Image Integration giving very low background ADUs in the stacked image

Niall MacNeill

New member
Sep 20, 2020
1
0
I have been processing my captures of NGC 6744. All was going well until it came to the Green channel. All the Registered Lights have a background ADU of ~ 110-120. The MasterStack however had a background ADU level of ~ 14. How does the integration of images end up with so much lower a background ADU than the individual subs. None of the other colour channels were affected. Note I added a pedestal of 100 at the time of calibration. I wondered if this would occur for a smaller range so I split it into two. The 2nd half when integrated gave me ~ 175 ADU, but the first half of the Register Lights gave me the 14. And so I honed it down. Eventually by going backwards and forwards it seems to be down to one sub, which if included gave a low background ADU. So then I selected everything and excluded that sub. You guessed it, the background ADU was still 14. I tried excluding one or two each side....same result. In the end, I chose from about 3 subs after the one I found to be suspect and then selected the remaining subs, some 60 in total and the final result came out with a background ADU of 175.
I am quite perplexed. Any ideas?
Regards, Niall
 

ngc1535

PTeam Member
Feb 1, 2014
560
77
AdamBlockStudios.com
I do have some ideas- but it would only lead to me saying "yeah huh!" and you saying "nuh uh!" .
So... it is probably best to make the data/images available to be evaluated. My guess is a quick look
at the data will glean important answers (or better questions to get the answers!).
-adam
 

XcalRocketMan

Member
Jul 9, 2014
9
0
Interesting ... I was just about ready to post about a similar issue, but a bit different than the OP. In my case, the Red and Green integrated lights show ADUs of about 310-420 while the Blue light shows 2055. Individual subs are fairly equivalent, generally in the range 300-420 or so. I can send images if you'd like.
(edit:)
I've attached a link to my Dropbox, which hopefully everyone can access. I've been able to determine that a group of blue lights is the center of the problem. They are numbered 4,5,6,7,8. If you integrate 4,5,6 all is good. Integrating 6,7,8, all is good. Both give ADU means of about 300 or so. But if you integrate 5,6,7 you get ADU mean of 2060. Yet, the image statistics for each file is reasonable and consistent.

Link: https://www.dropbox.com/sh/9yhdt6jdpacqzow/AAArTQ56qjGKg58lhkWKFawxa?dl=0
 
Last edited:

bulrichl

PTeam Member
Nov 2, 2016
859
65
La Palma, Canary Islands
In ImageIntegration's default setting, the option 'Truncate on out-of-range' is disabled. The mouse-over text says:

"If the output integrated image has saturated pixel samples out of the nominal [0,1] range, truncate them instead of rescaling the whole image.

No out-of-range values should occur after integration of a well-calibrated data set under normal conditions. However, sometimes saturated pixels may lead to out-of-range values after output normalization, depending on the frame selected as integration reference.

When this happens, the best option for integration of light or science frames is a linear rescaling, which preserves all of the integrated data. However, in some cases altering all pixel values is not admissable, so a rescaling operation is not applicable. This is the case for the integration of flat frames, where truncation is the only option available to preserve the correct illumination profile in the integrated master flat frame."

With option 'Truncate on out-of-range' disabled, the following results are obtained:
When integrating images 6, 7, 8, no rescaling is applied.
-> mean about 320 ADU.

When integrating images 5, 6, 7, rescaling is applied:
* Rescaling output image. Integration range: [-5.47401578e-06,1.55479729e-01]
-> mean about 2067 ADU.

With option 'Truncate on out-of-range' enabled, the following results are obtained:
When integrating images 5, 6, 7, no rescaling is applied:
-> mean about 320 ADU.

My conclusion is: for the integration of calibrated, registered light frames, the default setting (option 'Truncate on out-of-range' disabled) is OK. The high level of the blue channel will be corrected in the color calibration.

Bernd
 

XcalRocketMan

Member
Jul 9, 2014
9
0
In ImageIntegration's default setting, the option 'Truncate on out-of-range' is disabled. The mouse-over text says:

"If the output integrated image has saturated pixel samples out of the nominal [0,1] range, truncate them instead of rescaling the whole image.

No out-of-range values should occur after integration of a well-calibrated data set under normal conditions. However, sometimes saturated pixels may lead to out-of-range values after output normalization, depending on the frame selected as integration reference.

When this happens, the best option for integration of light or science frames is a linear rescaling, which preserves all of the integrated data. However, in some cases altering all pixel values is not admissable, so a rescaling operation is not applicable. This is the case for the integration of flat frames, where truncation is the only option available to preserve the correct illumination profile in the integrated master flat frame."

With option 'Truncate on out-of-range' disabled, the following results are obtained:
When integrating images 6, 7, 8, no rescaling is applied.
-> mean about 320 ADU.

When integrating images 5, 6, 7, rescaling is applied:
* Rescaling output image. Integration range: [-5.47401578e-06,1.55479729e-01]
-> mean about 2067 ADU.

With option 'Truncate on out-of-range' enabled, the following results are obtained:
When integrating images 5, 6, 7, no rescaling is applied:
-> mean about 320 ADU.

My conclusion is: for the integration of calibrated, registered light frames, the default setting (option 'Truncate on out-of-range' disabled) is OK. The high level of the blue channel will be corrected in the color calibration.

Bernd
Thanks for the info. So I gather I'm OK with proceeding as normal? Nice to know the background reason for the odd values showing up :)