  I am really trying to learn here and cant figure out what im doing wrong but more importantly why im doing it wrong



#16 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 22 November 2020 - 07:46 AM

What I see is the temp varied from subs and darks 30C to around 26C



#17 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 22 November 2020 - 11:51 AM

im not the only one who is using this setup and everyone of them not dithering, and doing exactly what I do, I dont always get this type of noise I have done Orion and Andramada without this much noise, atleast to the point of not workable,  



#18 sharkmelley

sharkmelley

    Soyuz

  • *****
  • Posts: 3,912
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 22 November 2020 - 01:32 PM

I downloaded and took a look at your raw files.  I averaged the 4 darks to make a master.  This dark master worked well for last couple of lights but not at all well for the first couple (which were a higher temperature).

 

The changing temperature of the lights is causing a problem for the dark calibration so I definitely recommend dark optimisation, which scales the dark using a mathematical procedure for matching the noise pattern in the master dark.  I know DSS has a limitation which prevents scaling upwards (the scaling coefficient applied is between 0 and 1, not greater than 1) but I'm not sure about APP.  For that reason create a set of darks at the top end of the temperature range i.e. 30C for your current image.  The master dark will then be scaled downwards for the lights which have lower temperature.

 

Mark



#19 Huangdi

Huangdi

    Viking 1

  • *****
  • Posts: 901
  • Joined: 24 Jul 2019

Posted 22 November 2020 - 01:53 PM

There is a set of Photoshop actions that has a couple of very interesting tools.

Two of them are horizontal and vertical banding reduction. Now you might or might not suffer from banding, that doesn't matter. If you increase your canvas size and rotate the image so that the walking noise appears to. Go from top to bottom, the vertical banding reduction tool will do a really nice job at removing the walking noise. It's worth a try :) saved some otherwise unusable data for me. https://www.prodigit...olsActions.html
  • stardustborn likes this

#20 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 22 November 2020 - 09:53 PM

Next clear night I am going to open up the apeture, increase ISO and take shorter exposures, I am just very frustrated lately with my images.



#21 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 7,040
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 22 November 2020 - 10:19 PM

Next clear night I am going to open up the apeture, increase ISO and take shorter exposures, I am just very frustrated lately with my images.

The root cause is due to the gradual drift in frames over time coupled with an inherently noisy sensor. You PA is simply not good enough, nor your mounts tracking abilities, to allow you to shoot without guiding+ dithering.....

Using higher ISO/shorter subs is going to result in noise of a different kind...with correlated noise aka "walking noise" still always present in the background..

You need a better mount or the other unpopular option of a better camera like a Nikon D5300/D5500/D5600.


Edited by 17.5Dob, 22 November 2020 - 10:26 PM.


#22 Jerry Lodriguss

Jerry Lodriguss

    Vendor

  • ****-
  • Vendors
  • Posts: 7,792
  • Joined: 19 Jul 2008
  • Loc: Voorhees, NJ

Posted 22 November 2020 - 11:54 PM

Images plus is now free and has auto dark matching.

Depending on how old your Canon is you may have to convert to dng first.

Jerry

#23 Tragusa113

Tragusa113

    Explorer 1

  • -----
  • topic starter
  • Posts: 54
  • Joined: 31 May 2020

Posted 23 November 2020 - 11:49 AM

so does this mean all this data is basically useless?



#24 sharkmelley

sharkmelley

    Soyuz

  • *****
  • Posts: 3,912
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 23 November 2020 - 12:19 PM

so does this mean all this data is basically useless?

No, not at all.  As I said earlier, at worst you just need to shoot a new set of darks - at the higher temperature.  Then use dark scaling or dark optimisation or auto dark matching - different software calls it slightly different things.  Also, experiment with the threshold for pixel rejection during stacking.

 

You won't completely eliminate the walking noise but you should be able to reduce it considerably.

 

Mark


Edited by sharkmelley, 23 November 2020 - 12:22 PM.

  • Tragusa113 likes this

#25 DubbelDerp

DubbelDerp

    Apollo

  • *****
  • Posts: 1,237
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 23 November 2020 - 12:20 PM

If you can match your lights and darks more closely, then no, it is not useless. Before I started guiding and dithering in RA with my Skyguider Pro, I would go out of my way to try to match the temperature between the lights and darks within one or two degrees C. This had a significant impact on reducing walking noise. It's probably not as accurate as dark optimization as mentioned above, though.

 

Are you calibrating with a bad pixel map in APP? If not, that may help.

 

If you can guide and dither in RA with your Skyguider Pro, it will have a major impact on reducing walking noise and make your setup much less sensitive to the mismatch between the lights and the darks.


  • Tragusa113 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics