Vignetting correcti...
 
Share:

Vignetting correction not working in my case  

  RSS

 UN73
(@un73)
Molecular Cloud Customer
Joined: 4 months ago
Posts: 3
13/04/2019 3:19 pm  

Hello,

whenever I try to use the CV tool I get a result which I can´t work with, see examples. Any ideas what´s happened? The LPR and CB tool seem to get rid of it overall as well but I am under the impression that a little bit of the vignetting still survives.

vignetting3
vignetting1
vignetting2

Best wishes,

Ulf

This topic was modified 3 months ago by UN73

astrogee and Acidlist liked
ReplyQuote
Topic Tags
(@gs-ascionegmail-com)
Molecular Cloud Customer
Joined: 2 years ago
Posts: 3
16/04/2019 9:57 am  
Posted by: UN73

Hello,

whenever I try to use the CV tool I get a result which I can´t work with, see examples. Any ideas what´s happened? The LPR and CB tool seem to get rid of it overall as well but I am under the impression that a little bit of the vignetting still survives.

I get the same results, every time i try to use this tool. My camera is a QHY128C, but I think this is independant by the camera. I wonder what is the right way to use the CV tool.

This post was modified 3 months ago by Acidlist

ReplyQuote
(@wvreeven)
Main Sequence Star Customer
Joined: 10 months ago
Posts: 42
16/04/2019 10:13 am  

Perhaps a silly question but do you use flat frames? If not, why not? Flats should allow APP to get rid of vignetting very well and eliminates the need of using the "correct vignetting" tool.


ReplyQuote
(@gs-ascionegmail-com)
Molecular Cloud Customer
Joined: 2 years ago
Posts: 3
16/04/2019 11:21 am  
Posted by: wvreeven

Perhaps a silly question but do you use flat frames? If not, why not? Flats should allow APP to get rid of vignetting very well and eliminates the need of using the "correct vignetting" tool.

I use them, actually it's the only way to get rid of vignetting. Tried to use the CV tool too, when flats are not perfect or if i coudn't use them for some reason. In that case I was not able to use CV tool at all, getting the same results showed by UN73


ReplyQuote
(@pete_xl)
Brown Dwarf Customer
Joined: 2 years ago
Posts: 8
16/04/2019 10:14 pm  

+1 from my side. The tool never produced any usable result for me. I come back to try again once in a while but the results are always unusable. I wonder what goes on or what I do wrong....


ReplyQuote
 UN73
(@un73)
Molecular Cloud Customer
Joined: 4 months ago
Posts: 3
21/04/2019 12:06 am  

ok, so it seems to be a common problem. perhaps mabula can say something about it, when he´s reading this. in the tutorial it seems to work like it should.

"Perhaps a silly question but do you use flat frames? If not, why not?"

with my eos 70d i use flats (and darks and biasframes). however in the m45 pic above i had faint diagonal noisy stripes which i couldn´t really get rid of with my usual processing technique. without flats, darks and bias the result was better.

and my modded 700d has sensor issues. when i use darks, flats and bias with that one, i get holes in the final pic. the holes result from pixels in the calibration frames that stand out but are not problematic in the lights. the lights however have sort of hot pixels which stand out colorful as the signal is only high in one or two channels and near zero in the other channel(s). sigma clipping takes care of that when i use the sw star adventurer as that one produces drift in dec. but the standardnoise is a pain. anyway, i don´t use calibration frames with that camera.


ReplyQuote
(@mabula-admin)
Quasar Admin
Joined: 2 years ago
Posts: 2091
25/04/2019 10:27 pm  
Posted by: UN73

Hello,

whenever I try to use the CV tool I get a result which I can´t work with, see examples. Any ideas what´s happened? The LPR and CB tool seem to get rid of it overall as well but I am under the impression that a little bit of the vignetting still survives.

vignetting3
vignetting1
vignetting2

Best wishes,

Ulf

Hi Ulf @un73, @gs-ascionegmail-com, @wvreeven and @pete_xl,

First of all, the Correct Vignetting tool is only to be used when you don't have flats for a dataset for whatever reason.

Secondly, shooting flats is always the best and recommended method 😉 to get the most optimal results.

Now, in your case, you are getting bad/improper results because of the way that you have placed the area select boxes. If you only place the select boxes in the 4 corners, there will not be enough data to correctly calculate a vignetting profile. You need to place the area select boxes evenly over the entire image, especially in the center of the image to make this work in the correct vignetting tool. A good method is to shoot a single frame on a part of the sky where there is no real object visible and use that to create an artifical flat. That artifical flat can then be used to correct your object data 😉

Data like your M45 shot, shown in the screenshots, will never be corrected nicely I think, in the correct vignetting tool, since the image data is simply filled with nebulosity. So I really recommend that you shoot proper calibration frames to get better results 😉 . The image data is filled with M45 so there is just very little information available to create a correct vignetting profile from well placed area select boxes.

"with my eos 70d i use flats (and darks and biasframes). however in the m45 pic above i had faint diagonal noisy stripes which i couldn´t really get rid of with my usual processing technique. without flats, darks and bias the result was better."

Perhaps there is an issue with how you create the calibration frames? Diagonal stripes seem odd as well. Usually such a camera has either vertical or horizontal stripes or both, but not diagonal.

Can you share such an example of diagonal stripes? are they visible in the bias and/or dark frames as well?

and my modded 700d has sensor issues. when i use darks, flats and bias with that one, i get holes in the final pic. the holes result from pixels in the calibration frames that stand out but are not problematic in the lights. the lights however have sort of hot pixels which stand out colorful as the signal is only high in one or two channels and near zero in the other channel(s). sigma clipping takes care of that when i use the sw star adventurer as that one produces drift in dec. but the standardnoise is a pain. anyway, i don´t use calibration frames with that camera.

Have you tried using a Bad Pixel Map? That would the correct way to correct pixels that behave oddly/non-linearly.

Kind regards,

Mabula

Main developer of Astro Pixel Processor and owner of Aries Productions


ReplyQuote
(@wvreeven)
Main Sequence Star Customer
Joined: 10 months ago
Posts: 42
25/04/2019 10:43 pm  

I have a modded 700D as well and flats work perfectly well for me. I use an Aurora flat panel with a folded white t-shirt to dim the panel so I can take 3 to 7 second exposures. Beware that too short exposures on flats may make the flipping mirror cast a faint shadow on one side of the flat frames which may ruin your images.

 

HTH, Wouter


ReplyQuote
(@rcfmitch)
Red Giant Customer
Joined: 10 months ago
Posts: 57
10/05/2019 5:47 pm  

On your 700D and in lower light what you are saying is that at 6-7 sec exposure at let say ISO 800 all your motes/vignette is gone ?

NO matter what speed i shoot my flats even at lower ISO they always overcorrect ( turn white) and that even happens when im set to AV in BYEOS or APT 3.63.2. any suggestion would help. mine are made late in the afternoon with Tshirt Sky flats.

never have problems with CCD/CMOS camera only DSLR and i have 3 of them .Also tried SGP wizard ,,, no go !!

sorry for butting into the thread

Mitch. 

 


1CM69 liked
ReplyQuote
(@1cm69)
Neutron Star Customer
Joined: 2 years ago
Posts: 121
10/05/2019 6:21 pm  
Posted by: rcfmitch

On your 700D and in lower light what you are saying is that at 6-7 sec exposure at let say ISO 800 all your motes/vignette is gone ?

NO matter what speed i shoot my flats even at lower ISO they always overcorrect ( turn white) and that even happens when im set to AV in BYEOS or APT 3.63.2. any suggestion would help. mine are made late in the afternoon with Tshirt Sky flats.

never have problems with CCD/CMOS camera only DSLR and i have 3 of them .Also tried SGP wizard ,,, no go !!

sorry for butting into the thread

Mitch. 

 

I am having this exact same issue with my Flats.

I do use a light panel along with T-Shirts to lower the light level, I even drop the ISO to 100 in order to get a longer exposure but still only achieve 0.6s with my Lum filter.

Kirk


ReplyQuote
(@rcfmitch)
Red Giant Customer
Joined: 10 months ago
Posts: 57
10/05/2019 6:35 pm  

Yes , the thing here is, its not the light source as much as it is the right ADU for a flat field.

my AA183C  ( 12 bit)  need 1850 ADu out of 4094 ( full well ) thats 46 % which is about right .

my friend has a ASI1600MM and he has to go up to 42,000 ADU on a 12 bit camera out of 65,530.

which is way closer to 70% . 

i was told on a DSLR you set your camera to AV shoot whatever ISO ( 400) lets say and let the camera do the exposure .. it NEVER works . perhaps next time i'm going to watch the histogram and try to hit 50 % ..

if not i will keep pushing the histogram till it hits 80 %.

 


ReplyQuote
 UN73
(@un73)
Molecular Cloud Customer
Joined: 4 months ago
Posts: 3
10/05/2019 9:08 pm  

hello mabula (and others),

thanks for the answers.

"Now, in your case, you are getting bad/improper results because of the way that you have placed the area select boxes. If you only place the select boxes in the 4 corners, there will not be enough data to correctly calculate a vignetting profile. "

yes, that makes sense...

"Perhaps there is an issue with how you create the calibration frames?"

i don´t think so. flats with a flat white shirt in front of the optical train and a bright white laptop screen held a bit away covering the whole aperture evenly, exposure setting to AV resulting in 0,3 - 0,4 s exposures, about 30 - 40. bias with minimum possible exposure time, cap on at end of the session, about 50. darks at the end of the session, with a lens + cap on and aperture set to minimum value (i.e. f/22), same exposure time as the lights, same iso setting, about 25. (@wvreeven: haven´t had an issue with that yet).

"Diagonal stripes seem odd as well. Usually such a camera has either vertical or horizontal stripes or both, but not diagonal.
Can you share such an example of diagonal stripes? are they visible in the bias and/or dark frames as well?"

it´s a bit odd indeed. it´s not like the coloumns one might get with dslrs. they are a bit wider, not so hard and not straight stripes but bend a bit in the picture which is why they are so hard to take care of in post-processing. they are not visible in the bias or darks.

i uploaded the app file of m45 with the eos 70d here. : https://www.dropbox.com/s/0mzc8le9dkye2df/St-avg-5496.0s-MWC_1_3.0_none-x_1.0_LZ3-NS-ref-qua-add-sc_BWMV_nor-AAD-RL-noMBB_1stLNC_it1-cbg-SC.fits?dl=0

best wishes,
ulf


ReplyQuote
(@wvreeven)
Main Sequence Star Customer
Joined: 10 months ago
Posts: 42
10/05/2019 9:18 pm  
Posted by: rcfmitch

On your 700D and in lower light what you are saying is that at 6-7 sec exposure at let say ISO 800 all your motes/vignette is gone ?

NO matter what speed i shoot my flats even at lower ISO they always overcorrect ( turn white) and that even happens when im set to AV in BYEOS or APT 3.63.2. any suggestion would help. mine are made late in the afternoon with Tshirt Sky flats.

never have problems with CCD/CMOS camera only DSLR and i have 3 of them .Also tried SGP wizard ,,, no go !!

sorry for butting into the thread

Mitch. 

 

Yes, that’s exactly what I am saying 🙂


ReplyQuote
Share: