«

Azzz1588 writes:

> Intresting that you have several people here offering more info
> on the subject, yet you will not even keep an open mind.

I don't see anyone explaining the technology.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

On Thu, 23 Oct 2003 16:15:45 +0200, Mxsmanic <>
wrote:

>Azzz1588 writes:
>
>> Intresting that you have several people here offering more info
>> on the subject, yet you will not even keep an open mind.
>
>I don't see anyone explaining the technology.


Well, duh. Use your imagination. And a bit of common sense.

The technology is called image processing. Pattern recognition.
It's not that cutting-edge any more. Scores of textbooks, white
papers and PhD dissertations have been written on the subject.

And there is absolutely no reason why the inventors would divulge
any details of this technology, and every reason for them to keep
the technology as secret as possible.

Digging a little further into this topic, I learned a bit more.
The technology doesn't apply to inkjets. It doesn't apply to
monochrome lasers. At present, it is applied only to color
laser copiers.


rafe b.
http://www.terrapinphoto.com

In article <>, Mxsmanic
<> writes:

>I don't concern myself with the standards that others apply to my posts.

The same applies in how others look at your postings........

Why dont you consider that as well ?
Have you never made a mistake in a posting here ??
Have you ever admitted it ???

Or are you right all the time, and have no need for others inputs.....
























"Only a Gentleman can insult me, and a true Gentleman never will..."

In article <>, Mxsmanic
<> writes:

>I don't see anyone explaining the technology.

You dont read the posts here then ??
























"Only a Gentleman can insult me, and a true Gentleman never will..."

Rafe B. writes:

> Well, duh. Use your imagination. And a bit of common sense.

My imagination tells me that these systems cannot be nearly as accurate
as is being claimed here without either enormous expense or very high
false-positive rates.

> And there is absolutely no reason why the inventors would divulge
> any details of this technology, and every reason for them to keep
> the technology as secret as possible.

If it interferes with my legitimate use of the copier or printer, it
makes a big difference.

In particular, if printers are encoding serial numbers into the things
they print, that means they are deliberately degrading image quality.
I'm not at all keen on paying big money for a printer that is designed
_not_ to provide the best possible image quality.

> Digging a little further into this topic, I learned a bit more.
> The technology doesn't apply to inkjets. It doesn't apply to
> monochrome lasers. At present, it is applied only to color
> laser copiers.

I don't see how these copiers can consistently and accurately recognize
any of the world's currencies, but still copy everything else without a
problem.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

On Fri, 24 Oct 2003 06:10:37 +0200, Mxsmanic <>
wrote:

>Rafe B. writes:
>
>> Well, duh. Use your imagination. And a bit of common sense.
>
>My imagination tells me that these systems cannot be nearly as accurate
>as is being claimed here without either enormous expense or very high
>false-positive rates.

Where did you read any accuracy claims? At this level of
the discussion, we still haven't heard an acknowledgement
from you that the technology exists.


>> And there is absolutely no reason why the inventors would divulge
>> any details of this technology, and every reason for them to keep
>> the technology as secret as possible.
>
>If it interferes with my legitimate use of the copier or printer, it
>makes a big difference.

I suspect that Toshiba, Xerox, et. al have considered this
situation.

>In particular, if printers are encoding serial numbers into the things
>they print, that means they are deliberately degrading image quality.
>I'm not at all keen on paying big money for a printer that is designed
>_not_ to provide the best possible image quality.


The encoding is at a level that you would not notice
without a loupe or microscope. Again, the technology is
mostly implemented in digital copiers using color laser
printers -- machines expressly designed for heavy duty
office and corporate environments. You're *not* going to
see it on an Epson or Canon photo (inkjet) printer.


>> Digging a little further into this topic, I learned a bit more.
>> The technology doesn't apply to inkjets. It doesn't apply to
>> monochrome lasers. At present, it is applied only to color
>> laser copiers.
>
>I don't see how these copiers can consistently and accurately recognize
>any of the world's currencies, but still copy everything else without a
>problem.


No accuracy claims have been made or cited in
this discussion so far, IIRC.


rafe b.
http://www.terrapinphoto.com

Word has it that on Fri, 24 Oct 2003 06:10:37 +0200, in this august
forum, Mxsmanic <> said:

>Rafe B. writes:
>
>> Well, duh. Use your imagination. And a bit of common sense.
>
>My imagination tells me that these systems cannot be nearly as accurate
>as is being claimed here without either enormous expense or very high
>false-positive rates.

If you read the articles closely, you'll find that they've been
extremely careful not to say anything about the accuracy of such
systems, & only imply that they actually work. Personally, I think that
95% of it is pure marketing, intended to scare people away from using
those printers & copiers for forgery.
Like you, I find it impossible to believe that these machines can
/accurately/ detect images of currency. OTOH, I do believe the stuff
about watermarking the resulting prints, as that is perfectly feasible
with todays technology.

>In particular, if printers are encoding serial numbers into the things
>they print, that means they are deliberately degrading image quality.

If I were designing such a system, I'd implement it by using the
'signature' to tweak the pseudo-random numbers used to dither the ink
pattern used to reproduce subtle tones. Except for pathological cases,
such a system would be unlikely to degrade the vast majority of images.
I can also think of a couple of other methods that could be used without
/visibly/ degrading the prints, but would be detectable under extreme
magnification.

>I don't see how these copiers can consistently and accurately recognize
>any of the world's currencies, but still copy everything else without a
>problem.

IMO, they don't. I'm of the opinion that the 'currency detection' part
is pure bullshit, although I wouldn't be surprised if they make some
token effort to detect American notes - probably by some really crude
method, like looking for the particular shade of green, for example.
Again though, I don't have any trouble believing that they watermark the
prints.

--
W
. | ,. w , "Some people are alive only because
\|/ \|/ it is illegal to kill them." Perna condita delenda est
---^----^---------------------------------------------------------------