Maybe no stego on eBay afterall

[Peter Wayner]

At 11:14 AM -0700 7/19/02, Nelson Minar wrote:
>  >The basic scheme is very simple. Generally the inserted message has
>>higher entropy than the raw LSBs of an image. So the entropy of the
>>picture/message combo should be higher than the picture alone.
>
>That's only for lousy steganography. If you really want to mask your
>message, you match it to the statistical profile of the source data.
>Of course that's difficult to do perfectly and so is potentially
>detectable, but you can make detection much harder than "look for
>white noise in the LSB".


Yes, you're right. Good steganography isn't that hard to generate-- 
especially if you're not greedy.

>
>>So even though there's some cool science in the process, there's no
>>scientific way to draw the line.
>
>I agree. Maybe the good folks at WetStone just draw the line in
>whatever place is best for their business at the time. Or maybe they
>made an honest mistake. Either way, now their president is on record
>drawing the line in two radically different places.


Yes, I'm glad someone noticed this. It's a great datapoint!

---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at wasabisystems.com