Why galaxy appearance is 'cartoon-like' using PixInsight?

Hello Wayne,

I think the second image looks different from yours and appears more "cartoon-like" not because of PixInsight itself, but probably because of the different equipment, integration time, and processing choices.

Looking at the second image, it also seems more saturated, and I think the imager may have used selective processing techniques to enhance certain structures and colors.

So, in my opinion, those factors are the most likely reasons why the image has a more "cartoon-like" appearance.

Clear skies!

I don’t mind either one, it’s all to taste and context. I think both versions are lovely in their own ways.

Image #1 could use: BlurXterminator to improve star profiles and non-stellar detail.

Colour calibration seems off, hence saturation and hues will be less than optimal. PI’s SpectrophotometricColorCalibration allows calibration according to many criteria and here clearly one would want to at least see how it looks using both an AverageSpiralGalaxy and the actual galactic structure class of the target as White References.

Other detail and contrast enhancement tools such as LocalHistogramEqualization and MultiscaleMedianTransform could used to further enhance both aspects.

Other software such as PhotoShop or Affinity may be of further use.

Image #2 seems to have crept over the line in my opinion on the sharpening front as well as on the saturation/colour enhancement. Looks ‘way too detailed for me unless it comes from a very large aperture setup in a very good location…

No idea how the two compare in terms of total integration time, selective subframe scoring and inclusion, and a host of other important variables.

I guess its all kind of subjective…

People have a tendency to really PUSH the saturation into galaxies, to draw your attention in from what is usually a large expansive field of black sky…

However, I’ll say this… Comparing two images captured and procecssed by different people, of different skill levels, using different equipment and processing software is by no means a valuable comparison.

What you should aim to do is, get a trial license of PixInsight, BlurXterminator, NoiseXterminator and StarXterminator… Load your M83 data into it, and process it using the different toolset.

See what a difference the correct tools makes for image processing, and compare ONLY against the results YOU can obtain, with YOUR data in both cases.

Only then can you really compare if one processing methodology provides better/worse or even ‘cartoonish’ results…

Likewise, you could say there is a drastic difference between the M83 you referenced in your post, and my M83 data from last year that were both processed in PixInsight, with all of the Xterminator tools being utilised… Neither of the images bares striking resemblence to each other, nor do they bare much resemblence to your own M83 data.

Mine: 3h integration with a 4.72” APO at f/5.6 from my B6 back yard.
📷 image.png
Yours: Unknown integration, newtonian reflector (I assume)
📷 image.png
Other: unknown integration, unknown scope (refractor?)
📷 image.png
I think it would be overly reductive to say that any of them was right or wrong, as it would be to say that because PixInsight, the ‘Other’ one looks cartoonish or otherwise ‘off’, or mine for that matter.

The three images are the result of 3 very different people imaging a target using 3 very different scopes, under very different conditions, with varying level of image acquisition skill or experience and then processing them based of 3 very different levels of processing skill, in different software packages with very different subjective tastes with regard to colour/saturation, star sizes and sharpening…

Ignore other peoples images when considering your own… They may have been doing this 1 month less than you or two decades longer. They may LOVE a highly saturated image while you don’t… They may have access to equipment and skies that far exceed yours, or have access to software you do not have that aids (and sometimes detracts, but that’s a WHOLE discussion of its own…) in image output quality.

Spend time with your scope under the sky.. Capture as much data on a target as you feel is necessary (don’t get drawn into the whole ‘well so and so did 165h so I have to do it too…’ nonsense)… Practice processing your images with the software you have available to you… process the image 15 different times in different ways, or change the order that you do certain parts of your workflow…

Don’t waste your time comparing your images to other peoples, unless you specifically feel like destroying your own happiness… Dare I say it, any target you image, Wolfgang Promper has imaged it before, and done it better in ways you can’t even articulate…

Make your images, if they make you happy - you’re done. If they don’t, practice… practice and hone your skills with acquisition, practice and hone your processing skills in whatever software you have available and try new tools/software to see if that helps… Remember there was a time when PixInsight didn’t exist, and 90% of amateurs were using DeepSkyStacker and either Photoshop or GIMP, and producing incredible images…

Really Wayne your image is close to being where you want. A C8 can produce a very sharp image. Make sure your collimation is dead on (lots of tutorials on internet) and that your focus is as well. Are you using something like a bahtinov mask for focus? If your image was just a little sharper and the stars less bloated it would have a beautiful very natural appearance. You’re almost there!

CS, John