View Single Post
  #13  
Old September 22nd 12, 11:53 PM posted to rec.photo.digital.slr-systems,alt.photography,rec.photo.digital
Alan Browne
external usenet poster
 
Posts: 12,640
Default Forensics v. Photoshop

On 2012.09.22 18:33 , Eric Stevens wrote:
On Sat, 22 Sep 2012 10:56:08 -0400, Alan Browne


That isn't what (or solely what) this software purports to do. Rather
it looks at fine grained differences in the actual image data to
fingerprint the effects of its sensor/d-a/n-r circuits/algorithms. That
will not be common from a given manufacturer except for a few closely
related variations of a model.


It's quite possible that a model X and model Y may share the same
sensor and image generation software. But there will be differences in
the EXIF data which should show up in the file. But that doesn't


Though it analyzes that as well, it appears to be the duller edge of its
sword.

matter. The question is, can anyone alter the image in such a way that
the fact that it has been altered is indetectable?

I'm sure that it will be possible to stuff another image inside the
file but to do so indetectably would require that the new image be
constructed in exactly the same way as the camera would have done it.
I expect that finding out how that should be would be very difficult.


Unless you can emulate the fine grained noise of the camera being
spoofed the s/w at discussion claims to be able to sniff out a fake.
Such emulation is not out of the question but you would have to do as
much work collecting the statistics (as they did) to be able to carry it
out.

Or get the data from the company at discussion perhaps. ;-)

--
"There were, unfortunately, no great principles on which parties
were divided – politics became a mere struggle for office."
-Sir John A. Macdonald