WBPP Local Normalization Failed

Hi guys,I ran wbpp and over half of the lights failed during local normalization. I really don't know the reason why. I ran 'Blink' before Wbpp and the frames looked fine though I didn't zoom in to check the stars. What could be the reason?

Was this a rerun of the same target. If so, try running again but clearing the cache (right side of main screen) before running. I've seen something similar when rerunning the same target after clearing out files. If that's not it… never mind…

Yeah it was a rerun but it failed the first time as well. I read the smart report and all of the 364 10 seconds subs and 20 subs at 60 seconds failed

Where is the histogram on the subs before stretching. I ran into this problem too when stacking images on an F2 newt running 3nm filter that were not shifted. My mean average was like 700 across alll the images. It took me a while to figure it out because the images looked fine in NINA while imaging. Under processes there is a statistics process you can run and it will tell you the mean. My guess is the images are too dim for the stacking software. But without seeing the data it’s just that, a guess.

In Local Normalization parameters check the grid size. I once had a number of LN failures and setting the grid size to 4 in my case solved the problem. Your optimum grid size may be different.

I've had issues where WBPP's normalization reference had issues. In the WBPP script, choose your reference instead of using Auto. Another option is to run LocalNormalization on the calibrated files to get a better idea of any errors.

Kevin Price:
Where is the histogram on the subs before stretching. I ran into this problem too when stacking images on an F2 newt running 3nm filter that were not shifted. My mean average was like 700 across alll the images. It took me a while to figure it out because the images looked fine in NINA while imaging. Under processes there is a statistics process you can run and it will tell you the mean. My guess is the images are too dim for the stacking software. But without seeing the data it’s just that, a guess.

*** Hi Kevin,thanks. I haven't stretched it yet ***

Abdul Thomas Jnr:
Kevin Price:
Where is the histogram on the subs before stretching. I ran into this problem too when stacking images on an F2 newt running 3nm filter that were not shifted. My mean average was like 700 across alll the images. It took me a while to figure it out because the images looked fine in NINA while imaging. Under processes there is a statistics process you can run and it will tell you the mean. My guess is the images are too dim for the stacking software. But without seeing the data it’s just that, a guess.

*** Hi Kevin,thanks. I haven't stretched it yet or am I missing something?***

Peter Hannah:
In Local Normalization parameters check the grid size. I once had a number of LN failures and setting the grid size to 4 in my case solved the problem. Your optimum grid size may be different.

*** Hi Peter. The grid size is on default of 4.0***

Charles Bradshaw:
I've had issues where WBPP's normalization reference had issues. In the WBPP script, choose your reference instead of using Auto. Another option is to run LocalNormalization on the calibrated files to get a better idea of any errors.

*** Hi Charles,how do I choose the reference frame manually?***

In the script choose reference frame auto/manual just above the output directory. In local normalization you have a choice of "single best frame" or integrated. You can play with both. Remembering back, the issue with reference frame I had was with registration. It may not affect LN, but worth a try.

Remember to adjust your path name allowance if this hasn't been done. Not doing so can cause precisely this issue.
-adam

I saw on Facebook group today that someone had similar problem after recent update of pixinsight. If you also updated to latest version, it may be the root cause.

Adam Block:
Remember to adjust your path name allowance if this hasn't been done. Not doing so can cause precisely this issue.
-adam

*** Hi Adam,thanks for your input. How do I adjust the path name allowance ? ***

I saw on Facebook group today that someone had similar problem after recent update of pixinsight. If you also updated to latest version, it may be the root cause.

*** it could be but i dont think mine has been updated recently. Unless it manually update itself***

What is the algorithm you use for the Subframe Weighting of your frames, PSF Signal Weight (default setting) I assume? Try PSF scale SNR, since I use this algorithm, I never have an error with LN (the influence by the weight granted could be very low on subframes that are in default).

Abdul Thomas Jnr:
Adam Block:
Remember to adjust your path name allowance if this hasn't been done. Not doing so can cause precisely this issue.
-adam

*** Hi Adam,thanks for your input. How do I adjust the path name allowance ? ***

https://www.youtube.com/watch?v=J-zLANFB2WU
-adam

Christian Proulx:
What is the algorithm you use for the Subframe Weighting of your frames, PSF Signal Weight (default setting) I assume? Try PSF scale SNR, since I use this algorithm, I never have an error with LN (the influence by the weight granted could be very low on subframes that are in default).

*** Okay will try this and see what I get. Thanks ***

Related discussions