So I will give it one last try…..

[Christian Frantz Toussay]

...still hoping someone with the required expertise and skills can hep us detect flaws in what I am doing, here is a summary of the technical points of the debate so far:

*I have presented a process, based on iterative interpolations/extrapolations of identical data sets. This process considers the "original image" as just one of n numbers of potentially valid (with varying values of course)derivations of the "true objective data" thas was recorded in the first place.

It thus consider the image not as an end product to be marginally enhanced optically, but as a raw set of data that has to be analyzed from all sorts of differnt "angles", to be more fully understood.

I used the undisputed Belzner image to establish, as objectively as I could, the process validity, only upgrading the undisputed HSCA findings of "an individual in dark clothing" to the more precise data level of "an individual in dark uniform"

*the first argument raised was the validity of the process of interpolating ("adding") data with data to get meaningful, exploitable feedback in return.

I explained that it is done all the time in our everyday life, and that this approach is part of countless scientific or technical operational procedures. Actually, interpolating data is indeferentiable from human experience.If you were not able to do this, you would not only just "not be there" (like you could be somewhere else), you just would not be.You'd be curled up inside a singularity outside the objective universe.

*the second argument was: Ok, data interpolation does bring valuable information feedback. But you said yourself that these are extrapolations, and so carry also a certain amount of "wrong" information, data generated by the observation process itself (note as an aside that this is true for all observation processes...). Now how can you sort out the good info from the bad?

The process is based on the very simple concept of resilience of information. Though I have never found it expressed as such, I would definitely be very surprised that it had not been expressed long ago in much more sophisticated ways. It simply states that real, objective information will have a greater propension to manifest itself coherently, regardless of how it is examined, than random noise will.

The process is about agregating layers and see where coherence is reinforced (eventually evolving into visual images, since we are analyzing pictures here. But remember, this was not developped for processing pictures to begin with...), and where it eventually dissolves, thus indicating the presence of unsupported artefacts.

The process focuses on shared data carried along by successive iterations.

*the third counter argument is the "randomly generated hypothesis". The argument is now: Ok, you have those apparently credible images. Let's take it from there. How can you verify they could not be obtained randomly, by mere iteration of meaningless derivations of the original set?"

I offered 2 tests, actually I described 2 possible way of running such tests, resting on known quantified variables value that could be possibly used to verify this.

I am out of my depth here.

I have no scientific qualification whatsoever: I am only using my brain to imagine how the parameters involved could be soundly checked. Help welcome here...

But I offered a low-complexity approach to answering, at least in part, this argument:

-the random noise hypothesis would require a "highly creative" noise generating aspect to the process, allowing for the artificial creation of very credible images but only in very definite areas (which are, more over, internally coherent: the image does'nt appear thru the wall, but at a location compatible with its being actually real, in the opened window...), while at the same time preserving overall valuable data virtually uncorrupted. That would seem to be a tough proposition.

This is particulrly striking if you use the white window frame in Hughes 3, partly hidden by the man's face, as a reference point.

Also, the fact that highly correlated images could be extracted from different set of data (Hughes 2 and 3) if there was not objective data of some sort "supporting" this manifestation thru the time sequence, is highly improbable and can I think be computed too.

So basically, that's where we stand right now. I will put several different iterations of Hughes n°2 for comparison to complete this section, and then I will move to some processed data pertaining to the the Picket Fence area.....

In the meantime, it can be interesting to put this new information (the presence of a law enforcement officer at the exact location the official theory claims the shots that killed JFK were fired) with an aspect of the case that has always troubled serious researchers.

It took the Dallas Police less than 2 hours to arrest oswald, and less than 24 hours to charge him with the crime, with what was presented as an uncontrovertible closed case loaded with physical evidence.

The images (granted, still not "confirmed"...) would maybe give an indication that the searchers suspicions about the case against Oswald (documented by numerous detailled specifics of the evidence, pertaing as much to the nature of the evidence as to how it was collected and preserved as evidence) could have a basis in fact.

The fact that a member of the law enforcement agency that charged (using highly contested "evidence") a man of a crime was present on the crime scene during the deed, while the alledged culprit is nowhere to be seen, should be of some consideration here...

The fact that this visibly innocent man died violently while under their assumed protection is also something to be taken into account...

Edited December 5, 2011 by Christian Frantz Toussay

[Thomas Graves]

"Christian,

When you get some free time (LOL), would you please apply your process to the photographic "captures" (stills), taken from the Robert Hughes film, which show Tan Jacket Man apparently handing off something to the guy I call "Blue Coated Cuban-Looking Man" in the parking lot? Chris Davidson has done some work on it and believes he's found the curved end of an umbrella handle in TJM's left hand immediately after the putative "hand off". I'm convinced that the handing off of a small back object from TJM to BCC-LM did take place, but so far I've been unable to convince anyone that I'm right..."

....I have noticed you have put a thread on this subject, but I have not had the time to check it.

I don't know the basics of this specific subplot but I assume it pertains to the umbrella-as -a-weapon theory.

Since you have already noticed the difficulty of image interpretation, even on much larger scale objects like human bodies, I am not sure you could extract data sufficiently convincing to definitely shut the case to the satisfactions of people with different opinions.

But let me see the images involved if you want...

Frantz,

Here's a capture and enlargement of one frame, compliments of Chris Davidson:

http://i140.photobucket.com/albums/r25/123steamn/2-2.png

Please visit my thread "Tan Jacket Man Secretively Hands Something To Blue Coated Cuban-Looking Guy In The Parking Lot" to see more images and links to images. If you can, please look at the slow motion film of this incident. The real issue is whether or not Tan Jacket Man handed off a small black object to Blue Coated Cuban-Looking Man. (Anyone recognize the coat-and-tie-wearing Latino dude at he far right in the above photo?)

Thanks,

--Tommy

Edited December 6, 2011 by Thomas Graves

[Christian Frantz Toussay]

...this is another iteration of the data content of the Sniper's Nest window, from Hughes film frame n°2, this time putting the image in a larger perspective relative to the window...

Note the box arrangement on the lower right quadrant of the window...

Now I think you have an idea about what I mean when I say that "cleaned up" versions of the assassination films can be produced, simply by applying the process frame by frame...

Edited December 5, 2011 by Christian Frantz Toussay

[Christian Frantz Toussay]

...still another one, carrying at the same time a large core of shared data, and additional information (the skin tones of the face and of the hand holding the object, for instance, have started to manifest)....

Edited December 6, 2011 by Christian Frantz Toussay

[Christian Frantz Toussay]

...and here is a composite showing the evolution of the image thru 3 stages...

[Christian Frantz Toussay]

..I post below a last vesion of Hughes 3, to illustrate why I think the optical illusion / randomly generated image hypothesesis is in a tight spot here:

*the image shows what I call "secondary shadows": if you look closely, you will notice that the cap actually casts a shadow over his eyes and part of his nose.

This would be an extraordinary level of sophistication for an optical illusion: an inexistent cap projecting an inexistent but very real-looking (respecting the light conditions of the scene)shadow.

And again, this optical illusion would have to manifest itself so credibly while the rest of the original data remains almost pristine, undisturbed by the process.

It actually can be claimed validly that the white line of the window frame is incomparably higher in definition in the processed version than in the original we started from.

This would mean that the process would at the same time objectively enhance data (the window frame) but also generate totally fake, very higly credible optical illusions (the man in the window)

This "optical illusion" would have to exhibit the very improbable caracteristic of showing up in the exact location where the official version claims that the shots which killed JFK were fired from.

Now that would be some sort of very stubborn optical illusion, I would think......

Edited December 6, 2011 by Christian Frantz Toussay

[David Josephs]

..I post below a last vesion of Hughes 3, to illustrate why I think the optical illusion / randomly generated image hypothesesis is in a tight spot here:

*the image shows what I call "secondary shadows": if you look closely, you will notice that the cap actually casts a shadow over his eyes and part of his nose.

This would be an extraordinary level of sophistication for an optical illusion: an inexistent cap projecting an inexistent but very real-looking (respecting the light conditions of the scene)shadow.

And again, this optical illusion would have to manifest itself so credibly while the rest of the original data remains almost pristine, undisturbed by the process.

It actually can be claimed validly that the white line of the window frame is incomparably higher in definition in the processed version than in the original we started from.

This would mean that the process would at the same time objectively enhance data (the window frame) but also generate totally fake, very higly credible optical illusions (the man in the window)......

still hoping someone with the required expertise and skills can hep us detect flaws in what I am doing, here is a summary of the technical points of the debate so far:

Franz,

There is no expertise or special skills required to see the basic flaw in what you are doing is that the process simply adds too much data to the original set of info, then manipulates the data based on the original "screen" and produces outputs that only SUGGEST what may be there -

just like your wet Tshirt analogy from way back that you dropped quickly once you saw it directly addresses the flaws in the process... If you took a photo before the water, and then again after, you SHOULD notice that the pixels in the two photos are now very different... all the processing in the world could not produce the 2nd image from the first or vice versa...

the "water" is only similiar to the process in that it CHANGES pixels by guessing and re interating - and then comparing to the original...

It will always give the appearance of some relationship to the original since obviously, that's where all the data is coming from, yet you still do not post the FIRST STEP, the vector mapping of the original... this is BEFORE the process... so please.. before posting the color enhanced EXTRA DATA versions, post a LARGE version of one of the enlargements...

The enlargement to vector process also adds info and drops other...

While I greatly appreciate the time and effort - the end result is no mare than computerized manipulation of the original image in such a way as to CHANGE the data of the original and therefore corrupt the end result

DJ

[Christian Frantz Toussay]

"just like your wet Tshirt analogy from way back that you dropped quickly once you saw it directly addresses the flaws in the process... If you took a photo before the water, and then again after, you SHOULD notice that the pixels in the two photos are now very different... all the processing in the world could not produce the 2nd image from the first or vice versa..."

..again, I still think that you are missing the point here: the wet t-shirt analogy was used to explain to you what is meant by the interpolation process. It is an analogy, and it is perfectly valid.

-you stated first that interpolating data is not a valid way to extract meaningful information. I demonstrated to you that it is.

I have not yet heard a valid counter argument from you on that very specific issue.

-You stated then that it is impossible for "hidden" data to be made available.

I used the wet t-shirt analogy to explain that it is perfectly possible, as long as the 2 different levels of information are in contact, ie already interpolating/exchanging information. This interpolation can be made more manifest by interpolating new data (the water, in the exemple taken). That is all.

You seem to be confusing the map with the territory here....

In the process, contrary to what you seem to understand, no water is added to the picture. The interpolation of data plays the role of the water on the t-shirt...

"It will always give the appearance of some relationship to the original since obviously, that's where all the data is coming from, yet you still do not post the FIRST STEP, the vector mapping of the original... this is BEFORE the process... so please.. before posting the color enhanced EXTRA DATA versions, post a LARGE version of one of the enlargements..."

...you already requested such. I have posted the original from Hughes 3, aside with the final processed result obtained. I have also posted on purpose several different iterations, for comparison purposes.

If you need more, let me know.

Also, I have advised you that the full Bond 4 file is 34,7 mo and that I am ready to send it to you for analysis as soon as we can find a way to make sure that no issue will later arise about what is in the originally sent file.

Clearly expressed, this means that I want to deposit this file with independent 3rd parties which will be able to verify the file content being made available to you.

Nothing personnal, you will understand, I think, I am taking only some reasonable precautions here.

This can be very easily arranged, so that should not pose any sort of problem for you to get acess to the files and do whaterver crosschecking you deeem necessary.

I also note, as an aside, that you don't seem to address the fact of the enhancement gain obtained on the undisputed image of Black Dog Man, nor the more technical issues I pointed out, like the fact these "fake" images would appear while still preserving a high amount of the original data, like the window frame in Hughes 3 and 2.

I'd be really interested in your explanation for this quite peculiar phenomenon.......

"... the end result is no mare than computerized manipulation of the original image in such a way as to CHANGE the data of the original and therefore corrupt the end result"

...this, any viewer can verify for himself, I think:

*in fact, the processed version clearly exhibits enhanced data pertaining to the window frame (it is more visible than in the original...): so the claim that the process "CHANGE the data of the original and therefore corrupt the end result", (Dave's own words...) is actually, and demonstrably, unsupported...

Edited December 6, 2011 by Christian Frantz Toussay

[Christian Frantz Toussay]

....I am posting below the first of a series of results obtained thru the processing of Zapruder frame n° 472. As usual, I will post different iterations of this same data set, with varying definition, but always carrying correlated information, for the viewer analysis.

I chose to present first an unknown image from Zapruder, instead of my work on the more known Moorman, , because I have always been intrigued by Zapruder last statement to the Warren Commission (after his questionning has offcially ended):

"They said it could all be done by one man...Do you know there was indication there were two?"

I personnally find this totally unsollicited statement extraordinary:

what could the man who had actually filmed the assassination meant by this? Why would he feel compelled to put this declaration into the official record, at a time when when the formal interrogation had ended?

I checked the Zapruder film, and noticed that several frames in the end actually showed the picket Fence corner area, a location that had long been suspected of being a firing post.

Zapruder had actually filmed the Picket Fence Corner, the exact area where HSCA experts had determined 33 years ago a shot was fired at JFK. They concluded that this specific shot had missed, still allowing for the Oswald-killed-Kennedy theory. Buyt they were definitive in their assertion that sombody had been there shooting at JFK.

If he had been there, and since zapruder filmed that very exact area, I decided I would give it a try with the process.

I captured several frames from this end sequence: I am not sure whether the work was done on frames from the "Images from an Assassination" DVD or from another source.

The frames are mostly shaky, but some of them are definitely....interesting.

I will present here results obtained essentially on frame 472....

These are, of course, up-to-now unknown images of the assassination...

Edited December 6, 2011 by Christian Frantz Toussay

[Christian Frantz Toussay]

...this is a different iteration of the processed data content of the Picket Fence Corner area, extracted from Z 472....

...searchers with at least a basic knowwledge of the evidence will of course notice that this is the exact area where a witness has been stating for almost 5 decades now that she saw a "policeman" fires at least one shot in the direction of the presidential motorcade...

Edited December 6, 2011 by Christian Frantz Toussay

[Duke Lane]

OK, I've been following this FAIRLY closely, at least enough to know that it's confusing a lot of people, or at least not clarifying things to the point that most people "get it," much less are able to reproduce it on their own. I'm don't think that I do or can either.

An important consideration to me is how this technique can be applied to other photographs/imagery, particularly those that don't necessarily conceal (or render indistinct) objects that are not obvious to the naked eye. I'd think that anyone who cannot discern a use for it outside of revealing conspirators hidden in bushes and shadows would take a dim view of its validity. It would seem to have to have other applications in order to be tested more fully.

Another thought is that this could be reviewed by someone who works with technical image processing regularly, such as a product reviewer at a technical/computer or photography magazine - or even a Photoshop guru! - as a "peer review," if you will, of the process. Hearing it from someone who knows the technical terms, etc., for this might also help to explain it in clearer terms for the rest of us.

What does anyone else think? I'm loathe to "throw out the baby with the bathwater" simply because I can't understand or appreciate what's being done, any more than I'm willing to accept something that appears to clarify something in an image that has otherwise been nothing more than a Rorsach blot that I've got to cross my eyes to make "sense" of.

I have a gazillion photos of the fronts and backs of homes, for example. Can this technique be used to discern or enhance the visibility objects that are behind the windows? Something like that can validate the process if, for example, it shows objects that are verifiably inside the house (or not). Thoughts? Other uses?

Up to this point, I have no opinion about this, but definitely find it intriguing. I just don't think - as David Josephs has said - that "finding what we want to see where we think it is" (my words, his idea) is necessarily the best test of the process. Other ideas?

[Christian Frantz Toussay]

"OK, I've been following this FAIRLY closely, at least enough to know that it's confusing a lot of people, or at least not clarifying things to the point that most people "get it," much less are able to reproduce it on their own. I'm don't think that I do or can either.

An important consideration to me is how this technique can be applied to other photographs/imagery, particularly those that don't necessarily conceal (or render indistinct) objects that are not obvious to the naked eye. I'd think that anyone who cannot discern a use for it outside of revealing conspirators hidden in bushes and shadows would take a dim view of its validity. It would seem to have to have other applications in order to be tested more fully.

Another thought is that this could be reviewed by someone who works with technical image processing regularly, such as a product reviewer at a technical/computer or photography magazine - or even a Photoshop guru! - as a "peer review," if you will, of the process. Hearing it from someone who knows the technical terms, etc., for this might also help to explain it in clearer terms for the rest of us.

What does anyone else think? I'm loathe to "throw out the baby with the bathwater" simply because I can't understand or appreciate what's being done, any more than I'm willing to accept something that appears to clarify something in an image that has otherwise been nothing more than a Rorsach blot that I've got to cross my eyes to make "sense" of."

...Duke, I can't really say how much I appreciate what you're saying here. Be sure I understand what it can be be to have to take into ccount what I am showing here. I will say it again: what I have found (and which is still to be verified, we agree on that...) contradicts my original reconstruction of the assassination, 50 carefully polished years in the making.

I could have argued aeons, convincingly, about my own pet theory.

If you want to know, I had a bunch of misguided cuban exiles shooting from the Sniper's Nest, while unbeknownst to them, some aditional help was added by CIA operatives, just in case...

So an apparent coordinated crossfire by people wearing Dallas Police uniforms was not exactly what I expected.

I am not much more advanced than you could be on this. Sure, I can operate the process, and I hope that I have been explicit enough in explaining it to others, but I am none the wiser.

I need help on this, to verify it. It still could be some very, very weird, freak occurence of some kind. It is just that I have checked it and rechecked it again over the years.

I think that maybe we could, with your help, test optical enhancement methods on the processed image. Those, I don't master, but I understand you do. It can maybe be verified if the fine details of some of the images obtained can be checked with the classic optical tools.

Optical tools do work: they are effective to bring out richer objective content. It is my assumption that that should at least give us an indication of the reality of such fine details as can be seen in Bond 4.

Optical illusions should, I would assume, dissolve under close scrutinity.

What do you think?

Edited December 6, 2011 by Christian Frantz Toussay

[John Dolva]

OK, I've been following this FAIRLY closely, at least enough to know that it's confusing a lot of people, or at least not clarifying things to the point that most people "get it," much less are able to reproduce it on their own. I'm don't think that I do or can either.

An important consideration to me is how this technique can be applied to other photographs/imagery, particularly those that don't necessarily conceal (or render indistinct) objects that are not obvious to the naked eye. I'd think that anyone who cannot discern a use for it outside of revealing conspirators hidden in bushes and shadows would take a dim view of its validity. It would seem to have to have other applications in order to be tested more fully.

Another thought is that this could be reviewed by someone who works with technical image processing regularly, such as a product reviewer at a technical/computer or photography magazine - or even a Photoshop guru! - as a "peer review," if you will, of the process. Hearing it from someone who knows the technical terms, etc., for this might also help to explain it in clearer terms for the rest of us.

What does anyone else think? I'm loathe to "throw out the baby with the bathwater" simply because I can't understand or appreciate what's being done, any more than I'm willing to accept something that appears to clarify something in an image that has otherwise been nothing more than a Rorsach blot that I've got to cross my eyes to make "sense" of.

I have a gazillion photos of the fronts and backs of homes, for example. Can this technique be used to discern or enhance the visibility objects that are behind the windows? Something like that can validate the process if, for example, it shows objects that are verifiably inside the house (or not). Thoughts? Other uses?

Up to this point, I have no opinion about this, but definitely find it intriguing. I just don't think - as David Josephs has said - that "finding what we want to see where we think it is" (my words, his idea) is necessarily the best test of the process. Other ideas?

Lots of ideas there to pursue for an answer. (Duke)

Franz, I've been giving this a rest since you posted a copy of the original you worked from and I can see straight away that that original itself is 'compromised'. One could just as well take a part of the sky and zoom in and do color splits of all kinds and generate images that show often what is rightly called rosarch tests afa seeing is in the eye of the beholder. Still, I think that it's good that you pursue this afa you can because in the process much can be learned, tho often not exactly that what one may expect.. I encourage you to continue.

[Christian Frantz Toussay]

"Lots of ideas there to pursue for an answer. (Duke)

Franz, I've been giving this a rest since you posted a copy of the original you worked from and I can see straight away that that original itself is 'compromised'. One could just as well take a part of the sky and zoom in and do color splits of all kinds and generate images that show often what is rightly called rosarch tests afa seeing is in the eye of the beholder. Still, I think that it's good that you pursue this afa you can because in the process much can be learned, tho often not exactly that what one may expect.. I encourage you to continue."

Thanks for the encouragement, John: , I appreciate that, like Duke, you want to keep an objective attitude about this, although you are not fully convinced by what I am presenting here, and maybe disagree with the potential conclusion where this is headed.

Actually, preaching to the choir would be of very limited interst here, as I see it: I'd rather have solid counter arguments instead...

When you say that the "original is compromised", I understand it as meaning "the original you worked on is not actually a truely objective representation of the recorded data" (correct me if I am wrong).

Now, as I explained, this is not actually a problem for the process, because it poses as a starting principle that "the original is just one of potentially true "readings" of the data at hand".

All readings are potentially "true", and the process is specifically about sorting them out: so it is the process which will enables us to rank the potential value of each iteration, by monitoring the shared data between successive iterations.

There is (almost) no seniority privilege here (with the exception of reinjecting regularly the original into the workflow, but this is considered an indispensable security check): no "birth-right pass".

Only correlation will do.

I think the same criticism could be leveled at the Belzner version I worked with: if compared to the "original", it would probably shows that it is not a clone-like duplicate of the original. Nevertheless, I think it still carries at least 99% of the original data recorded, so it should not interfer drastically with the way the process operates, and I have shown here, convincingly I think,that the process does bring added enhancement to that specific image.

Another way to describe the process at work would be to say that it works like an evaluation tool for probabilities. This means, by definition, that we admit from the start that an incompressible aspect of "fuzziness" will be present in the results obtained.

This is not, actually, much different from market research, when you allways factor in an "error margin", because since we are dealing with human material (testimonies, opinions, etc) we know that the data collected just cannot be "perfect": people will lie, try to hide what they really think, will want to answer questions in way most favourable to them,etc.

Actually, it could be argued that there is no such thing as "perfect data".....

I think I have already adressed the randomly generated argument: I will say again that the fact that the process would produce highly credible optical illusions while maintaining the preservation of the actual data (even obviously enhancing it, like the window frame in Hughes 3) is a solid counter argument..

Edited December 7, 2011 by Christian Frantz Toussay

[Christian Frantz Toussay]

...going thru the record looking for some other piece of data that might help us test the process in a satisfying way, like I think we started doing with the Belzner image, I think maybe a test could be done on the Oswald / Lovelady controversy from the Altgens picture.

Now, as can be seen in the definition of the material presented here, although critically enchanced, it does not claim in no manner whatsoever to be able to pick up enough facial details to make a positive identification between 2 persons who, and that's the problem here, look very much alike.

The process brings out new finer details, though, and that may help us get some clue from the clothing, not the face...

On the other hand, the Altgens picture was taken by a professional, and the work I have done on the Daltex Building definitely shows that this makes a big difference, as far as the quality of the data content. The Altgens picture is rich....

I have no theory about this specific issue: since Oswald was not in the Sniper's nest, logic dictates he had to be somewhere else.

I would think that if Oswald was a crucial element of the plot, and destined to become the scapegoat, the plotters would probably make sure that he was kept out of view to make the scenario credible.

A different theory would be that since the Dallas Police was "in", they did not doubt of their ability to control the inquest and the evidence.

So maybe this could be tested, with the mentionned limitations, on this material...