RC8 or 250PDS for Galaxies

[Dan_Paris]

1 minute ago, vlaiv said:

this is enlarged with use of nearest neighbor interpolation - stars are not "pixelated" even when undersampled - that is artifact of interpolation algorithm.

Yes it was enlarged in the viewer if this is what you mean. Otherwise I barely see anything on my small-pixels screen !

2 minutes ago, vlaiv said:

What do you use for FWHM measurement? Different software will give you different results.

 

I use the FWHMEccentricity script of Pixinsight. I believe it is rather robust, and measurements are done over the whole image, not a single star.

 

 

 

[vlaiv]

1 minute ago, Dan_Paris said:

I use the FWHMEccentricity script of Pixinsight. I believe it is rather robust, and measurements are done over the whole image, not a single star.

I don't use PI so I can't comment that particular script, but I do know that FWHM measurement is very sensitive to background levels and if background have been removed. After all - it is about maximum and maximum will depend on removed background.

AstroImageJ uses aperture photometry and as such it automatically removes background influence regardless of any background removal done prior to measurement.

[Mandy D]

29 minutes ago, vlaiv said:

Your Lanczos resampled image is slightly different dimension than other two - you probably used different scaling up and scaling down factor?

But it is in any case much sharper than other example:

Left is Gimp resampled and right is Lanczos resampled one - difference is obvious. Try applying Lanczos 3 times - but keep dimensions the same instead of changing them.

No harm in doubting - that is the way we come to truth and true understanding. Everything must have an explanation.

Wierd! I have just copied my Lanczos image on this page into GIMP and it has exactly the same dimensions as your original and my previously edited ones. I took great care with the dimensions each time I scaled: 498/2 = 249.

Yes, I can see the difference between the GIMP image and the Lanczos one and I did compare the three times resampled one with the equivalent GIMP one and, yes, it is better, but not perfect. All the physics of this makes sense to me and should result in what you are stating, but what I am seeing is different. I suspect (as I vaguely suggested at the start) that the image processing software is the problem, not the physics.

Thank you for the interesting discussion. It has certainly advanced my knowledge of the subject.

[vlaiv]

1 minute ago, Mandy D said:

Wierd! I have just copied my Lanczos image on this page into GIMP and it has exactly the same dimensions as your original and my previously edited ones. I took great care with the dimensions each time I scaled: 498/2 = 249.

Yep, my bad - I expressed myself wrong - images are in fact the same dimension - but content is "zoomed" in. This is clearly software artifact.

You can see it if you open both images in browser - each in its own tab and blink between them. They are the same size, but contents is a bit zoomed in in second image. Here is animated gif made from two images (Baseline and your Lanczos x3 resampled example):

Don't know why it happened though.

[Mandy D]

29 minutes ago, vlaiv said:

Yep, my bad - I expressed myself wrong - images are in fact the same dimension - but content is "zoomed" in. This is clearly software artifact.

You can see it if you open both images in browser - each in its own tab and blink between them. They are the same size, but contents is a bit zoomed in in second image. Here is animated gif made from two images (Baseline and your Lanczos x3 resampled example):

Don't know why it happened though.

No problem, you are forgiven!

I think it is something that this website does with small images.