For real lulz: print this pattern of small yellow dots on all your business correspondence with big companies, just to screw with their document management systems. Hours of endless fun for all the bureaucrats!
While looking for that I stumbled across patterns and text that only shows up after photocopying due to effects such as aliasing and low-pass filtering, also pretty neat!
https://en.wikipedia.org/wiki/Void_pantograph
Ok, I worded this wrong. I meant it as in "Wow, that Perl script is from 2008, I'm impressed and surprised because the article gave the impression that this is a relatively new trick." Can we stop the downvoting torrent now? Good grief!
I've done some preliminary work that measures the amount of "attention" an image processing algorithm pays to a given pixel (in terms of amount of computation done on it). Hard to tell at the moment how well it's working, but the results for the £10 note do seem to focus more on a region that includes a Eurion:
It's based on dynamic taint analysis. We assign each byte the input file a unique label and then track the propagation of those labels throughout a computation. Additionally, we track the "compute number"; basically, if you have an operation in involving tainted data like
A = B + C
Where at least one of B or C is tainted, the compute number is:
TCN(A) = max(TCN(B),TCN(C)) + 1
So if you track the propagation of tainted data and the associated compute number, you can make a map between the byte label and compute number. Then the only remaining step is to map back from bytes in the file to pixels in the image; I used an input file in BMP format so that the correspondence between pixels <-> bytes would be fairly direct.
All this is implemented using PANDA [1], which has nice facilities for dynamic taint analysis. I just recorded Paint Shop Pro opening the currency image, and then replayed with the taint analysis to generate the raw data used for that image.
The limitation currently is that the granularity is not great (you can see that there are relatively large rectangular sections of the image with the same compute number). My guess is that this is because a lot of image processing algorithms at the lowest level end up being "multiply the image data matrix by this other matrix and pick out the nonzero entries" or some such -- so all the data there has the same amount of computation done on it. But apparently there is some sort of multi-pass algorithm at work here, which gives rise to the differing amounts of computation done on each region. I'm looking at ways to improve the granularity right now, possibly by incorporating the magnitude of the transformation on the data somehow.
Yup. These weren't as exciting as I'd hoped. The former (eurion_attention.mp4) basically shows a sliding window of what pixels had computation done on them over time; the most recent 1000 bytes (~330 pixels) get "lit up" in each frame of the video. You can see in the early part of the video how it scans over the image multiple times, but the later parts are not very interesting (and the whole thing is absurdly long; I recommend watching at 50X).
The second (eurion_tcn.mp4) one is simpler, and just tracks the mapping between pixel and compute number [1] over time.
This patent describes how it is recognized: https://www.google.com/patents/US20060279767 The short version of which is "find all the circles, measure the distance between them; if you find five with the right distances you've found EURion".
So wait it doesn't account for dense circle patterns? Like if I have a bunch of circles, then it rejects the image if any five of them have bad distances? I would think it would account for blank space somehow.
Ah, good old Markus Kuhn. Always up for poking security features using any way possible. If you're interested in this kind of thing, you should read https://www.lightbluetouchpaper.org/ which is UoC's security research blog.
While I agree with the law, I take exception to my own possessions being hijacked for law enforcement. Apparently there's no need for a warrant to sneak in robotic cops in people's homes and offices.
"In the late 1980s, a new counterfeit hundred, the most perfect counterfeit yet made, began appearing in circulation. It looked identical to the real thing, betrayed visually, at least, under only the most rigorous forensic testing. (Some minor flaws were visible after enormous copies of the bills were made, but these were probably purposeful. Its makers didn't want to be suckered by their own handiwork.) Although the counterfeit came to America mostly on boats from gangs in China, it was eventually traced to North Korea, where it was believed to have been manufactured by the North Korean government on its own presses. Since then, new generations of the same counterfeit have appeared, including a big-head version, mimicking the redesigned hundred that entered circulation in 1996. This family of bogus notes has been given its own title, one that befits its almost mythical stature: the North Korean supernote."
"Some stories about the supernote sound more like legend than fact — like its being laundered by a bank in Macao called the Banco Delta Asia, or several thousand of them somehow appearing overnight in Lima, threatening to tip over the entire Peruvian economy. But there remains one truth in the supernote's history that has never been forgotten: It was first detected at the Central Bank of the Philippines by a teller, given pause only by the same nebulous flaw that betrays the majority of counterfeits. It just didn't feel right."
Fascinating. I wasn't aware of this, so I had to look it up. Apparently even the Deutsche Mark (which preceded the Euro in Germany) was made from "cotton paper" (i.e. 100% cotton fibres).
American banknotes are apparently made from 75% cotton and 25% linen. I'm not sure how the different banknotes internationally and across history stack up in terms of durability but the newer Euro notes (the 5, 10 and 20 notes introduced in 2013, 2014 and 2015 respectively) are supposedly more durable than the older ones.
When I go swimming I put a $20 note in my board shorts wrapped around my id with an elastic band. I don't really take anything else with my so leaving my wallet buried in the sand on its own isn't a great idea.
Cotton money gets all soggy and you can't really use it without drying it out. Plastic money is great for this... wipe it off and buy an ice cream.
Fun fact: don't do this with tickets either. Most tickets are printed on thermal paper, either for economical reasons (cheaper and more reliable than ink/pigments) or to prevent fraud. You may up with a black piece of paper.
Northern Ireland was actually the first part of the UK to have these, they were introduced back in 2000 to celebrate the new millennium, you can still find them in circulation.
I'm playing devil's advocate but a lot of the reason why places refuse to take Scottish notes is that in the early 2000s, the English notes had been updated with tons of security measures, while the Scottish notes had not, so forgers targeted Scottish notes (since they could produce very believable replicas) and the market got flooded with them. So for every
Thus why the bridge series was introduced in 2007, to add much needed security features and to make Scottish notes respectable again.
I can't find this pattern on swiss bank notes. Any idea if the swiss bank notes can be copied on xerox machines? Or do the swiss banknotes use a different pattern?
You can't falsify banknotes with regular printers and regular paper, but you may need to scan and print them for some art project in which case you should just use Linux and avoid those silly protection measures.
Even the "art project" use may be illegal in some jurisdictions. For instance UK law (Forgery and Counterfeiting Act 1981 s18) states:
"It is an offence for any person, unless the relevant authority has previously consented in writing, to reproduce on any substance whatsoever, and whether or not on the correct scale, any British currency note or any part of a British currency note."
(This is separate from the actual counterfeiting offence, which requires intent that you or another person will pass off the copy as genuine.)
This is one reason why pounds on TV shows and in movies is often bundled. They cannot legally print both sides on one sheet of paper, so they print one side on one and one side on another, and then put them at the ends in wads. If you ever see both sides of a note in TV or film then it is the real deal (at least as far as UK pounds go).
A lot of "stage money" has a washed out fake on the other side just in case it gets caught on film (so at least the colors are right). Same colors, the words are nonsense, and none of the detail remains.
Ahem: you can't falsify banknotes with regular printers and commodity paper well enough to fool an alert human being, but the scanners in some vending machines are another matter. With 7 colour 1200+ dpi inkjets readily available for photo printing, if you could find a vaguely passable source of paper you could do wonders -- and one option forgers have used is to take low value notes and bleach them before overprinting with a higher denomination print pattern. (I think this is the rationale for high-denomination Euro banknotes -- €50 and above -- having embedded RFID chips.)
> used is to take low value notes and bleach them before overprinting with a higher denomination print pattern. (I
> think this is the rationale for high-denomination Euro banknotes -- €50 and above -- having embedded RFID chips.
Citation needed
You can't take a lower denomination bank-note and print a higher denomination one on an Euro note: the sizes are different (of course, maybe you can pass one note for the next in sequence, but it is slightly harder)
Night guard at work came to the front desk one night, pockets jangling with quarters. "Did you know the change machine will accept a photocopied bill?" Desk called the police, who called the FBI who came and took him away. Never saw him again.
and it says
"This case was the result of an investigation by the U.S. Secret Service, the Minneapolis Police Department, and the Federal Bureau of Investigation."
> you should just use Linux and avoid those silly protection measures
I think the protection is often done in the hardware (firmware), not drivers. At least it has to be for any printer/scanner that can work as a standalone copier.
(There's got to be a PDF/postscript file/print filter for the pattern somewhere. Anyone? EDIT: Ah, found it here: http://wildsparx.com/eurionize/ -- download has vanished, but source is on github: https://github.com/jplona/code/blob/master/eurionize.pl )