Re: Complex Theoretical One Way Hash Question

Matthew Harrison wrote:
> I wish to know if it is possible to embed an MD5 of an image (e.g. a
> JPEG) in the image such that the MD5 is human readable in the image AND
> is an MD5 of the modified image including the readable MD5.
>
> That is - if I execute an MD5 hash sum off the entire image, that MD5
> has will be the same as the MD5 shown in human readable format in the
> image (I don't mean a JPEG tag, I mean literally on the image canvas
> itself).
>
> The questions:
> 1. Is it theoretically possible ?
> 2. Is it feasible ?
> 3. Has it already been done ?
> 4. Would other one way hash algorithms be more suited ?
>

It's possible to find two random 260 byte blocks with the same MD5
starting from any initial hash value, so you can fill a file with n
blocks and have 2^n permutations to work with. If you can get a JPEG,
GIF, etc. decoder to interpret each block as a pixel, you're set. The
compute time will be large; 2^33 operations for each block IIRC, but
probably feasible. There may be some tricks you can pull with other
image or page description formats to interpret each block as a bit and
then hex code it and print it. To keep it interesting you just have to
be careful not to choose a sufficiently advanced filetype that can both
print quines and compute MD5s...
Received on Mon May 1 02:01:38 2006