Using R456 composites to distinuish between berries and pebbles

Hi Hort, et al

I don't know if you've been following an interesting discussion on Doug's Unmanned spaceflight forum related to "strange mineralogie feature in sol 616 R345 image.

Essentially, a poster Harkepper, passed on a claim that by manipulating Infra red images obtained from the raw R456 pancam striking differences in colours of the pebbles and berries at erebus can be obtained. He is wondering if these colour differences could be related to real minerological differences between the pebbles.

here

Of course the Scientists in the Unmanned spaceflight have pooh-poohed the idea based mainly on the fact that the raw jpeg stretched images are not RAD corrected.

Here is my processed sol 616 image made from the R7; R4+5/2; and R2 Oppy images

This image essentially corroborates Norbert Gasch's work ( The Scientist who originally did the image processing) but there are many aspects that could probably be discussed here.

Eg. Is it possible that, even though the raw images are not colour corrected and the processing has a somewhat greater Art component as compared to a strictly scientific Image processing component, the colours might still offer an indication of different constituents in the objects that take on a different colour. e.g. The very small berries are of a different colour than the irregular clasts. Could this be an indication that they are of different chemical makeup even though they show up similarly coloured in the Left Eyed colour composites and even the untweaked Right eyed ones?

Also, even though the science might not be top flight, could this system be used to differentiate between the berries and the clasts of similar size?

And, Since, the concretion theory now needs all the help it can get, could this technique offer some indication of what are berries and what are not?

Winston

I have created some infrared composites from the right pancam raw images and they all seem to suffer from a severe vignetting ( as seen in your example above ), so I would not trust any visual comparisons not on a circle centered in the image.

Having said that, yes, there are differences in the infrared between rocks that appear "the same" in the left filters.

The "red" rocks ( brightest in the "far" infrared ( R7 -- 1 micrometer) ) are the most intresting to me because, um, green vegitation is "bright" in this region and soil is not.

See here for a sample spectra.

Hi Hort

Thanks for commenting.

For comparative purposes. Here ar 3 more images;

THis one is a sol 314 Oppy colour composite using my usual L257 images. Note all the berries are blue.

This one is from the R2457 raw jpeg images of the same sol, Sol 314. There is some differentiation of colours of the berries, some are red some are blue.

This last one is from the R2457 RAD corrected img files from sol 314. There is a very similar differentiation of colours as seen in the jpegs above, indicating to me at least, that the effect spans both the jpegs and the img files.

Winston

I seriously think this should be look into much deeper, and i almost, everytime, tweak my images to see these differences, in the materials.

Like horton says however, you must stay in the center, for the real differences.

Look up vignetting.

Hi mann / Horton

I'm an absolute neophyte re. the theory of image processing. I therefore am not clear about how horton's comment about vignetting and a circle centered in the image and mann's "stay in the center for the real differences" square up.

How big is the circle around which vignetting is likely to occur? Would the superb quality of the rover's lenses be expected to reduce vignetting to a manageable radius? If so, what is this radius?

On another related matter, some of us are postulating that some of the berries might be fossils, be alive or be products of life elaborated by living microbes, if so, could this technique, or a variant of it, give some indication of textures or sites that might be examples of "life" in both Meridiani and Gusev?

On Doug's site, after some conjecture that the effects were faked, there now seems to be a consensus gathering that the effects are probably real but are due to the variations in the light angle at which objects are imaged as the filters are changed. Does anyone here think this is an adequate explanation of the effects?

Winston

Hi All

Here are 2 more images for comparison purposes with the earlier images above.

The first is a normal L257 composite for a spirit sol 587 pancam series

The second is the R7542 composite for the same sol 587 series. The image was processed in exactly the same way as the Oppy images above.

There are no outrageous colours seen in this second image. The sun angles should have operated in the same way with the spirit cobbles here as they would have with the oppy cobbles in the images in my earlier posts.

Winston

Hi Bill Harris

I've just read your comments on this matter in Doug's Unmanned Spaceflight forum. care to repeat them or otherwise comment here where I can respond appropriately?

Winston

Interesting conjectures, Winston. I hope that some mechanism can be developed to distinguish between berries and round pebbles.

Back when Oppy was MI'ing many berries, this was a pretty easy distinction to draw. There were surface differences between at least some berries (the larger ones, at least) and rocks.

But now NASA/JPL has stopped looking for berries for the most part. We have seen some suspicious-looking things lying about in pancam shots, but these have not been MI'd to any extent. This is a pity, because I'm still looking for the ONE BIG BERRY which would push the concretion model back towards favour, by establishing the possibility of a lognormal distribution for Martian blueberries.

Hi Henry

It's quite likely that the technique will not be able to distinguish between berries and similar sized cobbles or clasts in the Pancam images.

However, I think that it should not be thrown out merely on considerations based on imaging theory but that it merits a more in-depth study.

Just to add another comparison of a Spirit IR 2457 highly saturated image with an L257 image.

Again, like the other spirit comparison, there are no significant colour changes except around the edges of the highlighted hole.

Winston

Here are the infrared bands, to help distinguish, between Hematite and Goethite.

Maybe there is a difference, between the berries, In outcrops, and the cobbles, always not eveloped, in anything but the soils.

They need to find a larger enough chunk to grind, or just run a few of those odd pieces over, and over, if nessasary.

Is it just a coating, A silica/salt crust?
Gothite?
Hematite?
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=11765784&dopt=Abstract

http://mann.smugmug.com/gallery/100790/62/44136907/Large

Hi All

Bill Harris did not respond here although he responded to me obliquely on Doug's forum yesterday.

Bill Harris is an occasional poster to the MarkCarey Blog. I thought it might be of some interest to a few of us to see what he wrote about this Markcarey Blog when posting on Doug's blog.

A Bill Harris quote from Doug's forum;

"I personally think that the Good Doctor (Norbert Glasch) is straining at a gnat and trying to infer things that ain't. Over at the culinary forum down the hall ( "Kooks Online" ) they think he's on to something and might be using this technique to find fossil chicken lips in the meridiani plain..."

Another Bill Harris quote from Doug's forum to show that he read my comments above;

"Exactly, Scott. It is wondrous that we can download image files taken through bandpass filters and easily combine them into 24-bit color images with readily available software on home computers. But the downside is that even "an absolute neophyte re. the theory of image processing" can do the same thing and with some time and little experience can arrive at wild conclusions. It's a perfect example of the computer principle of GIGO".

It may be of some interest also to note that I started this topic here, having seen the discusssion on Doug's blog, and that at no time did I or any of the other 2 persons who joined the discussion, conclude that the technique could find "fossil lips" or any other bete noir or otherwise jumped to any wild conclusions.

Could this be a testimony to Bill Harris' fairness and ability to comprehend simple statements?

Winston

Get used to it Winston, There will always be someone like Bill Harris, who will want to be part of what He believes is the In crowd.

Then there will be people who just want to figure out, one of the tools that Nasa is using to see different minerialogies in the images.
Organics is one of the differences, including fossil biofilms, varnishes, And And Microbes in salts.
Spectra, can show us whats what.

One of my very first posts was something like, "whats up with the crappy Black and White images" Posters wrote, "get a life" these are the same images that Jpl Is looking at" Uh Hu.

Winston
Don't be disturbed you still have friends.

While I am here I would like to bring up the R2 457 image on your 818 post.
Has anyone commented on the serated edges of the thin bedded laminae. Looks almost like what you see when tilted thin sheets of ice melt. Anyway just an observation.

The true fossils here are individuals like Bill Harris.

There is nothing bad about being a fossil!

Anyway, the images are NOT inconsistent. Getting the fine calibration on CCD data to be as subtle as a human eye is mostly an art. The images show a color DIFFERENCE - while it would be nice to know exact level of this difference (i.e. have good calibration with all of the steps) it is not required to understand what the images are telling us.

The big cobbles of shiny rock is hematite. This is practical proven by the near match to the hematite "berries" as well as the examples of larger hematite cobbles eroding out of the sulfates rich rock. There are many examples. Besides, have you looked at the cobbles? Just regular rocks eh?

Some of the berries in photos are red because to your eye they would have a slight red hue. This is not inconsistent with them being mostly made of hematite. If they were green, then that would be a problem, but not red because if you take any piece of gray hematite and scratch it, it looks red.

I have a bunch of pieces of gray hematite on my desk right now. I just walked outside and took a piece that was polished, rubbed in on the concrete, and now I have red scratches on the concrete as well as a red scratched spot on the piece of hematite. The spot on the concrete is only very slightly pail red, like a red dust exists in the scratches. The scratched spot on the piece of polished hematite is even more subtlety red.

I started a conversation with Henry about measuring the diameter of "berries" a while ago. I wanted to bring him around to understanding that the "berries" are just one form of the hematite, and lacking a framework we need to measure the diameter of all hematite, thus "berries" will stand out. But alas I have been very busy and not posted much. There is much to do with that analysis…

On a another topic, which I will post separately on, do you see the evidence for surface flow?

Fossil
I have not seen evidence for surface flow. Tell me where to look and I will see if I agree with you.
I agree with you about the source but,do you have an explanation of how the berries formed in the layered beds?

Come on Fossils, PLEASE, show me an image that shows a Large cobble, eroding out of the layered beds.
Just a link to Raw, would be fine.

They are on top, on top of the soil mostly, very rarely on the layered evaporates even.

They appear to have formed, either, on the surface, or Between the beds in the soil, or in the soil, but not in the evaporates.

Mann, I will start with a simple example. One that is often put aside because it seems anomalous. Yet, all you need to do is match each item in the photo to its likely source in the rock.

It is usually impossible to find the spot in the rock where a piece of hematite fell from or where the rock around it simply eroded away leaving only a resting point and tell tail strata disruption. But it is clear where a piece of hematite sources from in some cases.

It is good to think that you can see the rock eroded away from the abstraction of a time machine. From the time machine, you look as the rock melts (erodes) away, leaving bits of hematite to wiggle around drop down as the rock around and below disappears. When you stop the machine, and get out to look closely, you say, I see, this piece came from here and that piece came from there.

If you were to draw lines following the strata in some photos, you might get a good picture of what the surface was like at the time of deposition, the dips, the ripples, the bumps, etc when the sediment was deposited. The sediment collects covering everything. Things that are not flat, that is they have vertical relief, usually take a few layers of sediment to be completely covered. Dips and hollows may collect more sediment that other places flattening the bottom out. There are of course, complications.

What do you see here? (Note, the MI photos of the Upper Dells are upside down.)

Not much, you say. I can't tell exactly what that big piece of _____ is. So, take a step back and look.

What do the strata do around the big piece of _____. And, what are those odd little bits of _____ sitting near by? Do strata lines bend around hematite or does hematite intrude across strata?

Mann, you say "They [the large hematite lag cobbles] are on top, on top of the soil mostly, very rarely on the layered evaporates even."

This is just like the "berries" that everyone now accepts as hematite. And since the entire Meridiani surface was at least a meter higher before the billions of years transpired that sand blasted the sulfate rock to dust leaving the hematite lag covering much of everything. Aren't the cobbles lag also? The lag has in large part preserved the bedrock for us to examine.

The processes that have occured within one billion years is very hard to grasp. Three to four is almost unimaginable.

Good point on the eroded out berries not being on top either Fossils.

However Can we find any large cobbles IN the evaporates, Partly exposed, like some of the berries?

This piece is on top.

The berries appear to trancend layers.

Can you fix your last image link?

Mann, sorry, the link got messed up. I think the forum software does not like underscores. The second photo is the Upper Dells mosaic where you can see the large piece of hematite with the surrounding strata lines (right side up). This photos is available many places, but they usually do not have the best MIs of the large piece of hematite included. - Here is the Upper Dells mosaic again.

The anaglyph photo you post is nice - but upside down. When the photo is upside down, our minds need to transpose things (at least mine does) and it is extra hard to understand the photo.

When I look at the mosaic and draw the strata lines I find that the big piece of hematite fits, just about where it sits. Better when it is turned a bit counterclockwise. Are the flat lines through the large piece of hematite random? Do they line up with strata? If so, how are the strata spread around the piece? Strata lines pass through it, like many "berries", yet the strata also bend and narrow over the piece. There was a gradual dip in the bottom where this was piece of hematite ultimately formed and the sediment draped over the chunk in layers. Its possible that soft sediment layers were spread and compressed by a concretion as it grew, but that is in contrast to the nearby "berries" where if they are nucleating concretions they are said to completely replace the sulfate rock as they grow.

So is it a small cobble? Or a large "berry"?

What about the two pieces of lag nearby (below the large piece in the anaglyph)? Just bits of sulfate rock?

You should also know that I think many of the large angle-ly cobbles are meteorites. Actually, a meteorite lag from the dunes. The dunes are meteorite collectors…