Mars curiosity Sol2 Anomalies

reply to post by ainsley


Following the Link in your OP, I found that a fellow named Andrew Bodrov had created that panorama. The source images are linked, but when anyone uses software to create panoramas, such anomalies are the result of the software/individual creating the pan. (Just a few more pixels closely to have a direct overlay, Andrew!)

The anomalies are not really there. Remember that when viewing pans. Sometimes a panorama will have artifacts, sometimes not. Pends on the software and skill level of the user.

reply to post by Blue Shift


I do not accept that rational as valid, simply by the fact that if that was the case the "creative liberty" should be made clear and evident, even by generating multiple shots to show what had been done. By providing "altered" images without further explanation it falls into the concept of being deceitful, even if that was unintentionally it erodes the validity of what is presented.

What is the point of the image if parts of it do not reproduce factual data and that is not made clear from the onset, as it is, it is nothing more than an artistic rendering based on the data gathered to a point that it becomes detrimental to any scientific scrutiny and debatable if it serves any real purpose.

reply to post by Panic2k11

Well then. You should give Andrew Bodrov a piece of you mind because he's the one who made the panorama.
www.npr.org...

reply to post by Phage


What purpose would that serve. I recognize that the creator has the right to do the work, and it would be hard to attribute him any blame in the misrepresentation of the work. There is no law that forbids or sanctions the creator itself, at best one could act against the publishers or editors that make use of the image without providing sufficient indication that it is not a representation of reality, only a neat trick that at best shows the artistic capability of the author and the power of the software used in its generation.

Editor's Note Aug. 14, 2012
Earlier today, we published a panorama that purported to be stitched together from images taken by the NASA Mars rover Curiosity. Since that time, we have learned that the author of the panorama has said he used Adobe Photoshop to add a sun to the sky. According to Talking Points Memo, Andrew Bodrov used images from a 2005 Mars rover to approximate the size and appearance of the sun. Below is the interactive as it originally appeared.

This seems insufficient in face of what is being discussed on this thread as the additions (alterations) are not limited to the sun and sky.

reply to post by Panic2k11

There is no law that forbids or sanctions the creator itself, at best one could act against the publishers or editors that make use of the image without providing sufficient indication that it is not a representation of reality, only a neat trick that at best shows the artistic capability of the author and the power of the software used in its generation.

Then what is your point?

reply to post by Phage

Then what is your point?

I think it was obvious, that the publication was at fault and has yet to properly correct the issue, even if some effort appears to have been made. That does not obligate me to take any action, but permits me to call attention to the decrease in reputation of such publication (and those responsible) that such facts expose.

What I said seems a reasonable analysis of this case and other similar cases. Beyond that I had no point to make.

reply to post by Panic2k11


I think you're over reacting. Mosaic images like that are often bound to have overlap/stitching issues (such as the ones discussed in this thread), unless they are put together very carefully and using professional software.

In this case, all of the original images used to make the mosaic panorama are available to everyone online. The easy thing to do here is to use those originals if you want to search the Martian landscape in detail, while using the panorama only as a guide to provide context.

The originals would probably be a higher resolution anyway, because they are the originals. The panorama will never show you more information than the set of originals.

reply to post by Soylent Green Is People


I classify information that is miss classified by lack of due diligence or processing as erroneous or error inducing. If that type misinformation is distributed it becomes disinformation as soon as there is intentional dereliction or an established consensus that it is in fact misinformation. It then falls into two categories unuseful or outright degrading to competing but fully valid information. I see this as a truism.

In any case I was commenting and replying to posts, not taking an overcritical position, having said that any publication, editor or publisher has the obligation to protect their reputation and impart an ethical standard to their work. This is a core aspect of our society and the preservation of knowledge.

I know that the WEB is very prone due to the lack of rigid standards to permit the degradation of useful information and limit access to it (by the share volume of data), but if falls down to the public not to be passive an let it contaminate our culture. My only intention after Phage remark was to center my intervention in this call of attention, doing my bit to make this issue and implication known...

Originally posted by Panic2k11
What is the point of the image if parts of it do not reproduce factual data and that is not made clear from the onset, as it is, it is nothing more than an artistic rendering based on the data gathered to a point that it becomes detrimental to any scientific scrutiny and debatable if it serves any real purpose.

As far as I understand it, the point of the image is to give people a rough idea of that area looks like, as many (most?) people are more interested in that than in what that area is composed of.

I don't think that anyone wanting to do scientific (or just a little in depth) work will use a panorama as their source (I know I do not), because they know that a panorama is just a composite of the original images, distorted and overlapped to make them fit in a spheric (or cubic) projection.

If you want to do serious work with those photos, you get the originals.

PS: It's like using Google Mars; it's easier to find things but the images are worthless for doing any kind of work.

reply to post by ArMaP

the point of the image is to give people a rough idea of that area looks like

All fine but by the presentation and first editorial correction, showed clearly that limited use and lack of accuracy was not obvious and fallowed, hence we this thread. My comments regarding this example and clarifications regarding similar issues are on that point nothing more.

Wouldn't you agree that for instance if Google Mars indicated clearly the limitations on accuracy (I do not know if it does but by the ATS posts I see I presume it doesn't) and technical methodology for the generated modeling even access to the raw images. life would be generally better and less confusing ? This is the reasoning behind much of the care that is given to publishing and peer-review to weed out the unnecessary noise and disturbance on the acquisition of information.

PS: I understand the obvious limitations but the common folk generally does not or even takes the time to consider that they should demand more quality and transparency, especially on the WEB the noise level is so great that it is deafening us all. Even ATS is a perfect example on how the problem spreads.

reply to post by Panic2k11


Using your argument, every single photograph ever taken should have a disclaimer. "The colours presented are not the actual colours as seen by the eye, but an approximation limited by current technology".

Anyway... there will never be an accurate Mars panorama because Curisosity doesn't have a proper panoramic camera system. As the camera rotates, each image is shot from a slightly diferent perspective. To create a proper panorama the camera needs to rotate around the entrance pupil of the lens (often mistakenly called the nodal point) in order to avoid perspective errors. And even this fails for nearby objects.

reply to post by BagBing


I'm tiring of this discussion since we are not getting anywhere consider the point that you are defending, and what I have said, you are not making any sense.

A non textual photo/image alone does not make any claim it is the context were it is presented in that can contextualize it.

Originally posted by Panic2k11
reply to post by BagBing

 

I'm tiring of this discussion since we are not getting anywhere consider the point that you are defending, and what I have said, you are not making any sense.

A non textual photo/image alone does not make any claim it is the context were it is presented in that can contextualize it.

I'm not sure what you are saying, but the context in which it is being presented (gigapan, or a similar website of stiched-together panoramic images) is basically only good for "pretty pictures", and it shouldn't be assumed that images presented on such site are anything more than that. I think the limitations of those gigapan images are obvious, and a disclaimer is not necessary.

For example, I don't think an urban planner would be using a gigapan image of a city to do detailed urban planning or building design. There would be no reason for a planner to trust the gigpan image is to a consistent scale over the entire image, or if the image has missing information at the overlaps. However, I don't think gogipan images of cities need to have disclaimers warning people that the scale of the images may not be consistent.

A gigapan of a city is simply a fun and cool way to view a city. This Mars panorama is simply a fun and cool way to view Mars.

Also in the case of this Mars panorama, the original images are publicly available and easy for anyone on the internet to access if they wanted to use the images for a proper investigation. Those stitched-together panoramas are certainly impressive and useful for the "big picture" context they provide (literally and figuratively), but I see no reason why gigapan would need to provide a disclaimer.

Thank you all for your replies,

The reason for this post was, to find out what methods were employed to create the duplicate effect on the images. Whether it was due to stitching alone or a photoshop tool (clone).

A number of points have been raised due to the discourse so far, one is;

The images portrayed in my 1st post (1st thread also!) show clearly the type of image artefacts left behind from "stitching" tools.

From further reading I found that various algorithms leave different artefacts, if anyone would like to include more photos please do so.

On the whole this has been quite interesting and it has pointed out specifics so that others do not jump to conclusions. Is there a key or chart on ATS for your typical "camera/digital space imagery" artefacts? It would be most helpful I'm sure.

Kind Regards,

ains,

Originally posted by Panic2k11
reply to post by BagBing

 

I'm tiring of this discussion since we are not getting anywhere consider the point that you are defending, and what I have said, you are not making any sense.

A non textual photo/image alone does not make any claim it is the context were it is presented in that can contextualize it.

I'm not sure what you're asking... but I'll give an answer anyway....

First of all, when photographing Mars, you need a white point. Which is impossible. No one has ever been there...

Secondly, the cameras on board Curisosity are not panorama cameras - some people think that Curisosity is a tourist. Only designed to take photos. That is so wrong it's untrue!

OK - I'd like tyo talk about white point. There is no such thing as white in the real world. It's whatever the photographer wants it to be. Consider this... You hold a peice of white paper half in sunlight/half in shade. The sunlight half would be yellow, and the shadowed half would be blueish. Yet - to our brains, both 'look' white. So which one is really white? The answer is, either. Whatever you (the photographer) prefers! Martian colours are totaly subjective. And will always be...

I can really see the duplication here. At first I thought it was just as the camera took several shots and they stitched them together to make one big shot, but one of the pics you can see at the bottom almost a line (in a kind of zig zag shape) where the duplicate rocks have not been placed/married over the previous pictures rocks and if you follow the line up betwee the images you come to a set of 3 rock formations. The one on the left is one shape, the one on the right is a different shape and the one in the middle is a mix of bits of both shapes. So the theory of the stitching of the images does not work here due to the fact that there is now an extra 'fusion' rock of the two different images where the ones below are just duplicates.
Nicely spotted.

Originally posted by ainsley
reply to post by KEMIK

 

It can be quite hard to see, my mind works differently to most, I see tessellation and patterns in the most mundane.

Here's another image that should highlight exactly what I mean.

Thanks for the reply

Ains,

Holy S*** your right.

The "bone" like rock is even just half of it on the other part.

You can't even argue it. I'm trying to. I want to, but I can't. We can't.

Nice find.

Originally posted by CthulhuMythos
So the theory of the stitching of the images does not work here due to the fact that there is now an extra 'fusion' rock of the two different images where the ones below are just duplicates.

What you describe is a common side-effect of making a panorama with some automated tools. If you want to see if there was really some cloning besides the one used to make the panorama just look at the original images, as they are freely available.

reply to post by ArMaP


I sometimes get the same problem when trying to take a panoramic pic with my Sony Xperia phone, that plus if anything is moving in the sweep it can duplicate. Kind of funny when it is a person and their head appears to be floating on its own, lol.

reply to post by ArMaP


Ah thank you for that! Will do.