Implementation of steganography using CES technique

Over the last few decades, security of data exchanged over the network has become a major concern. Two major techniques have existed to achieve the same, namely cryptography and steganography. Cryptography alters the structure of the text itself whereas steganography hides the text behind some other digitally representative media, thus transmitting it unsuspectingly. But with the increase in networks usage and advancements in technology, it has become increasingly complex to protect the information. So the need arises for an almost unbreakable system. In this paper a technique named CES is proposed for exchanging secret data/information between two ends. This technique preprocesses the data before masking it behind a cover image. Preprocessing involves compressing the data to reduce its size and then altering it using a key. Compression reduces the size of text and allows more data to be hidden behind the same cover image. Encoding text with the help of a key makes discovering the original text difficult even if some intruder succeeds in obtaining the modified data. Finally, the processed data is masked behind an image using frequency domain based steganography. The proposed technique is implemented using MATLAB and its strength is compared by calculating the MSE and PSNR of the cover image to the stego image. The various results obtained for different cover images demonstrate that the stego image obtained by applying this technique is not visually distorted because the PSNR is high and MSE is low. Thus, experimental results demonstrate that the technique performs well as compared to other image steganography techniques.