Strange pattern in Light Pollution Tool

That is quite the strange pattern indeed. I'm wondering if this is an issue in the data itself. Can you try to integrate without any calibration data and then see what happens?

Sure thing Vincent, I will do so now and post a photo in a few hours...

Well this isn't good!  When I stack just my lights, you can clearly see this pattern embedded in the image... then it is really brought out in the light pollution tool and you can see the way it is processing the data.

Very strange... this is 250 30 second subs with an asi294MC pro at -10, through an 8" celestron edge HD and .7 focal reducer.  No filters, no other thing I can imagine that would cause this.  Doesn't look like a faulty cable.  How strange.  Any ideas?

@luckybudd Are you shooting 8 bit or 16 bit images? Can you check in the data? If you load a light in APP and double click it in the file list, it will be opened in the image preview window and the FITS header gets displayed. There it should inform you about this.

Ha, good one! I was also thinking about maybe a data issue with power-noise on a dataline, but this indeed looks more like 8-bit or something.

Hey! 

Ya. When looking at my lights, under “color space” it says “16b RGB”. The only change I see is that under “Bad Pixel Map” it says “8b Gray”

@luckybudd Do you get the same strange noise pattern when you redo the integration but without using the BPM? Just to see if that makes a difference?

Wouter

I can try that. Didn’t know I could leave that out! I still have the file open so I can try! 

Well, I don't know if this is related, but when I integrate without a Bad Pixel Map, the RGB integration file comes out as a 32 bit image.  I assume that is normal.  Otherwise, everything in JUST A STACK OF LIGHTS is 16b.

@luckybudd But do you still see the strange noise patterns in the stack without BPM?

@wvreeven I do.  I even tried cropping RIIIIIGHT into the galaxy and trying the tool from there and same result.  So, there is either an issue with the lights themselves (a camera issue?) or it is picking up something REAALLY wonky in the processing...  correct?

@luckybudd Right. Would you mind uploading your data set to https://upload.astropixelprocessor.com using upload2 for both username and password? Please create a directory called luckybudd_wonky and put the files in there. Then we can have a look and see if we can find out what's going on.

I have put in a call to ZWO to discuss with them if it is something on their end, so any info would be GREATLY appreciated 🙂  I just tried the link you sent and apparently it isn't working currently.  Is there another?

Sorry, I changed the url to https://upload.astropixelprocessor.com and it works.  Uploading now.  Thank you VERY much!!  

@luckybudd Sorry for the typo in the site name 🙁

@wvreeven It is all good!  I really appreciate your help!  I am able to work with what I am stacking in lightroom to get a passable image, but obv there is something very wrong here!

Oh!  Do you want me to upload all data?  Or just lights?  I can also do a master dark/flat/dark flat to save time if you like

@luckybudd Yes, files of all the types you have please!

On their way!!  Thank you!!!!

Ok!  All files are uploaded!  

@luckybudd Thanks. I started downloading earlier this day and played a bit with the data. I haven't been able to find out what's going on yet but will keep on looking.

Much appreciation!  I tried processing on two different computers (a Mac with M1 and an old Lenovo running Windows 10) and when I use that light pollution tool - same result.  

Hey,

So I went back and restocked my Rosette from last month.  This image was taken with my Sharpstar 61 (f4.5) and an IDAS NBZ filter.  The pattern the Remove Light Pollution tool came up with is like amazing abstract art!  Anyhow, I dont know if this is helpful, but it this has been an issue with every target I have used this camera on.  I have only had the camera a little over a month...

I think I figured it out!!!

When I batch modify my stack and crop it, I have been saving the files as 16 bit TIFF files.  From there, if I "remove light pollution" it makes the image all crazy and wavy like the images I have been showing you.

I just did a test where I batch modify to crop and save as a fits file.  From there if I remove the light pollution, the lines are gone!  See attached images.  

I will add the darks and dark flats now and process the full image.  Were you seeing anything else of concern with my data that would cause this?  Or are we all good?

@luckybudd Ah yes that would indeed explain this. Glad that you figured it out. I didn't find any issues with your data.

@wvreeven thank you so very much for your time!!  I can't say enough about the help and support the community here has offered!  I am learning so much and really appreciate it! 

Hi @luckybudd & @wvreeven,

I understand from Wouter that the issue is solved?

Do I understand correctly that you were cropping and saving the data in another application as TIFFs?

Is there a particular reason to start cropping? Normally this should not be needed.

Those TIFFs probably were saved with a lossy compression causing an effect called posterization, that is what you see in the data.

Cheers,

Mabula

Hi Mabula,

I am only using APP.  What happened is that I took these images as a series of 30 second subs on an Evolution Alt-Az mount with no guiding.  So, I get some drift.  The final stack looks like the attached image below.

My process is that I Batch Modify to crop and make the image square.  At this point APP asks me if I would like to save as a FITS, TIFF or JPG.  For no real reason (habit perhaps), I click on TIFF and then move on to Remove Light Pollution (also asks if I want to save as FITS, TIFF or JPG). Then I move on to Star Colour or just go right to stretch and save.

So, the weird banding was happening when I would save the Batch Modify as a TIFF instead of a FITS file.  I have learned that I will NOT save anything as a TIFF until my final stretched image.

I was thrilled to learn that there is no issue with my data or the software, just my process!  Thank you!

Hello Lucky Budd, I have had a look at your data. I also use the ASI294MC Pro and the EdgeHD. The problem seems to come from the flats and darkflats.

The MasterFlat has a color gradient.

The DarkFlat has an eye-catching circular pattern in the center. Looks like a light leak to me (possibly the side vents from the main mirror and or a filter drawer).

The MasterDark looks good.