London, November 29 : A researcher at the New Jersey Institute of Technology in Newark, US, has come up with a way to spot messages hidden in noisy and complex digital images.
Yun Shi says that does so by analysing the special features that noise may have in an image, and looking for elements that do not match them.
He points out that though regions of texture within an image are often noisy, they have specific characteristics that differ from any image or text hidden in the data describing that region.
Shi says that he has written a software program that uses this idea to look for hidden messages in images, reports New Scientist magazine.
The researcher says that his approach significantly improves on current abilities to spot information hidden through steganography, the art and science of concealing messages in such a manner that nobody but the sender and receiver knows they are there.