Cookie consent

AstroBin saves small pieces of text information (cookies) on your device in order to deliver better content and for statistical purposes. You can disable the usage of cookies by changing the settings of your browser. By browsing AstroBin without changing the browser settings, you grant us permission to store that information on your device.

I agree

Some doubts about an image...

Jean-Baptiste_Paris
11 Oct, 2019 08:43
Good morning all,

Sorry, I usually do not do that, and I feel a little embarrassed to ask the question so directly, but … what do you think of this picture?


https://astrob.in/6wmljt/0/

I ask the question because this image has been pushed into the list of the IOTD and can soon be a TP.

The resolution obtained on this image is impressive, especially the very fine details in the dust bands; equivalents or better than those obtained for example with a telescope of 500mm (an example with one of my images made in remote in Chile : https://astrob.in/409229/D)… while this image is supposed to have been taken with a refractor of 80mm diameter, 480mm focal and 8nm filters !

I suspect a "fraud", not on the image as a whole, but I think the details may have been "stolen" from another image …

Maybe I'm totally wrong… what do you think?

Thank you !
JB

PS: I may have post this in "IOTD Staff", but I think it is better to avoid a large visibility.
Edited 11 Oct, 2019 08:47
Andys_Astropix
11 Oct, 2019 09:21
smile Interesting & well spotted.
Great image but the image scale is also wrong imo. An 80mm refractor & that sensor should surely have the entire rosette Nebula in frame.
Looks more like a 130mm to me?
Hmmmm - thoughts? 🤔
Jean-Baptiste_Paris
11 Oct, 2019 09:25
Well, sorry again… I don't know what's going on today… am I becoming totally paranoid?  smile

But what do you think of that image too :


https://astrob.in/1lqnt8/0/

This image is supposed to have been taken with a 250mm scope and a 0.8" sampling.
I made an image of this galaxy last year with a RC 1m scope and 8m focal (0.23" sampling) : it was very far from that resolution (https://astrob.in/367700/L/).

Ok, it was in remote and in such case you don't have the possibility to easily judge the quality of seeing, and I am not a processing superstar either… so my image is not a reference.
But there are very few images of this galaxy made by amateurs ; and even the image made by Damian Peach with a 20" telescop has a not a resolution as good as my image, and even more as this image (http://planewave.com/cartwheel-galaxy-by-damian-peach/).

More disturbing : this image seems to be even more detailed than the HST image, specially in the core of the galaxy (https://www.spacetelescope.org/images/potw1036a/).

How can you explain that ?
I would understand if such an image have been made with a large scope and lucky imaging process in order to void the atmospherical turbulence ; but here the luminance layer is made with 20min subs…

There are many other curious things… for exemple, the level of details in the galaxy is far more high than the stars, or even the very faint and more distant galaxy just aside. Even on my image, I have more details in this little galaxy in background than on this image… this is not consistent imo.

What do you think of that ? am I totally paranoid ?  smile

jb

PS : this image is now in the IOTD list.
Edited 11 Oct, 2019 09:50
Jean-Baptiste_Paris
11 Oct, 2019 11:59
Something really strange about this last image : compare the stars inside and outside of the disk of the galaxy… the stars inside the galaxy are about 10x finer than the same stars outside…

How the stars inside the galaxy can be so fine, compared to the stars of same magnitude outside of the galaxy ?
Moreover, the stars of same magnitude (as the stars inside of the disk) are nearly not even visible outside of the galaxy.

On my own image of this galaxy (https://www.astrobin.com/367700/L/), the stars of same magnitude nearly have the same aspect inside and outside ; what is logical I believe…

Please have a look at this comparison made on full images (the background image is mine, the insert one is Terry's image).



https://astrob.in/full/d4vpq2/0/?real=

Note that all stars of same magnitude have the same aspect and size on my image.
Also note the little galaxy in background : on my image, there is some details that are consistent with the level of details in the main galaxy.

The more I look at this image, the more I am convinced that something is wrong and even it can be a fake…

I hope I'm wrong… I leave him the benefit of the doubt for the moment but I would like to know how he can achieve such processing performance !

jb
Edited 11 Oct, 2019 13:46
jtrezzo
11 Oct, 2019 13:31
I agree, something fishy about the first one you posted. Especially given its only the first two images posted by the guy.
Terry's image you seem to be on to something as well. There's a few guys out there that secretly enhance their images with Hubble Luminance data, and if you call them out on it will go haywire, but it is so obvious. I think that may be what we are seeing here. I've seen another fairly well known imager using a 130mm scope trying to pass off that he is splitting the same close stars as Hubble, which is just not possible. What you have posted of the Cartwheel up close certainly seems indicative of that - there is no way that much detail was captured at 0.8" pixel scale on a 1'x0.9' object. If so I'd also like to know how. It looks the exact same detail as the Hubble image.
Edited 11 Oct, 2019 14:08
Jean-Baptiste_Paris
11 Oct, 2019 14:06
Jarrett Trezzo
Terry's image you seem to be on to something as well. There's a few guys out there that secretly enhance their images with Hubble Luminance data, and if you call them out on it will go haywire, but it is so obvious. I think that may be what we are seeing here. I've seen another fairly well known imager using a 130mm scope trying to pass off that he is splitting the same close stars as Hubble, which is just not possible. What you have posted of the Cartwheel up close certainly seems indicative of that - there is no way that much detail was captured. It looks the exact same detail as the Hubble image.

I didn't know that… what a pity !!

I also think it's pretty obvious on this image…

When you look at his gallery while having this in mind, some other things seems quite fishy…

For example his previous image (IOTD : https://www.astrobin.com/njk365/?nc=user) which is presented as a "reprocessing" of old 2017 data (IOTD too : https://www.astrobin.com/342470/?image_list_page=2&nc=&nce= ; well that's 2 IOTD for the same data, but whatever… ). Just compare both versions… the first one is realistic of what you can legitimely achieve considering the setup, despite you surely can improve the process… but improve it to the point of its 2d version, seriously ? Look carefully at the 2d version, you can nearly see what areas of the image have been reprocessed using a mask, while other close areas have not changed at all…

Look also at its M83 image. Same setup, 30h of exposure. https://www.astrobin.com/407828/?nc=user
Something strange ? Not at all ! The details are realistic ; but there is less details than in my image of this same galaxy made with the 1m scope (https://www.astrobin.com/354521/F/?image_list_page=2&nc=&nce=).

However, for "Cartwheel", these are the two same setup that were used. How would it be possible to obtain such details in ESO350-40 when it's not possible on a much more easy target such as M83 ?

As a reminder, we are talking about objects of this apparent size (yes, ESO350-40 is the very small one on the right-bottom edge… ) :



JB
Edited 11 Oct, 2019 14:07
Jean-Baptiste_Paris
11 Oct, 2019 15:24
For your information I asked their feelings to few well known astrophotographers (including Damian Peach), and they agree there is something strange concerning the insert image : probably a mix with HST data and in all case impossible resolution for a ground based system.

I send a private message to Salvatore in order to know what is possible to do.

I suggest, if Salvatore and other IOTD staff members are ok with that, to ask Terry in private for some explanations and let him the possibility to prove his good faith, by sharing a raw file whis us for example… We then could discuss about this seriously. If we consider it is a fake, we should block any TP or IOTD push for this image and "warn" him, or at least let him know what we think. If the image is real, then ok all is fine !

I believe it is not possible to allow fraud on Astrobin ; and in that case there are some serious suspicions.

What are your thoughts ?

jb
Edited 11 Oct, 2019 15:26
Die_Launische_Diva
11 Oct, 2019 19:47
I hope that the "details" we see in Terry's Cartwheel Galaxy are only deconvolution artifacts. My poor mind can't process the possibility of existence of individuals who have the luxury of spending 10K$-100K$ and then "enhance" their data with professional data just for winning a badge smile
Jean-Baptiste_Paris
11 Oct, 2019 21:16
Die Launische Diva
I hope that the "details" we see in Terry's Cartwheel Galaxy are only deconvolution artifacts.

That is what I believe at first sight… but in no way a deconvolution can lead to reduce 10 times a star size without visible artifacts ; and even less "create information".

Some of details on this image have a width of ~1 pixel ; that is just not possible, even with drizzle (especially with only 30 subs), even with an incredible seeing, with a 10'' scope.

Even being a processing wizard, you cannot break the laws of optics…

The galaxy is 1.1' x 0.9' ; the sampling is 0,8'' ; the minimum size of a detail must be higher than 1 pixel (even with drizzle), so you cannot have details smaller than 66/0.8 = 1/80 of the galaxy size ; what is nearly the case here… except the scope is not in space !

jb
Edited 11 Oct, 2019 21:27
Andys_Astropix
11 Oct, 2019 21:33
I'm not so sure - Terry is a member of my astronomy club - the ASV and is known for being a stickler for detail, never uses NR and is a perfectionist - It's out of character for him to do anything like this…. I know him personally - he's quiet and understated and not in it for show or awards. Hmmmm…. best tread lightly.
Edited 11 Oct, 2019 21:34
Jean-Baptiste_Paris
11 Oct, 2019 21:57
Andy 01
I'm not so sure - Terry is a member of my astronomy club - the ASV and is known for being a stickler for detail, never uses NR and is a perfectionist - It's out of character for him to do anything like this…. I know him personally - he's quiet and understated and not in it for show or awards. Hmmmm…. best tread lightly.

Gosh…  smile

Always a bit difficult when it comes to a personal aspect but, Andy, what is your own technical opinion about the image itself ?

IMO there is a clear problem on this image… not necessary a deliberate will of fraud, considering what you are saying about him, but a lot of things sound wrong.

Once again, his overall image looks real. That inset view top right is not from the main image. It looks nothing like the galaxy in the main image.
Such a resolution is not possible from a ground based system, even less with a 0,8'' sampling ; it would require a large mirror and adaptative optics to obtain such details.

Take a look at the image of this galaxy made by Capella Team with a 60cm telescop and 0,3'' sampling (and 2,6'' seeing with 300s subs) :



And Capella team members are quite far from being beginners in astrophotography, especially for galaxies…

JB
Edited 11 Oct, 2019 21:59
Andys_Astropix
12 Oct, 2019 02:01
I recommend you message him directly, say it’s an extraordinary result and ask how he did it.
Give him a chance to respond and if you have further concerns raise them with him directly.
Jean-Baptiste_Paris
12 Oct, 2019 07:29
Andy 01
I recommend you message him directly, say it’s an extraordinary result and ask how he did it.Give him a chance to respond and if you have further concerns raise them with him directly.

You are right Andy, I think it's a good way to do ! smile

I take a fresh look at his image this morning (in France it's right now… ) ; and I'm quite convinced after that this is not a "fake" (ie a mixing version with other data) because the details in the galaxy do not match any real detail that can be see on the HST image.

I then think it is only due to very heavy deconvolution and over-processing ; and the "details" visible in the galaxy are only artifacts.

I have to say it is a great relief that imo this is the good option…

JB
Edited 12 Oct, 2019 07:29
 
Register or login to create to post a reply.