Milky Way Pano Stacking and Stitching Worflow - Issue with Lens Profile Correction w/ TIFFs post-Stacking - Forums

Topic starter

May 11, 2022

Hi People,
I am looking for some suggestions on how to stack and merge a Milky Way Panorama shots.
I know the process and have watched quite a few tutorials on Youtube.
This might sound like a naive question but I just want to try my luck here.
The workflow I have been trying to follow is to stack the images in Sequator first (Output is TIFF files), Profile Correct (Lens Correction) in LR (TIFF files) and then merge the Pano shots.
The issue I am seeing with this flow is that when I try to apply the Profile Correction in LR to the exported TIFF files from Sequator, LR doesn’t show the Lens Profile for my lens.
If I load the the RAW (CR2) files in LR, it does let me choose the correct profile (for my lens that I used) however with TIFF I only have limited profiles available.
I read on Adobe forums that files other than RAW files, will not have the Lens profiles or have that EXIF data and we might have to do that manually but I haven't come across a process showing that or if even that is possible. BTW, I tried that today with no luck. Doesn't seem to be saving Lens Configuration to TIFF file.
I tried doing the Lens Profile Correction prior to stacking but that created some artifacts in my images (which is what I read and was doing the stacking first). These images were taken using Canon DSLR (APS-C).
I read somewhere that mirrorless or Sony Mirrorless doesn't create those artifacts if Lens Profile correction is done prior to stacking. I do have a Sony Mirrorless camera too but I haven't tried this flow with Sony (ARW) RAW files.

I also tried DSS and I think I didn't see any artifacts when doing Profile correction first and stacking them so might take this way but the only thing I am not sure about is that DSS doesn't let you choose the sky part (like in Sequator) so the landscape part stacking might not turn out well or is it something that DSS takes care off?
Can someone advise what I should or shouldn't be doing here and if there is a better way to do all this.
I am kind of curious to know how you people are doing it to get the lens profile information correctly populated for TIFF files for Canon DSLR shot images specifically or there is a way around it.
Constructive feedback will be much appreciated
Thank you in advance!
Clear skies to all

Helpful Engaging

Boyan Stiliyanov

May 11, 2022

When you work with a RAW file and convert it to TIFF then the metadata (lens corrections, white balance, focal lenght etc.) are removed. In fact you can see the precise Kelvin temperature on your white balance in RAW photos, but on TIFF files you are only given the scale -100 to 0 to 100.

Well I recommend stitching the photos automatically (Photomerge), but editing manually the TIFF file. What I mean is, yes Lens Corrections is useful to remove vignetting, distortion and chromatic aberrations, but shouldn't you be doing that manually in Photoshop? I'm not sure that Lightroom works well for astrophotos. In Photoshop you can precisely tune exactly the colour you want to remove (CA), or the amount of distortion and vignetting you want to change.

Also if you overlapped the images, which you should definetly do if you want to stitch, then the corners of the images shouldn't be too problematic, because they are cropped anyways.

Searching a way to get a lens profile on a tiff is too much hassle and in the end you have limited settings. Try Photoshop, see how to remove CA, distortion and vignetting and you have more freedom. Plus if you have a gradient on your image due to light pollution or other lights, you will certainly need to use Photoshop to fix it.

Helpful

nickastrobin

Topic starter

May 12, 2022

When you work with a RAW file and convert it to TIFF then the metadata (lens corrections, white balance, focal lenght etc.) are removed. In fact you can see the precise Kelvin temperature on your white balance in RAW photos, but on TIFF files you are only given the scale -100 to 0 to 100.

Well I recommend stitching the photos automatically (Photomerge), but editing manually the TIFF file. What I mean is, yes Lens Corrections is useful to remove vignetting, distortion and chromatic aberrations, but shouldn't you be doing that manually in Photoshop? I'm not sure that Lightroom works well for astrophotos. In Photoshop you can precisely tune exactly the colour you want to remove (CA), or the amount of distortion and vignetting you want to change.

Also if you overlapped the images, which you should definetly do if you want to stitch, then the corners of the images shouldn't be too problematic, because they are cropped anyways.

Searching a way to get a lens profile on a tiff is too much hassle and in the end you have limited settings. Try Photoshop, see how to remove CA, distortion and vignetting and you have more freedom. Plus if you have a gradient on your image due to light pollution or other lights, you will certainly need to use Photoshop to fix it.

Thanks Boven!

I am using LR for Merging the Pano shots as PS is giving me weird results (with either of the options). I was able to get past the issue I was seeing earlier with my stitched image so that's a good thing.

I did Lens Correction and adjusted white balance to my CR2 files and then stacked them using Sequator. Resulting TIFF images were merged in LR to get the stitched-Pano image.

Will I loose anything by editing the TIFF files and not the CR2 files?

I will be editing the stitched photo in PS as you mentioned and from what I have seen so far it gives more controls however I am just a beginner with PS so will be editing side-by-side while watching the videos on Youtube. If you have any recommendations on PS tutorials, let me know else I am going to look on Youtube anyway.

Regards,
Nick

James Peirce

May 12, 2022

In my experience, using lens profile corrections from Lightroom prior to exporting TIFF files does tend to create artifacts. And using the in-camera lens profile corrections which bake into RAW file's (e.g. Sony's Lens Shading Comp) have created their own artifacts. Sony's are particularly bad.

I've simply depended upon capturing flat frames to correct vignetting and a more basic soft of correction in post-processing (e.g. cropping, registration with distortion correction, or just doing a manual profile correction adjustment on the processed image). As well as stopping down the lens a little if stars are bad. If the distortion is just really terrible on the perimeter I plan that into the shoot and crop, or just allow it to slide on the grounds that an ultra-wide Milky Way landscape photograph tends not to involve pixel-peeping stars in corners.

Helpful Insightful Engaging

Wei-Hao Wang

May 13, 2022

The lens profile correction mainly does two things:

1. distortion correction
2. vignetting correction

For #2, it can be replaced with flat field correction during calibration and stacking, either in Sequator or DSS. You just need to take flats and bias. It is much more accurate than the vignetting correction in the profile correction, even if you can do profile correction without the problems you describe.

For #1, I am not very positive that you actually need it. When you try to stitch the panels into a mosaic, most of the stitching programs have some capability to automatically figure out the distortion and correct it for you. It doesn't have to be corrected by profile correction.

Regardless of the above details, generally speaking, my advice is to get away with profile correction once for all. Do standard bias and dark subtraction and flat fielding in your stacking. That's the right way that will lead you to high quality results.

Helpful

nickastrobin

Topic starter

May 13, 2022

James Peirce:
In my experience, using lens profile corrections from Lightroom prior to exporting TIFF files does tend to create artifacts. And using the in-camera lens profile corrections which bake into RAW file's (e.g. Sony's Lens Shading Comp) have created their own artifacts. Sony's are particularly bad.

I've simply depended upon capturing flat frames to correct vignetting and a more basic soft of correction in post-processing (e.g. cropping, registration with distortion correction, or just doing a manual profile correction adjustment on the processed image). As well as stopping down the lens a little if stars are bad. If the distortion is just really terrible on the perimeter I plan that into the shoot and crop, or just allow it to slide on the grounds that an ultra-wide Milky Way landscape photograph tends not to involve pixel-peeping stars in corners.

Thank you James for sharing your thoughts and inputs. Much appreciated!

Well Written Respectful

nickastrobin

Topic starter

May 13, 2022

Wei-Hao Wang:
The lens profile correction mainly does two things:

1. distortion correction
2. vignetting correction

For #2, it can be replaced with flat field correction during calibration and stacking, either in Sequator or DSS. You just need to take flats and bias. It is much more accurate than the vignetting correction in the profile correction, even if you can do profile correction without the problems you describe.

For #1, I am not very positive that you actually need it. When you try to stitch the panels into a mosaic, most of the stitching programs have some capability to automatically figure out the distortion and correct it for you. It doesn't have to be corrected by profile correction.

Regardless of the above details, generally speaking, my advice is to get away with profile correction once for all. Do standard bias and dark subtraction and flat fielding in your stacking. That's the right way that will lead you to high quality results.

Thank you Wei-Hao! I will keep that in mind. Your inputs are much appreciated!

Well Written Respectful Supportive