International Journal of Computer Science Trends and Technology (IJCST) Volume 5 Issue 2, Mar Apr 2017

RESEARCH ARTICLE OPEN ACCESS

A Review: A Digital Image Steganography

Mrs.M.A.Wakure [1], Mrs.S.A.Wakure [2]
Department of Computer Science & Engineering [1]
Dr. BAMU University, Osmanabad
Department of Electronics & Telecommunication Engineering [2]
JSPMS COE, Pune University
Pune - Maharashtra

ABSTRACT
Now a days, a lot of applications are Internet based and in some cases it is desired that the communication be made secret,
digital communication has become an essential part of infrastructure. Information hiding has an important research field to
resolve the problems in network security, quality of service control & secure communication through public & private channels.
Steganography is the science that involves communicating secret data in an appropriate multimedia carrier, e.g., image, audio,
and video file. Steganography has various useful applications. Steganographys ultimate objectives, which are
undetectability, robustness (resistance to various image processing methods and compression) and capacity of the hidden data,
are the main factors that separate it from related techniques such as watermarking and cryptography. This paper provides a
state-of-the-art review of existing methods of steganography in digital images.
Keywords :- Information hiding, Security, Steganography, Staganalysis.

I. INTRODUCTION DATA HIDING

Internet has an become essential the most effective and fastest
media for communication and transmitting Large amount of
data in different part of the world. The safe & secure
STEGANOGRAPHY WATERMARKING DATAMANAGEMENT
transmission of long distance communication remains an issue.
Hence, the need for secret communication is required.
Cryptography [1,2] and Steganography are the two fields for
data security.[3,4]. In Cryptography, the data is encrypted so
TEXT IMAGE AUDIO/VIDEO
that it cannot understood by any one else. The encrypted data
is unreadable but is not hidden from the eavesdroppers. Fig. 1. Classification of Data Hiding
Cryptography has helped in data surety but it has some
disadvantages. The encrypted data will arous suspicion to Steganography technique based on the two approaches i.e
unwanted users and there is possibility of it being decrypted spatial domain and frequency domain approach. In spatial
or being suppressed. The Steganography solve this problem domain approach secret message are embedded into least
by embedding d in the cover object so that it is hard to detect. significant pixels of cover image. They are fast but sensitive
to image processing attacks. In frequency domain
There are other two techniques related to steganography are transforming the cover image into the frequency domain
watermarking and fingerprinting. Watermarking is the process coefficients before embedding secret messages in it. The
that embeds data called watermark tag or label into a transformation can be either Discrete Cosine Transform
multimedia object such that watermark can be detected or (DCT) and Discrete Wavelet Transform (DWT) . Though
extracted to make an assertion about the object. In these methods are more difficult & slower than spatial domain.
fingerprinting unique marks are embedded in distinct copies They have an advantage of being more secure and noise
of the carrier object that are supplied to different customers. tolerant[5].
Steganography is the art and science that hides the information This paper is organized as follows. Section II
in an appropriate cover carrier like image , text, audio and describe the spatial domain method which involves encoding
video media. Following fig.1 shows classification of data at the LSBs level. Section III describes the frequency domain
hiding techniques [14]. methods such as discrete cosine transform (DCT) and discrete
wavelet transform (DWT). Section IV describes analysis of

ISSN: 2347-8578 www.ijcstjournal.org Page 33

 International Journal of Computer Science Trends and Technology (IJCST) Volume 5 Issue 2, Mar Apr 2017
different existing methods of Steganography. Finally section Spatial Steganography generates unusual patterns
V describes conclusion. such as sorting of colour palettes, relationships between
indexed colours, exaggerated noise , etc, all of which leave
II. STEGANOGRAPHY IN THE SPATIAL traces to be picked up by Steganalysis tools. There is a serious
DOMAIN conclusion drawn in the literature- LSB encoding is
In spatial domain methods a Steganographer modifies the extremely sensitive to any type of filtering or manipulation
secret data and the cover medium in the spatial domain, which of the stego-image. Scaling, rotation, cropping, addition of
is the encoding at the level of the LSBs.This method has the noise, or lossy compression to the stego-image is likely
largest impact compared to the other two methods even destroy the message. Furthermore an attacker can easily
though it is known for its simplicity [6, 7]. Embedding in the remove the message by removing (zeroing) the entire LSB
4th LSB generates more visual distortion to the cover image plane with very little change in the perceptual quality of
as the hidden information is seen as non-natural. Potdar et the modified stego-image [6]. Almost any filtering process
al., used this technique in producing fingerprinted secret will alter the values of many of the LSBs [11]. By inspecting
sharing Steganography for robustness against image cropping the inner structure of the LSB, Fridrich et al., claimed to be
attacks[8]. This paper addressed the issue of image cropping able to extract hidden messages as short as 0.03bpp (bit per
effects rather than proposing an embedding technique. The pixel) [12]. Xiangwei et al., stated that the LSB methods can
logic behind their proposed work is to divide the cover image result in the pair effect in the image histograms. This pair
into sub-images and compress and encrypt the secret data. The effect phenomenon is observed in Steganography based on the
resulting data is then sub-divided and embedded into those modulus operator. This operator acts as a means to generate
images portions. To recover the data a Lagrange Interpolating random locations to embed data. It can be a complicated
Polynomial was applied along with an encryption algorithm. process or a simple one like testing in a raster scan if a pixel
The computational load was high, but their algorithm value is even then embed, otherwise do nothing [13]. The
parameters, namely the number of sub-images (n) and the least-significant bit (LSB) insertion method is the most
threshold value (k) were not set to optimal values leaving the common and easiest method for embedding messages in an
reader to guess the values. If n is set, for instance, to 32 that image. The basic idea of LSB embedding is to embed the
means we are in need of 32 public keys, 32 persons and 32 message bit at the rightmost bits of pixel value so that the
sub-images, which turns out to be unpractical. Moreover, data embedding method does not affect the original pixel value
redundancy that they intended to eliminate does occur in their greatly. The formula for the embedding is as follows:
stego-image. Shirali-Shahreza [9] exploited Arabic and
Persian alphabet punctuations to hide messages. While their X2 = X X mod 2k + b
method is not related to the LSB approach, it falls under the where k is the number of LSBs to be substituted. The
spatial domain. Unlike English which has only two letters extraction of message from the high frequency coefficients is
with dots in their lower case format, namely i and j, given as: b = X mod 2k
Persian language is rich in that 18 out of 32 alphabet letters
have points. The secret message is binarized and those 18 There are two types of LSB insertion methods, fixed-sized and
letters points are modified according to the values in the variable-sized. The former embeds the same number of
binary file. Colour palette based Steganography exploits the message bits in each pixel of the cover-image. On embedding
smooth ramp transition in colours as indicated in the colour fixed four random bits in the four LSBs of each pixel, some
palette. The LSBs here are modified based on their positions false contours can occur. The unwanted artifacts may arise
in the said palette index. Johnson and Jajodia were in favour suspicion and defeat the purpose of steganography. To treat
of using BMP (24-bit) instead of JPEG images. this problem, either fewer bits must be used for message
embedding or a variable-sized method needs to be applied.
Their next best choice was GIF files (256-color). BMP as well For the variable-sized embedding method, the number of
as GIF based Steganography apply LSB techniques, while LSBs in each pixel used for message embedding depends on
their resistance to statistical counter attack and compression the contrast and luminance characteristics. Thus the most
are reported to be weak . BMP files are bigger in size than important requirement is maintaining the image fidelity while
other formats which is improper for network transmissions. adapting these local characteristics to estimate the maximum
However, JPEG images were at the beginning avoided embedding capacity [14]. Direct LSB techniques handles large
because of their compression algorithm which does not payload But often offset the statistical properties of the image .
support a direct LSB embedding into the spatial domain[10].

ISSN: 2347-8578 www.ijcstjournal.org Page 34

 International Journal of Computer Science Trends and Technology (IJCST) Volume 5 Issue 2, Mar Apr 2017
III. STEGANOGRAPHY IN THE image for hiding the secret message. Vijay kumar [18]
FREQUENCY DOMAIN proposed an algorithm in which secret message is embed in
different bands of cover image. PSNR has been used to
The need for enhanced security, has led to the development of measure the quality of stegano image and it gives better PSNR
other algorithms. LSB technique has weak resistance to by replacing error block with diagonal detail coefficients (CD)
attacks. So to overcome this shortcoming, researchers found a as compare to other coefficients. A new image steganography
better way for hiding information in areas of the image that technique based on Integer Wavelet Transform (IWT) and
are less exposed to compression, cropping, and image
Munkres' assignment algorithm was introduced. IWT converts
processing. Steganographic methods of the second type spatial domain information to the frequency domain
employ the transformed domain of a host image to hide secret information. For embedding secret data, assignment algorithm
data[15]. Transformation functions like the discrete cosine is used for best matching between blocks. Stego image is
transform (DCT) or discrete wavelet transform (DWT) are subjected to various types of image processing attacks and it
first exploited to transform the pixel values in the spatial
shows high robustness against these attacks. The experiments
domain to coefficients in the frequency domain. Then the on the Discrete Cosine Transform (DCT) coefficients showed
secret data are embedded in the coefficients. A lossless and promising results and redirected researchers attention towards
reversible steganography scheme has been introduced that use this type of image. In fact acting at the level of DCT makes
each block of quantized discrete cosine transformation (DCT) Steganography more robust and not as prone to many
coefficients in JPEG images for embedding secret data [16].
statistical attacks.
In this scheme, the two successive zero coefficients of the Prabakaran G. proposed a steganography approach
medium-frequency components in each block are used to hide for hiding a large-size secret image into a small-size cover
the secret data. This method results in a high image quality of image. Transformation is performed to scrambles the secret
stego image and successfully achieves reversibility. A image. Both secret and cover images are decomposed using
reversible data hiding scheme that use the histogram shifting
discrete wavelet transform (DWT) and followed by Alpha
method based on DCT coefficients was proposed [17]. blending operation [20]. Discrete wavelet coefficients are used
for hiding the data in order to maximize the hiding capacity.
Cover images are partitioned into several different frequencies, This DWT based approach provides high security and certain
and the high-frequency parts are used for embedding the robustness.
secret data. For hiding secret data, this method of histogram
shifting shifts the positive coefficients around zero to the right IV. ANALYSIS
and the negative coefficients around zero to the left . It
Three common requirements, security, capacity, and
improves the hiding capacity and quality of the stego-images.
imperceptibility, may be used to rate the performance of
On reversing the frequency domain stego-image back to the
steganographic techniques. As a performance measure for
spatial domain image may cause underflow and overflow
image distortion due to embedding, the well-known peak-
problems. Wavelets transform (WT) converts spatial domain
signal-to noise ratio (PSNR), which is categorized under
information to the frequency domain information. Wavelets
difference distortion metrics, can be applied to stego images.
are used in the image steganographic model because the
wavelet transform clearly partitions the high-frequency and
low-frequency information on a pixel by pixel basis. Many (1)
practical tests propose to use the Wavelet transform domain Where MSE denotes the mean square error, which is given as
for steganography because of a number of advantages. The
use of such transform will mainly address the capacity and
robustness of the Information Hiding system features. A Haar (2)
discrete wavelet transformation (HDWT)- based reversible
data hiding method was proposed in 2009 [19]. In this method Where and are the image coordinates, and are the dimensions
a spatial domain image is transformed into a HDWT-based of the image, is the generated stego-image and is the cover
frequency domain image and then the high frequency image. Also represents the maximum value in the image. The
coefficients are used to embed the secret data. This method original cover image sized and the stego image sized, and the
provides a high hiding capacity and a good stego-image and are pixel located at the row and the column of images and
quality. In the recent year DWT based algorithm for image , respectively.
data hiding has been proposed that uses CH band of cover

ISSN: 2347-8578 www.ijcstjournal.org Page 35

 International Journal of Computer Science Trends and Technology (IJCST) Volume 5 Issue 2, Mar Apr 2017
Spatial domain techniques have large payload but Computer Forensics, Cyber crime and Steganography
often offset the statistical properties of the image. It is not Resources, Digital Watermarking Links and
robust against lossy compression, image filters, rotation, Whitepapers , Apr 1999.
cropping and translation noise. As LSB insertion is simpler [7] Kermani, Z. Z. and Jamzad, M., A Robust
and good for steganography, we can try to improve one of its Steganography Algorithm Based on Texture
major drawbacks: the ease of extraction. We dont want that Similarity using Gabor Filter. Proceedings of IEEE
an eavesdropper be able to read everything we are sending. 5th International Symposium on Signal Processing
DCT based domain techniques are less prone to and Information Technology , 18-21 Dec. 2005,
attacks than the spatial domain methods at the expense of 578-582.
capacity. It is not robust against rotation, cropping and [8] Potdar, V. M., Han, S. and Chang, E., Fingerprinted
translation. Embedding in the DWT domain gives better result. Secret Sharing Steganography for Robustness against
Image Cropping Attacks. Proceedings of IEEE's 3 rd
International Conference on Industrial Informatics
V. CONCLUSION (INDIN), Perth, Australia, 10-12 August 2005.
Steganography is the art and science of writing hidden [9] Shirali-Shahreza, M. H. and Shirali-Shahreza, M.,
messages in such a way that no one, apart from the sender A New Approach to Persian/Arabic Text
and intended recipient, suspects the existence of the message, Steganography. Proceedings of 5th IEEE/ACIS
a form of security through obscurity. It is therefore a book on International Conference on Computer and
magic. It is emerging in its peak because it does not attract Information Science (ICIS-COMSAR 2006) 10-
anyone by itself. In this paper reviewed stegnographic 12July 2006, 310- 315.
techniques. In spatial domain LSB technique have high [10] Johnson, N. F. and Jajodia, S., Exploring
payload but they fail to prevent statistical attacks and thus Steganography: Seeing the Unseen , IEEE
easily detected. The emerging techniques such as DCT, DWT Computer, 31 (2): 26-34, Feb 1998.
and Adaptive Steganography are not an easy target for attacks, [11] Anderson, R. J and Petitcolas, F.A.P., On the Limits
especially when the hidden message is small. That is because of Steganography, IEEE Journal of Selected Areas
they alter bits in the transform domain, thus image distortion in Communications, 16(4): 474-481, May 1998.
is kept to a minimum. Generally these methods tend to have a [12] Fridrich, J., Goljan, M. and Du, R., Reliable
lower payload compared to spatial domain algorithms. In short Detection of LSB Steganography in Grayscale and
there has always been a trade off between robustness and Color Images. Proceedings of ACM, Special Session
payload. on Multimedia Security and Watermarking, Ottawa,
REFERENCES Canada, October 5, 2001, pp. 27- 30.
[13] Xiangwei Kong, Ziren Wang and Xingang You.,
[1] Bruice Schneier, Applied Cryptography Protocols,
Steganalysis of Palette Images: Attack Optimal
Algorithm and Source Code in C, Second edition.
Parity Assignment Algorithm, 860- 864, 06-09 Dec
Wiley India edition 2007.
2005.
[2] W.Diffie and M. E. Hellman, Exhaustive
[14] S.K.Muttoo,Sushil Kumar, A mulltilayred secure,
Cryptanalysis of the NBS Data Encryption Standard,
robust and high capacity image steganographic
IEEE Computer, Vol. 10, 1977, pp. 74-84.
algorithm, 2011.
[3] Abbas Cheddad, Joan Condell, Kevin Curran, Paul
[15] A.M. Fard, M. Akbarzadeh-R., and F. Varasteh-A.,
Mc Kevitt, Digital image steganography: Survey
A new genetic algorithm approach for secure JPEG
and analysis of current methods Signal Processing ,
steganography, in Proc. of IEEE International
(2010) 727752.
Conference on Engineering of Intelligent Systems
[4] S. Katzenbeisser, F.A.P. Petitcolas, Information
ICEIS, 2006. pp. 216-219.
Hiding Techniques for Steganography and Digital
[16] C.C. Chang et al., Reversible hiding in DCT-based
Watermarking, Artech House, Norwood, MA, 2000.
compressed images, Information Sciences 177
[5] Ahmed A. Abdelwahab and Lobha A. Hassan, A
(2007) 27682786.
discrete Wavelet Transform based technique for
image data hiding, 2 nd National Radio science [17] Yih-Kai Lin, High capacity reversible data hiding
Conference, Egypt,2008. scheme based upon discrete cosine Transformation,
[6] Lin, E. T. and Delp, E. J., A Review of Data Hiding The Journal of Systems and Software 85 (2012)
in Digital Images. Retrieved on 1.Dec.2006 from 2395 2404.

ISSN: 2347-8578 www.ijcstjournal.org Page 36

International Journal of Computer Science Trends and Technology (IJCST) Volume 5 Issue 2, Mar Apr 2017
[18] Vijay Kumar and Dinesh Kumar, Performance [20] Prabakaran. G and Bhavani.R, A Modified Secure
Evaluation of DWT Based Image Steganography, Digital Image Steganography Based on Discrete
2010 IEEE 2nd International Advance Computing Wavelet Transform,International Conference on
Conference. Computing, Electronics and Electrical Technologies
[19] Y.K. Chan et al, A HDWT-based reversible data [ICCEET], 2012.
hiding method, The Journal of Systems and
Software 82 (2009) 411421.

ISSN: 2347-8578 www.ijcstjournal.org Page 37