Novel Image Steganography Algorithm Using Two Hidden Thresholds

This paper describes a novel image steganography algorithm, with the central idea being the use of two hidden thresholds. By implementing the described method, various types of information (sub-images, numbers, symbols) can be hidden inside a usual image. The article presents how to apply this procedure for a gray image: after recovering the two threshold values, hidden inside the least significant bit of each of the first 16 pixels on the edges, an image can be processed by comparing the gray level to the threshold levels (making sure the gray level is between the thresholds) and thus reveal the secret message. The algorithm can be extended to different types of data and image representations and could also suggest another method of embedding metadata inside an image, such as the time of download, the original website from where it was downloaded or information on its author, etc.