Prev: now, you need to encrypt a test .pdf file, what better than the original RSA paper?
Next: cracking an old zip password
From: Mok-Kong Shen on 2 Jun 2010 04:41
jmorton123 wrote:

> No. You cannot choose to embed into only a specific color byte
> although I could program it to do so. Remember that if you did this a
> cracker would be able to eliminate two thirds of the bytes in the
> bitmap image and just concentrate their efforts on the remaining one
> third.

What I (and I think quite many users) would like is this kind of
generality: One can specify exactly in which pixels (and which colour
of a particular pixel, i.e. different colour for different pixels) of
the LSB of a given image file is to be set to the stego bit one wants
to put in. Could you say with a few words whether this is the case
with your software? If not, I suppose it shouldn't be too difficult
for you to achieve that, since you have alredy done quite a lot
of the work.

Thanks.

M. K. Shen
From: Dave -Turner on 2 Jun 2010 05:01
Yes - the lack of mathematical proof that demonstrates that it IS
bulletproof ... :)


From: mike clark on 2 Jun 2010 09:30
On Jun 1, 11:54 pm, jmorton123 <jmorton...(a)rock.com> wrote:
> On May 30, 10:52 pm, Mok-Kong Shen <mok-kong.s...(a)t-online.de> wrote:
>

> I also suggested a protocol in my original post that was not quite
> right.  Here is what I wanted to say:  place your source files into a
> folder.  Zip this folder.  Use the XOR Utility Program availabe at
> KingKonglomerate.com to XOR this folder using random binary numbers
> generated with BulletProof Random Binary Number Generator, also
> available at KingKonglomerate.com.  These random binary number files
> will have file names such as FA000001.  Next, place the FA000001 file
> into another folder and zip it.  The reason for this is that you need
> to preserve the file name in order to know what file was used to XOR
> it.  Now you use the Stega_BP_v3.exe software to randomly embed this
> into a bitmap image.  

Embed what? The zipped random file or the zipped source files? I'm
guessing both.

> When you extract the file from the bitmap image,
> you already know it is a zip file.  Just add the .zip extension and
> unzip it.  In the unzipped folder you will find FA000001.  You then
> use FA000001 to XOR to get the folder with your original source files
> in it.
>

So basically you are encrypting our plaintext with a random file. Then
storing that random file and the ciphertext in (basically) the clear.
If the attacker knows your algorithm he can also extract the zipped
key file, add the .zip extension, unzip it and use the resulting
random file to decrypt the ciphertext.

From: jmorton123 on 2 Jun 2010 22:04
Currently, the stega utility program embeds into the LSB of a ranomly
chosen byte in the bitmap image. This random byte is chosen based
solely on the random digit string in the Digits.txt file.

The following could easily be implemented: select a particular byte
in the bitmap image and do anything you want with any particular bit
in that byte.

Each pixel consists of three bytes: red, green, blue. All the bytes
in a bitmap image are in this order: blue, green, red.

So it would be straight forward to do what you describe.

The BulletProof source code is available on a non-exclusive basis but
requires, along with a non-disclosure agreement, a rather large sum of
money.

This is also the case for the Stega program.

JM

On Jun 2, 1:41 am, Mok-Kong Shen <mok-kong.s...(a)t-online.de> wrote:
> jmorton123 wrote:
> > No.  You cannot choose to embed into only a specific color byte
> > although I could program it to do so.  Remember that if you did this a
> > cracker would be able to eliminate two thirds of the bytes in the
> > bitmap image and just concentrate their efforts on the remaining one
> > third.
>
> What I (and I think quite many users) would like is this kind of
> generality: One can specify exactly in which pixels (and which colour
> of a particular pixel, i.e. different colour for different pixels) of
> the LSB of a given image file is to be set to the stego bit one wants
> to put in. Could you say with a few words whether this is the case
> with your software? If not, I suppose it shouldn't be too difficult
> for you to achieve that, since you have alredy done quite a lot
> of the work.
>
> Thanks.
>
> M. K. Shen

From: jmorton123 on 2 Jun 2010 22:27
On Jun 2, 2:01 am, "Dave -Turner" <ad...(a)127.0.0.1> wrote:
> Yes - the lack of mathematical proof that demonstrates that it IS
> bulletproof ... :)

Ah, hah! The old if it can't be proven mathematically it can't be so,
routine.

In the Help files, along with a description of the BulletProof
software, there is an argument I make to this exact point. It is
thoroughly addresssed in, The Grand Theory.

Tell us, what in that explanation do you disagree with?

JM




First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5 6
Prev: now, you need to encrypt a test .pdf file, what better than the original RSA paper?
Next: cracking an old zip password