Image Specialists - please help in image analysis ideas

Originally posted by qmantoo
This topic is possibly getting hot and slightly controversial and being in the science forum may not attract those willing to discuss or debate that side of things.

Tell me more, please.

reply to post by qmantoo

If you zoom in to the rocks on the right, the black areas have a kind of overlay around the bottom edges and overlapping the black a fair bit. While there, you can see this 'overlay' extending to the rest of the ground. This is not a regular compression instance as it would either compress the black area or it would not compress any of the black area.

When I magnify the area in question, I can see noise but no indication of masking. But then, I'm far from being an expert.

Originally posted by qmantoo
If you zoom in to the rocks on the right, the black areas have a kind of overlay around the bottom edges and overlapping the black a fair bit. While there, you can see this 'overlay' extending to the rest of the ground. This is not a regular compression instance as it would either compress the black area or it would not compress any of the black area.

Maybe you can explain why you think it's an overlay?
You said forget about NASA, now we're back on NASA images, which is fine.
With Mars images specifically, one type of "overlay" which can occur might result from dust on the lens. If it's on the lens it will not be in focus, but it can still distort the image a little in the same area in each photo where the dust particle is. In this early image from curiosity, the whole image is affected by dust.

But in order to see a repetitive pattern like an overlay from dust, you'd need to find similar artifacts in similar locations on each photo, somewhat like the gif image I posted on the previous page, though it wouldn't be so dramatic on the mars rover because it's digital, not film, so the dust won't show up the same way because the mechanisms are different, but same general concept of an "overlay" seen across multiple photos might be seen.

I don't know how you can determine an overlay is being used from just one photo, nor can I see what you mean by an overlay in this one.

Originally posted by Mary Rose

Originally posted by qmantoo
This topic is possibly getting hot and slightly controversial and being in the science forum may not attract those willing to discuss or debate that side of things.

Tell me more, please.

Apparently when a thread is not getting any replies, that means the topic is "getting hot"?
I thought hot topics were the opposite and would get lots of replies?
I am just replying now because I just noticed you finally posted an example for us to look at. Thanks for the example, though I'm not sure it helps, until you can explain better why you think there's a mask or overlay.

I'm not an expert. Mostly just a useless person.

But is it the speckles in this area that have you curious?



As others have said, if you're making a comparison between images please just show them both? The lack of attention is likely due to there really being nothing to see or do here I suppose. This thread is waaaaaay at the bottom of my thread lists like a Vanilla Ice single.

@Pinke - I am not comparing images.

@Arbitrageur -

Maybe you can explain why you think it's an overlay?

I have said on page 1 why I think it is some kind of overlay - because as far as I understand it, areas of single colour do not get compressed with jpg algorithm and there is a section of what looks like compressed image ( at the bottom-right of the black area explained below). I think this is some kind of 'overlay' on the image which is separate to the compression process.

OK, well lets consider the NASA image I posted a link earlier. Considering the 3 large black 'shadow' areas of the rocks in the right of that image (going in a line towards the middle right edge). If we look at the first leftmost of these for a moment.

There is an area at the bottom right of this shadow which looks like compression artifacts. If you look closely however, you can easily see that this extends over the grey ground beneath and to the right of the black shadow. The ground around this rock is totally featureless at that zoom level.

This is not anything to do with compression as there are numerous other grey areas bordering this black shadow which are not similarly marked. Equally, there are other areas apart from this where this 'compression-like overlay' is evident.

-------
The issue of the white dots is another topic but seems to be a function of the images taken with one particular filter as I have noticed it before. It is inconceivable that there could be that number of dead pixels which make the white patches. All I can think is that there may be some kind of ionising radiation causing it perhaps?

reply to post by qmantoo


Is the term for what you're talking about "image manipulation"?

I've been googling looking for information and one of the results was this:

I have always enjoyed editing images, especially so that the observer is unable to tell if the finished result was one single image or a result of several images stitched together. That’s where most of my image manipulation experience lies, creating believable images to the degree that the observer doesn’t even know it’s a doctored image.

Originally posted by qmantoo
as far as I understand it, areas of single colour do not get compressed with jpg algorithm

Actually, they do. You can verify this by saving various images as jpeg and seeing the artefacts that are created.

OK, well lets consider the NASA image I posted a link earlier. Considering the 3 large black 'shadow' areas of the rocks in the right of that image (going in a line towards the middle right edge). If we look at the first leftmost of these for a moment.

There is an area at the bottom right of this shadow which looks like compression artifacts. If you look closely however, you can easily see that this extends over the grey ground beneath and to the right of the black shadow. The ground around this rock is totally featureless at that zoom level.

These look like garden variety compression artefacts. Compression works on a cluster of samples. Areas of high frequency content and high contrast ratios are particularly troublesome for most compression algorithms.

This is not anything to do with compression as there are numerous other grey areas bordering this black shadow which are not similarly marked. Equally, there are other areas apart from this where this 'compression-like overlay' is evident.

The differences in these areas, whilst appearing somewhat similar to the eye, can be "distinct" enough from the algorithm's point of view to result in different artefacts. None of what you have pointed out hints at an overlay of any kind but looks very much like typical compression artefacts. This is unsurprising considering the image has been compressed with a lossy algorithm.

Thanks for all these comments and your time too. As always I learn a lot from everyone.

I'll go back under my rock for a while now. :-)
until the next time.
Q