Roswell - New scans of the Ramey Memo : Can it now be enhanced/deciphered?

originally posted by: lovebeck

originally posted by: MoshiachIusDei
a reply to: IsaacKoi

I don't think it's the words you need to focus on, but the 'not words'.

Here's what is being hidden, near the thumb.

93

Huh? Could you elaborate please? Ty.

Also, the "editing" of the image makes it very cartoon like and really just kind of cheapens the subject matter. No need for fancy graphics and all that jazz...IMHO, anyway.

Don't blame me for the editing. That's how it was intended to be viewed by the 'author'.

Firstly, can you see the angry alien? If not, this isn't for you. Stop reading.

I used an image from the Camera section, one that was quite sharp and hadn't been 'tampered' with. I noticed the blacks were much blacker than the the text which, being typewriter ink, should be very black on a fresh document. Typewriter ink is designed to last and not fade.

I analysed the blacks by setting them to transparent in PowerPoint. The algorithms are quite good. Then I realised the image is a composite of three: two copies of one image, one rotated by around thirty degrees, and a third which is the memo, over-exposed lying over the top.

I took two copies of the image and set the two principle blacks to transparent, one on each. Then I rotated the top image until the alien with bared teeth appeared. The white 'streaks' aligned. I coloured the top image red which didn't affect the alien, showing the memo image is not relevant to the content.

Hope that's enough elaboration.

a reply to: jeep3r

Excellent idea you have:

"One way of solving the puzzle could be to recreate the exact settings in that photo in a studio setup. With similar lighting, the same camera at the same distance etc. Then one could create different sample memos written on the same typewriter (one that had most likely been used at the time) with the same font. An OCR algorithm could be trained using these samples and in the end, we could feed it with the original photo. If we're lucky we might get some good results or at least pretty reliable recommendations for what the wording might be."

Extremely good work on presenting this research.

J Bond Johnson had said in one article written in 1998 (entitled [url=http://www.abduct.com/features/f25.php]“that:

“I was given the wire service "flash" announcement of this rapidly developing story by my city editor and I headed for the air base.
…
When the General entered the room I handed him the "flash" announcement printed from the news wires. He read it with interest.
I then took a couple of shots him, still wearing his hat in his office, examining the debris with the "flash" announcement held in his hand”.

If the original "news service flash" telex text is available, then it should be relatively straightforward to determine if this is the document that Mr. Johnson provided to the General.

In looking at the image, the bottom of the page does appear to me to be torn, as if from a telex machine.

Also, having some information regarding the typeface and font may be helpful in deciphering the message. Typewriters and Telex machines of that vintage no doubt used standards.

Anyway, I'm just throwing a few ideas out there...

-dex

a reply to: IsaacKoi

Might I suggest you practice writing in all caps, evenly spacing the characters, much like it appears on the paper? If you get a "feel" for the writing style you might pick up on what some of the words are. Language specialists can recognize patterns based on the lengths of words as well.

originally posted by: athatguy
a reply to: Pinke
these are tools for solving captcha, and some you can even submit your own images too. this would be a great way to decipher what is written on that paper

Hiya athatguy!

This is a great first post, please stick around.

The issue with captcha is that, in general, it is using 'trained' image / character recognition.

Captcha has the benefit of being able to get a dataset of correct (and incorrect) answers with which to train the recognition process. You get five hundred or a thousand captchas from the same engine and you run it through your process. The application 'learns' from its wrong answers by basically going, 'I thought this was a B you're telling me it's a P, from now on I'll claim it's a P.'

You can vary thresholds in that process and, particularly with captchas, you can even have 'guesses' sometimes. If the application comes across a particular scenario that it gets wrong 50% of the time then there is no harm in the application trying both answers.

The difference we have here is that, without the original hardware to create a training data set, there is no way we can test or train our application's answers. It's a bit like the whole Rosetta Stone concept. Give an analyst a few good examples of the code or language intact and we can reproduce the whole thing, but without that we're a bit in the dark.

It's a great idea though and I like where your head is at. If we had some more definition or even just some of the same imagery where we know the answers we could make some significant progression in that direction.

PS - with the original camera hardware and film stock it (or really close to it) might be possible to create a training data set that could be used for testing. That could then be used to verify any results from the original imagery. Could also be used to test theories regarding it being a known document.

originally posted by: MoshiachIusDei
a reply to: IsaacKoi

I don't think it's the words you need to focus on, but the 'not words'.

Here's what is being hidden, near the thumb.

93

I'm not sure if anyone took the time to look at this image but whatever you did to it, the memo in red is easier to read, you can make out letters and words easier. I urge people to try photo enhancement on this image as it may be able to show more.

I personally find this image easier to make out words and letters in than the original ones posted in the OP, just my opinion though.

a reply to: ChaoticOrder

actually there are 3 images of that note taken at slightly different times and angles.
I didn`t check the OP`s scans to see if all 3 are there but here are three slightly different photos (from google):

www.alienscientist.com...
www.nicap.org...
www.roswellufomuseum.com...

a reply to: IsaacKoi

Fair enough then. Only time will tell i guess. Oh, and might i say. Good job on the post.

I played around a little with the Focus Magic filter for Photoshop and wasn't as successful as I wanted. I resorted back to manual adjustments using Fireworks MX and Photoshop and it's still a guessing game of what the words actually are. The first readable line seems to be the big part of the contention. The first line is said to read:

AND THE VICTIMS OF THE WRECK YOU FORWARDED TO THE

The pic below is a few of my adjustments in which I believe the first line reads:

AND THE FINDING OF THE _____ YOU CONNECTED TO THE


Even with the poor quality of the image, I can see the word CONNECTED over FORWARDED. The first letter in the claimed word VICTIMS seems to be closer to an F or P. The second letter looks to be an I with the third and second to last letters being I N.

I also make out-
AT FORT WORTH
ON THE "DISC" MUST
WEATHER BALLOONS WOULD MAKE

All assumptions of course.

a reply to: clt1994

that first line reads "AND THE FINDINGS OF THE MAJOR WAS CONNECTED TO THE"

To my eyes at least.
my best guess is this paper is a press statement showing the balloon story.

still stellar work by the OP and the image pro's on this thread.

A similar kind of 'challenge' was issued over at unmannedspaceflight and it might have some ideas as to what to try.

It was trying to de-blurr some characters on an envelope printed with an inkjet printer. The answer was known but the methods people used might help in this situation perhaps.

originally posted by: Tardacus
actually there are 3 images of that note taken at slightly different times and angles.

In my OP, I embedded 4 different photos which included Ramey holding the piece of paper. Those images were the ones numbered 1, 2, 3 and 4 at:
www.abovetopsecret.com...

Unfortunately, the side of the paper with text printed on it only appears to be visible in the first of those 4 images

originally posted by: qmantoo
A similar kind of 'challenge' was issued over at unmannedspaceflight and it might have some ideas as to what to try.

Mmm. That forum certainly seems to have a lot of expertise in image processing.

I'll sign up to that forum and post in that thread from 2009 to see if any of the members there are interested in playing with these scans.

Just a thought. One possible way to see if a consensus can be reached regarding the words would be to cut them all out individual using the best possible resolution, mix them up randomly, and then have as many people as possible tell what they think the individual words are. This would possibly help eliminate a lot of the tendency of people to try to force words and meanings into the spaces by assuming context. It's basically using a multi-scan technique, but with people's eyes and brains doing the interpretive scanning. Interpretation by consensus, I guess.

For instance, if you took this word...

... and asked as many people as possible what they thought it was, you'd get a range of guesses. But if you had 25 people say that it was "HEATING," then that could be a good indication that is what the word actually is. Then you put all the words back in context and see if the message makes any sense.

originally posted by: Blue Shift
One possible way to see if a consensus can be reached regarding the words would be to cut them all out individual using the best possible resolution, mix them up randomly, and then have as many people as possible tell what they think the individual words are.

I agree that would be an interesting little experiment.

By the way, Kevin Randle was involved in an experiment a few years ago to see how different people interpreted the "memo" (including people outside ufology). He co-authored an article on that experiment at:

Houran, James and Randle, Kevin “A Message in a Bottle : Confounds in Deciphering the Ramey Memo from the Roswell UFO Case”, Journal of Scientific Exploration, Volume 16, Number 1 pages 45-66 (2002)

David Rudiak is, um, not a fan of that experiment. He wrote a rebuttal at:
roswellproof.homestead.com...

originally posted by: Ectoplasm8
I played around a little with the Focus Magic filter for Photoshop and wasn't as successful as I wanted. I resorted back to manual adjustments using Fireworks MX and Photoshop and it's still a guessing game of what the words actually are.

Thanks for posting the results of your adjustments using Fireworks MX and Photoshop.

I've avoided participating in the guessing game that you (rightly, in my view) mention. I'm still hoping that some enhancement(s) will remove some of the element of speculation.

(By the way, I am reading all the posts but not responding to each one individually. I am grateful for the various comments and attempts to enhance the images but it probably would not be productive for me to post a separate response thanking each person for each comment...).

originally posted by: IsaacKoi
By the way, Kevin Randle was involved in an experiment a few years ago to see how different people interpreted the "memo" (including people outside ufology). He co-authored an article on that experiment at:

I think I read that at some point. We can also try to identify as many individual letters as possible, then find people who are good at Jeopardy.

Well, I think there is one thing we can say for sure.

This letter doesn't say...

A bunch of intelligent people mistook a few ounces of foil and sticks.

For a SPACECRAFT apparently with a few dead ALIENS.

Back in those days when you were told to keep your mouth shut...You did.

Roswell happened. Maybe not exactly how they tell it in the books.

No New Mexican Rancher in those days drives 75 miles to show the AF that junk.

originally posted by: whyamIhere
Well, I think there is one thing we can say for sure.
This letter doesn't say...
A bunch of intelligent people mistook a few ounces of foil and sticks.
For a SPACECRAFT apparently with a few dead ALIENS.

Well, the Army pretty much invented the word SNAFU, and there's practically nothing they can't miscommunicate.

A line of words most people can make out is "OF WEATHER BALLOONS WOULD MAKE." So it could easily be saying something along the lines of "What idiot who saw a bunch of weather balloons would make such a mistake?"

a reply to: Blue Shift

Point taken.

However, we went from my Dad loading our TV vacuum tubes to go test.

To solid state circuitry pretty damn quick.