2

I am developing a project " separable reversible data hiding in encrypted image" based on an ieee paper in java. I understood how to encrypt the image but cant really understand how the data is embedded in the image by modifying the LSB. Can someone explain to me how it is done or just point me in right direction.

I was thinking of using RC4 for encrypting the image.

The base paper can be found at http://www.sundaychennai.com/IEEE%202012%20Dotnet%20Basepaper/Separable%20Reversible%20Data%20Hiding%20in%20Encrypted%20Image_moorthy.pdf

Amar C
  • 374
  • 5
  • 17

1 Answers1

0

Without reading the paper, my understanding of how this could work, is this:

We could store information in the picture, by letting the LSB in of the R, G and B components be carriers of the information we want to "hide".

It would be very hard for the untrained eye to see "fluctuations" of these bits, especially in an image displaying a scenery or something like that.

An image of the dimensions 1024x768 pixels, would would then be able to store 2359296 bits, or 294912 bytes.

Of course, lossy compression algorithms like JPEG could not be used, without the information being destroyed.

Werner Kvalem Vesterås
  • 10,226
  • 5
  • 43
  • 50
  • I will be using uncompressed 8 bit gray scale image. So i guess i can modify 2-3 bits of some random pixels to embed the data. This is my basic understanding. But i want to do this as explained in the paper. – Amar C Feb 02 '13 at 11:25