[Solixious]

Hello friends. I would like to share a software that I've just finished making moments ago. I've decided to call it Solixia, a name similar to my own username. It can be used to hide and extract some secret text within an image. Of course the text you store has to be put in there by this software itself. After the modification is done, save the image in which the secret message is stored. This image would be saved in png format. Other formats of image didn't work with the algorithm I used since it requires preservation of the image as best quality possible.

Requirements : Java Runtime Environment 1.2 or higher

Screenshots:



Virus scan: Link

Download link: Solixia v1.0.rar
Download link: Solixia v1.0.rar

Note : Password to start the application is "hackcommunity" (without quotes of course) and you'll hopefully figure out how to use it since it is not very complicated.

========Updated Version======

On suggestion from Deque I've made few changes in Solixia v1.0 and releasing it as Solixia v1.1. I've removed the password that was required to start (since it got me irritated while testing it a lot)
And few more changes have been made which includes a change in the algorithm used in hiding the text in the image.
Thus the text hidden by previous version cannot be read by this one, but this technique is much better than the previous one and will arouse least suspicion when one normally looks at the image with the secret text. Their is not much change in looks and feel of the software and process of using it remains exactly the same. So the screen shot is same as that of the last version

Virus Scan : Link

Download Link :
Solixia v1.1.rar
Solixia v1.1.rar

Cheers!

[Deque]

Hi Solixious.

Thanks for this program. I've had a close look to it via JAD. It hides the characters in regular intervals within pixels. The problem is, that it changes the pixel significantly, so it is visible by the naked eye.

Here is an image demonstrating what I am talking about. You see the dots in the left side:



If I put even more info in these dots become lines:



You can avoid that by changing only the least significant bit of a pixel, which won't be visible for a human at all.
That is still detectable in the histogram though. Avoiding that is harder, but I don't know if people really do steganalysis on a regular basis in oder to find such stuff.

Additionally you can avoid regularity/patterns in the data locations. Use a random generator with the same seed every time or apply some kind of encryption to define the locations the data is stored at.

Regards
Deque

[Solixious]

Thank you for the suggestion Deque. If we change the least significant bit, the storage capacity would reduce to about 1/3rd of the current capacity. Could there be some way in which the storage would not drop so much? Would it be too much if we changed all the pixels instead of doing it at regular intervals?

[Deque]

Thank you for the suggestion Deque. If we change the least significant bit, the storage capacity would reduce to about 1/3rd of the current capacity. Could there be some way in which the storage would not drop so much? Would it be too much if we changed all the pixels instead of doing it at regular intervals?

That with the capacity is true, but you have a small input field for the message/password to save in the picture. It must be a really really big password to get over the capacity.

You can use all pixels without it being visible to the eye. I did this in my steganography program too and it is only visible if you have an almost plain picture to mask the image and if you use more than one bit per pixel color. (I actually made that program to see how it looks, other than that it is not the best program for steganography)

Using the least significant bit you can store 3 bits of information per pixel (one bit per color) and that should be enough for most messages. I.e. a 200x200 pixel image will provide you with 120 000 bits, which is 15 KB, which can save 15 000 characters. You don't have to save it using the chars, save it using the bits insteads.

To make sure that your program is working for large input, you can compute the max capacity by counting the pixels of the picture beforehand and only let the user input that much characters that are possible.

If you need any help with programming, you can ask me. You can also use my stego source and modify it, if it helps.

[Solixious]

I'd try doing it myself first. If I get stuck, I might need your help again

[Solixious]

I've updated few things. Please see the first post for the updated links. Kindly leave reviews about it or report about bugs in the software if you find any.
Thank you

[Deque]

Works like a charm. Well done.