File talk:Total Solar Eclipse 8-21-17.jpg

Doubt about the image

Hi. I explain here a doubt that has emerged on this facebook thread on the closed group "Amateur Astronomy - Telescope, Binocular and Imaging Forum".

it's curious nobody had noticed it so far.

This is an image from Wiki Science Competititon 2017.

It went through many passages after its upload.

1. It was a national winner in the USA competition
2. It was declared a quality picture and a featured image here on commons.
3. It was inserted on many articles on different language editions receiving many visits per days. As far as I know, nobody noticed any issue so far.
4. When the image became international semifinalist, due to the high concentration of astronomical images in such selection I asked a further comment from someone in the field. Such person (both a researcher and a amateur astronomer) agreed with me that many of the astronomical finalists were not particularly original but he found the image overall nice and suggested me to check the shape of the solar flares to be sure they were not copied from a previous different eclipse. I did so and in most of the images I saw the moon was dark so I didn't notice anything again. The previous community discussions analyzed other issues such as the red flares, everything seemed again in order.

The image was than selected few weeks ago as an international winner and before a full press release was sent, it was being shown on line fur further feedbacks and comments. That's where one probable mistake was finally noticed.

The moon is rotated of circa 180 degree, this NASA image shows as it should it be.

According to people I have asked, it is something that probably only someone with expertise of HDR photography who also took images of that eclipse, could have spotted easily. Even on the group on facebook the first reactions were positive, it took a guy with such profile to actually notice it.

I am now linking this from some projects and article talks. I have sent a mail to the author, I hope he can fix it because it is used on many pages. I am informing jurors and (inter)national organizers, too, but the goal here should be to fix it because it is used a lot on articles.

Comments welcome.

Tomorrow I can also ping the users who commented during the procedures here, if you think it is a appropriate.--Alexmar983 (talk) 20:58, 10 November 2018 (UTC)

@Alexmar983: Thanks for being thorough. I'm not quite sure I follow, though. If the only issue is that it's rotated 180 degrees, why can't we just rotate it another 180 degrees? That seems like an easy fix that we can do without the creator. I would ping participants of the FPC for good measure, but I cannot imagine it would affect anything. Is there another outstanding issue? — Rhododendrites ^talk |  23:45, 10 November 2018 (UTC)

User:Rhododendrites I was sleeping but I woke up so I write you immediately. Just want to be sure I give all the details so people can discuss them. I hope I don't make some mistakes because I am sleepy. So... if the whole image was rotated it wouldn't be a problem. If only the moon is rotated this is a mistake that certainly is not worth a prize, because it is unfair for other uploaders that you can fix it after the deadline. BTW, we are not discussing the prize, but the content we are showing on so many articles.

So the main issue is reported in this comment by Alexis Kwasinski, on the FB group, which I copy and paste: If this is processed as an HDR image as implied in the description then it makes no sense that the moon ended up in the wrong position. Besides, the moon looks unnatural. Just compare it with the many images you get when googling "earthshine eclipse 2017." The more I look at the image the more I believe that this may be a composition of two images taken by someone else (otherwise, as the author of the images you would usually figure out the right orientation of things even when messing up somehow the processing....although I still don't see it possible to have the orientation issue when processing it in HDR). that's why the author should tell us something.

Other people are less critical, for example Neil Butcher saysCompare it in the image I posted above, where I've shown it next to one of my images. I can see how he's processed the image and it's no doubt a simple mistake (or intentional for some strange reason), albeit a mistake that shouldn't happen to anyone giving proper attention to the image. It is an HDR image, but that doesn't mean the whole process is automated though, and I'd be very surprised if it were. It looks like he processed the corona images from a handful of different exposures with some kind of HDR merging software (Photoshop's built-in HDR merging, HDR Efex Pro, etc) and then put the moon in manually, as this is almost necessary for this type of shot. The artifacts all over are a dead giveaway that he was using some kind of automated solution for merging the corona images and enhancing the corona, but the moon is also clearly from the same time and nothing looks strange about it (ignoring the processing); you can even see the places where the solar prominences are bleeding light onto the disc of the moon, but they are 180° from where they should be (see the above comparison to see where these bright spots should be - lined up with the solar prominences you can see in his image in the corona). Nothing looks strange about the images, but it surely could have been processed better. Some people just have lower standards than others or don't put in the same time and effort as others, as is demonstrated here (no offense to Michael at all, but it is what it is). I'm sure that once he realizes it, if he hasn't already, he'll be face palming so hard that it hurts

I think it was good faith, but we need a clear reply from Adler. Without the mistake of rotation you can believe in good faith this is a smart processing. With such mistake you have doubt that this is a composition of another image, so the author should tell us something more.

In any case, the quality and featured image evaluation should be redone, that's also why we are discussing here. The presence of such procedures was also considered a positive factor in rearranging the final ranking for WSC.--Alexmar983 (talk) 00:40, 11 November 2018 (UTC)

@Alexmar983: I've emailed the creator and linked to this page. Hopefully he will be able to clear this up. — Rhododendrites ^talk |  00:52, 11 November 2018 (UTC)

This is the explanation I have received The moon is in the correct orientation relative to the sun as it is not a separate picture. It is part of the original picture that was a combination of 7 pictures taken from 1/8000 sec to ½ sec within 1 second. The software takes the dim part of the image from the longer exposures and the brightest parts from the shortest exposure. It is called high dynamic range(HDR) imaging. The overall orientation may be different than other eclipse images you have seen.... I still don't understand how the orientation can be so different.

This comment was sent as a reply to a first question, I have sent him also the NASA link this late morning. He also knows about this discussion. At this point since nobody commented from the astronomy projects of English wikipedia I will ping the users who took part to the the featured image discussion.--Alexmar983 (talk) 20:19, 11 November 2018 (UTC)

@Rhododendrites, Ikan Kekek, and Charlesjsharp: @Martin Falbisoner, Uoaei1, and W.carter: @Johann Jaritz, Schnobby, and Cayambe: @Daniel Case, Agnes Monkelbaan, and Basotxerri: @Bijay chaurasia, King of Hearts, and Basile Morin: @Poco a poco, Llez, and Colin: @Pugilist, Draceane, and Benh: @Yann, Ximonic, and Meiræ: you are all invited to share your comment about this picture.--Alexmar983 (talk) 20:22, 11 November 2018 (UTC)

• IF he didn't use a tripod there can be a very simple explanation for this. The photo is in upright orientation and a 180 deg mishap is easy to do. You shoot a few photos with the camera in one position (left hand on top), then your arms get tired or for some other reason you switch to the other position (right hand on top). Most photographers use both positions when holding their camera and don't really think about which one they use. Later at home when the HDR is merged from several photos, it might not be that easy to remember how you held the camera for the different shots. It's a very honest mistake to make with a digital camera since you don't have the reliable old text on the film to tell you the camera position. Or maybe his software turned one or two of the photos as they were merged to better align them with the rest. Software can do very unpredictable things sometimes. Looking forward to hear his version. --Cart (talk) 20:56, 11 November 2018 (UTC)

Me too. I am quite sure he is good faith, that's why I have the feeling I can learn something interesting from how this could have happened.--Alexmar983 (talk) 21:12, 11 November 2018 (UTC)

Fixing things

I am very very embarrassed about this. The moon is 180 degrees off as has been observed. All 7 exposures were done at the same time on a telescope that was tracking the sun so there was no orientation issue with the original imaging. I then did automated HDR processing on Photomatix Pro and at first did not realize that I had captured detail on the moon surface. When I did realize this I did some manual HDR processing in Photoshop to augment the moon brightness by taking the ½ second exposure that was taken at the same time as the other six images and adding it to the automated HDR image. Unfortunately as some(Neil Butcher) mentioned in the discussion I inverted this image so it shows the moon image upside down. I have redone the image with the moon in the correct orientation. I think the image is close to identical to the original except for the moon now being correctly oriented. Here is a link to this corrected image.

http://astrogab.ning.com/photo/solar-eclipse-2017-jackson-hole-wy?context=user

Michael Adler 11-11/18 — Preceding unsigned comment added by Msadler13 (talk • contribs) 22:37, 11 November 2018 (UTC)

ok, thank you for the explanation and for being available on a week end. Please Msadler13 in the section "file history" there is a link with the text "Upload a new version of this file". Could you use it and upload such corrected image of the file? This way, if people agree, we can keep the flag of featured image and quality image, and we can keep it used in all the articles.--Alexmar983 (talk) 22:51, 11 November 2018 (UTC)

Perhaps, once it is uploaded, we should specify an amount of time (another 2 days, maybe) for FPC participants to voice objection before calling this issue resolved. I cannot imagine anyone will object, though. — Rhododendrites ^talk |  01:52, 12 November 2018 (UTC)

"if people agree" implies IMHO a certain amount of time to check if they do. However, unless someone who is pinged here clearly opposes, there is no reason to remove or question the flag of quality and featured image so far if a new version is uploaded... someone should state that (s)he does not believe the explanation. So far, we all look quite tolerant. --Alexmar983 (talk) 12:50, 12 November 2018 (UTC)

In any case the modified image is a version of this one with no originality, so copyright is probably fine and we can also download and upload, but it is just nicer if he does directly.--Alexmar983 (talk) 12:51, 12 November 2018 (UTC)

He uploaded a different file... anyone can fix this easily? I guess we another step of contacting a sysop. I though my guideline was quite clear, I sent him a link to the re-upload interface.--Alexmar983 (talk) 16:28, 12 November 2018 (UTC)

I have moved File:Total_Solar_Eclipse_2017.jpg on top of the old File:Total Solar Eclipse 8-21-17.jpg. Platonides (talk) 18:23, 12 November 2018 (UTC)

Thanks! I have restored the assessments and categories to the file. If anyone has a problem with this, feel free to nominate the now correct photo for delisting, that would be the proper way to deal with this. --Cart (talk) 19:34, 12 November 2018 (UTC)

So it's called "delisting"... I am not an expert of these procedures. On one side all the reasons the file is considered good still hold IMHO... on the other side someone might still think it's a composition. Personally, based on the quick reaction, I don't think it is the case, but people were pinged so it's up to them. At least the file shown on wikipedias is now correct. Thank you all.--Alexmar983 (talk) 20:26, 12 November 2018 (UTC)

Yep. 'Delisting' is a process where someone who thinks an FP is not up to standards can call on a general vote to remove the star. The process is described here: Commons:Featured picture candidates#Featured picture delisting candidates. --Cart (talk) 20:32, 12 November 2018 (UTC)

Question about the colors

Hi, does anyone know what we see on the right rim of the moon. There are details which are pink/red and look like solar flares. Any info on what they are and about their color?Vasilis Theofylaktopoulos (talk) 10:06, 21 August 2019 (UTC)