selection of MureDenoise script parameters

[Dmitry]

Hello!
I really like the work of the MureDenoise script!
Unfortunately, some processes (LocalNormalization, ImageIntegration, DrizzleIntegration, DynamCrop with image rotation) lead to a change in the image noise scale and the size of the pedestal.
Selection of parameters is not complicated, but takes a lot of time.
Is it possible to add a script that could be applied to the image preview to evaluate the best pair of parameters “Offset” and “Gaussian Noise”?
The user could set the range of interest and the increment for each parameter. Then the parameter selection script could launch MureDenoise with each pair and choose the combination that will create the best signal-to-noise ratio of the image.
Sincerely, Dmitri.

[mschuster]

Thank you Dmitry for your comments and suggestions. Unfortunately the script's current denoising technology is based on a limited noise model. If the noise statistics differ from the model, the results are undefined. In this situation (ie, the results of several of your processes) I know of no way to pick parameters to meet a desired criteria, other than basically what you are doing: trial, error, and subjective evaluation. I do have a plan to support LocalNormalization, beyond that however the situation remains an open problem.

[Dmitry]

Thanks for your reply. In my humble experience, the MureDenoise script gives very good results after LocalNormalization and after DrizzleIntegration and after DynamicCrop with rotation.
It is only necessary to select the best pair of parameters “Offset” and “Gaussian Noise”. The only exception is the DrizzleIntegration process, for which the Bilinear method gives slightly better results compared to Lanczos-3.
The SNR criterion for the selected preview with the background area is all I need.
Here are my values ??for processing the galaxy M110, channel Blue, the integration of 16 images with a pedestal 320DN. The image scale is 0.5 arcseconds per pixel. Camera ASIZWO 1600mm_s, gain G76. :
"Mask: ATWT1" - additional noise reduction by the ATWT process: removal of the first layer with an inverse brightness mask.