Academic Journals Database
Disseminating quality controlled scientific knowledge

Data Hiding in JPEG Images

Author(s): S. K. Muttoo | Sushil Kumar

Journal: BVICAM's International Journal of Information Technology
ISSN 0973-5658

Volume: 1;
Issue: 1;
Date: 2008;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Steganography has been an important subject since peoplestarted communicating in writing. Steganography meanshiding a secret message (the embedded message)within a larger one (source cover) in such a way that anobserver cannot detect the presence of contents of thehidden message. Today the growth in the informationtechnology, especially in computer networks such asInternet, Mobile communication, and Digital Multimediaapplications such as Digital camera, handset video etc. hasopened new opportunities in scientific and commercialapplications. But this progress has also led to many seriousproblems such as hacking, duplications and malevolentusage of digital information. Steganography finds its role inattempt to address these growing concerns. We know that,with the use of steganographic techniques, it is possible tohide information within digital audio, images and video fileswhich is perceptually and statistically undetectable. Themethod of embedding secret message (which can be plaintext, cipher text, or even images) is usually based onreplacing bits of useless or unused data in the source cover(can be audio files, sound, text, Disk space, hidden partition,network packets, digital images, software, or circuitry).There are two common methods of embedding: Spatialembedding in which messages are inserted into the LSBs ofimage pixels, and Transform embedding in which a messageis embedded by modifying frequency coefficients of thecover image (result is called the stego-image).Transformembedding methods are found to be in general more robustthan the Spatial embedding methods which are susceptibleto image-processing type of attacks. However with respectto steganography robustness is not a critical property but theperceptibility (i.e., whether the source cover is distorted byembedding information to a visually unacceptable level).There is another important issue of steganography, namely,capacity, i.e., how much information can be embeddedrelative to its perceptibility [5, 1].We shall use digital images as the cover object in this paper inwhich we embed the hidden information. The challenge ofusing steganography in cover images is to hide as much data aspossible with the least noticeable difference in the stego-image.Steganographic algorithms operate on basically three types ofimages: Raw images (i.e., bmp format), Palette based images(i.e., GIF images) and JPEG images. JPEG images areroutinely used in Steganographic algorithms due to the mostpopular lossy image compression method. Usually it is foundthat an algorithm used to hide large amounts of informationtypically result in lower perceptibility (i.e., greater change tothe image appearance) and a more robust algorithm result intolower embedding capacity. The JPEG image generation firstdecomposed the input image into a number of 8 x 8 blocks.Then DCT of each block are computed and the resultant DCTcoefficient matrix is quantized using a standard quantizationtable. Finally the inverse DCT of quantized coefficient matrixare evaluated and the final JPEG image is obtained afterrounding the values.
Affiliate Program     

Tango Jona
Tangokurs Rapperswil-Jona