Re: [steghide-devel] Sorry! Ignore the previous e-mail

[Stefan H. <sh...@ch...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sunday 27 April 2003 06:05, Adam Brickman wrote:
> I found your suggested tip to copy the module files to the steghide
> directory after I sent the e-mail asking for help. Sorry about that.
> Do you have any documentation on how your particular algorithm works, in
> more detail? I understand that it modifies the LSB's of the cover file.
> In my report, I would like to briefly explain how your algorithms works,
> including how it choose s which LSB's to modify, what it does with the
> other LSB's, its proposed strengths/weaknesses...anything you might have
> would help.
To understand how it chooses which LSBs to modify you should take a look at=
=20
the "distribution method" as described in the man page (the usage.txt file =
on=20
Windows) and implemented in the dmtd_* functions in stegano.cc. The other=20
LSBs are not modified. A strength is that the default distribution method=20
automatically chooses the maximum possible interval length (also see man=20
page) thus resulting in an equal distribution of the embedded bits in the=20
cover-file. A weakness is - as with every LSB-based technique - that it is=
=20
relativly easy detectable with first-order statistical tests if you embed=20
large amounts of data (see for example: Andreas Westfeld, Andreas Pfitzmann=
:=20
Attacks on Steganographic Systems. Information Hiding, 3rd International=20
Workshop, Proceedings, Dresden September/October 1999, LNCS 1768, Springer=
=20
Verlag, Berlin 2000, 61--76). The upcoming 0.5.1 version will use an=20
algorithm that is not detectable by first-order statistical tests.

Greetings,
Stefan Hetzl
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+q5VFKN/5Ocgn19gRAjNsAJ9k6e92wzyAGpESxOML/Kh+/YBc8wCfWaWm
pjwaN+LG8xEAu5M+w/HBAe0=3D
=3D/KbF
=2D----END PGP SIGNATURE-----