[5037] in cryptography@c2.net mail archive
Re: Eason/Kawaguchi stego
daemon@ATHENA.MIT.EDU (Jay Holovacs)
Wed Jun 30 15:46:30 1999
From: "Jay Holovacs" <holovacs@idt.net>
To: "Russell Nelson" <nelson@crynwr.com>, <cryptography@c2.net>,
"David Honig" <honig@sprynet.com>
Date: Wed, 30 Jun 1999 08:43:45 -0400
----------
> From: David Honig <honig@sprynet.com>
> To: Jay Holovacs <holovacs@idt.net>; Russell Nelson <nelson@crynwr.com>;
cryptography@c2.net
> Subject: Re: Eason/Kawaguchi stego
>
> \begin{nuance}
> Except that encrypted LSBs will be perfectly uniformly distributed
> and normal noise won't. Its possible to reversibly sculpt crypto
> data to have a less conspicuous spectrum.
> \end{nuance}
>
Good stego can choose LSBs from pseudo randomly bit locations. LSBs from
these locations should be indistinguishable from random if appropriate
images are used.
jay
