[Solved] Problem with star analysis 1.075

Hello

I think I have a similar message with 1 Ha field with M42 on it. Other fields near this find 1000 stars where this one make an error with only 11 stars

I will try to send a link for calibrated FIT of 1 working image and 1 not working.

Cheers

It depends on the actual data I think, could you upload one of these subs that are failing?

hello

Here are 2 files (calibrated with MB, MD, MF under APP) and I also give you the original files

LINK TO IMAGES DIRECTORY

Try to make a mosaic with these 2 images 
My goal was to verify pointing of 12 fields and check overlap.

I got error message due to lack of stars on M8 field

cheers

philippe

Dear Vincent,

Please find enclosed two raw files as well as the calibrated files (bias corrected plus the new cosmetic correction features).

File number  2 is star analyzed, although with extremely low number of stars. File number 3 is one of the many files that failed.

Version 1.074.1 does not show these problems.

Best, Ruud

Thank you both! Sorry for the slight delay, life got in the way. I'll have a look at these asap.

Posted by: @ccd2048

Try to make a mosaic with these 2 images 
My goal was to verify pointing of 12 fields and check overlap.

I got error message due to lack of stars on M8 field

So yes I'm getting the same. It seems there is a problem already with the separate integrations. When I look closely at them I see artefacts like the picture below, that should be better first I think.

Posted by: @pol75rex

Please find enclosed two raw files as well as the calibrated files (bias corrected plus the new cosmetic correction features).

Thank you Ruud. There aren't many stars visible, but it should be enough. The only thing I see with the one that fails is that the stars are elongated. But I don't think that should be a problem. I'll let Mabula know, thanks for sharing.

Posted by: @vincent-mod

Posted by: @ccd2048

Try to make a mosaic with these 2 images 
My goal was to verify pointing of 12 fields and check overlap.

I got error message due to lack of stars on M8 field

So yes I'm getting the same. It seems there is a problem already with the separate integrations. When I look closely at them I see artefacts like the picture below, that should be better first I think.

 

Hi

I think you didn’t understand : this is NOT integrated image, but RAW images.

Artefacts, yes, created with APP !!!

But, for me (and not only me) is why it is working on 1.074.x and not on 1.075

please take RAW image, no artefact, same issue

PIXINSIGHT can find a lot of stars on any image

Dear Vincent / Mabula,

Version 1.074.1 detects around 40-50 stars. The latest version detects only 5 to 10 stars. An unexplained difference between both versions.

Stars are elongated due to oscillation in RA axis compared to Dec axis, but that is unrelated to the problem I identified.

Best, Ruud 

I have reported it to Mabula and send some of the data to him, I hope he has time soon to have a look at it as well.

Hi Vincent,

I tried on some of the subs I took lately ( that stacked well

in APP ) and analyzing 1 sub in 1.075 gives me aprox. 60 stars whereas

in PixInsight doing FWHMEccentricity on the same sub gives

me 89 stars.

Just analyzed the same sub in 1.074 that gives me 106 stars!

Food for thoughts?  :-O

Hope to see some more feedback on this! ( filesize is too big to attach, 32 Mb )

Best regards,

Mert

I'm tagging Mabula @mabula-admin so hopefully he can chime in on this one.

Hi Ruud @pol75rex, Philippe @ccd2048, Mert @lammertus & Vincent @vincent-mod,

First of all, if you look at the extensive release notes of APP 1.075, you will see that something changed between APP 1.075 and 1.074.1 with respect to Star Analysis/Detection :

• FIXED/IMPROVED, 3) STAR ANALYSIS, the star analysis module now fully supports severely undersampled data. Undersampled data means that the star sizes in Full Width Half Maximum (FWHM) are less than 2 pixels. APP will now be able to process these frames as well. Undersampling could occur due to having too large pixels for your focal length and atmospheric conditions, or due to binning of your camera's sensor for instance.

• IMPROVED, STAR ANALYSIS, further improvements have been implemented in star detection to make it more robust on both data with small and large FWHM values. Especially discrimination between hot pixels and small stars has been further improved. Further improvements were triggered by this topic where registration was done on hot pixels instead of stars: https://www.astropixelprocessor.com/community/main-forum/my-stars-wont-align-after-integration/

Clearly, these adjustments fail to work on your data for some reason. My apologies for this inconvenience, I did test it extensively on a wider range of images with different FWHM values. I will download all of your data and I will do further testing to make sure it works properly again in APP 1.076.

Kind regards,

Mabula

@lammertus, can you share some of your frames possibly? Maybe send it to me using wetransfer.com ? use support@astropixelprocessor.com 😉

Mabula

Hi Ruud @pol75rex, Philippe @ccd2048, Mert @lammertus and Vincent @vincent-mod,

After extensive work in the past few weeks, I have finally finished work on a greatly improved Star Analysis module:

https://www.astropixelprocessor.com/community/release-information/astro-pixel-processor-1-076-preparing-next-release/ "

BIG IMPROVEMENT, STAR ANALYSIS, extensive work has been done on improving star detection and analysis, due to some users reporting issues with Star Analysis in APP 1,075 (  https://www.astropixelprocessor.com/community/rfcs-request-for-changes/problem-with-star-analysis-1-075/ ).  I am happy to announce that all reported data that gave issues, now works flawlessly. The Star Analysis module is also twice as fast now while becoming much more robust and memory consumption is kept very low as well. Users will notice that APP will detect more stars on more datasets as well. Here are some performance graphs that show Star Analysis of 100 H-alpha frames of 5 different optical setups, for 4 CPU threads, 8 CPU threads and 16 CPU threads, APP 1.075 versus APP 1.076:

Ruud, on your data, 40 stars are again found 😉

Philippe, on your images, about 2500 stars are now detected robustly and I can register the 2 frames as well:

I will try to release a BETA as soon as possible so you can test the improvements yourself 😉

Mabula

Thank you Mabula ! Looking forward to test beta version 😉 

Dear Mabula,

Thanks for solving this issue. I will try the beta version asap after it's release.

Best, Ruud