P. J. Alling wrote:
Change the block of pixels in the Jpeg at the upper right hand corner, unrotated, and you'll change the encoding. Do it a couple of times and I'll bet it's unpredictable enough to screw this method all too hell. However, even if it doesn't let's see they used an example to identify cameras, used in kidnapings.
I haven't been able to come up with a good use for this technique beyond things like identifying the camera to use specialized (tuned) algorithms on the image in question. Just identifying a camera model doesn't do much good in nearly any other environment that I can think of. Still, it's interesting to me that they can "see" enough artifacts of the demosaicing algorithm to identify them uniquely. Hopefully I'll get the time soon to read the original paper and see exactly what they're doing.
-- Thanks, DougF (KG4LMZ) -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
