A Multi-threaded Neural Network approach for Steganography

Steganographic techniques are being applied across a broad set of different modern digital technologies. Steganography is basically the process of hiding one medium of communication (Text, Sound, and Image) within another. It can work on JPEG 2000 compressed images & stir Mark images. The steganographic method will be used for internet/network security, watermarking and so on. Steganalysis is the field of detecting the covert messages. The new methods of steganalysis are based on neural network to get the statistics and features of images to identify the underlying hidden data. We first extract the features of image embedded information, and then input them into neural network to get the output. Experiment result indicates this method is valid in Steganalysis . Almost all steganalysis consist of hand-crafted tests or human visual inspection to detect whether a file contains a message hidden by a specific steganography algorithm. The neural network in images is used to overcome the hurdles by hiding the data indirectly into graphical image using neural network algorithm to get cipher bits. The generated cipher bits are then placed in the least significant bit position of the carrier image. A Multi threaded back propagation algorithm is used in the neural network. Multi threading in the back propagation algorithm increases the speed of processing in the neural layers and thereby significantly increases the efficiency. The XOR propagation network model is used which acts as a multilayer perceptron