International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 3, Issue 1, January 2013)

Image Steganography: A Review

Shikha Sharda1, Sumit Budhiraja2
1

Student, 2Assistant Professor, Department of Electronics and Communication Engineering, UIET, Panjab University
Chandigarh

Abstract Steganography can be defined as the study of

invisible communication that usually deals with the ways of
hiding the existence of the communicated message. If it is
achieved successfully, the message does not attract attention
from eavesdroppers and attackers. The main objectives of
steganography are undetectability, robustness (resistance to
various image processing methods and compression) and
capacity of the hidden data. These are the main factors
which make it different from other techniques
watermarking and cryptography. This paper includes the
important steganography methods and the main focus is on
the review of steganography in digital images.

For decades people have hidden information in

different ways. Steganography is a type of hidden
communication that literally means covered writing
(from the Greek words stegano or covered and graphos
or to write). In 1550, Jerome Cardan, an Italian
mathematician, proposed a scheme of secret writing
where a paper mask with holes is used. The user needs to
write his secret message in such holes after placing the
mask over a blank sheet of paper. Then remove the mask
to fill in the blank parts of the page and in this way the
message appears as innocuous text [2].
The performance of a steganographic system can be
measured using several properties. The most important
property
is
the
statistical
undetectability
(imperceptibility) of the data, which shows how difficult
it is to determine the existence of a hidden message.
Other associated measures are the steganographic
capacity, which is the maximum information that can
safely embedded in a work without having statistically
detectable objects and robustness, which refers to how
well the steganographic system resists the extraction of
hidden data.
This paper is organised as follows. Section II briefly
discusses the basic idea of steganography technique.
Section III describes the spatial domain method which
involves encoding at the LSBs level. Section IV
describes the frequency domain methods such as discrete
cosine transform (DCT) and discrete wavelet transform
(DWT). Section V describes adaptive steganography
which is a special case of the spatial and transforms
techniques. Section VI represents analysis of the different
existing methods of steganography. Finally section VII
gives the conclusion.

KeywordsAdaptive steganography, Frequency domain,

Image steganography, Spatial domain, Steganography.

I.

INTRODUCTION

In this modern era, internet offers great convenience in

transmitting large amounts of data in different parts of
the world. However, the safety and security of long
distance communication remains an issue. In order to
solve this problem has led to the development of
steganography schemes.
Steganography is the science that communicates secret
data in an appropriate multimedia carrier, e.g., image,
audio, and video files.
Steganography is different from cryptography. The
main objective of cryptography is to secure
communications by changing the data into a form so that
it cannot be understand by an eavesdropper. On the other
hand, steganography techniques tend to hide the
existence of the message itself, which makes it difficult
for an observer to figure out where exactly the message
is.
There are other two technologies that are closely
related to steganography are watermarking and
fingerprinting [1]. These technologies are mainly
concerned with the protection of intellectual property,
thus the algorithms have different requirements than
steganography. In watermarking all of the instances of an
object are marked in the same way. On the other hand,
in fingerprinting unique marks are embedded in distinct
copies of the carrier object that are supplied to different
customers. This enables the intellectual property owner
to identify customers who break their licensing
agreement by supplying the property to third parties [1].

II.

STEGANOGRAPHY MODEL

A basic steganographic model is shown in figure 1.

The process of embedding can be explained as:
Let
represents the cover image, and
the stegoimage. Let
be a stego-key (as a seed used to encrypt
the message, which can be set to for simplicity), and
let
be the message that the sender wishes to send.
Then,
represents an embedded message and
represents the extracted message. Therefore,

707

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 3, Issue 1, January 2013)
A lossless and reversible steganography scheme has
been introduced that use each block of quantized discrete
cosine transformation (DCT) coefficients in JPEG
images for embedding secret data [6]. In this scheme, the
two successive zero coefficients of the mediumfrequency components in each block are used to hide the
secret data. This method results in a high image quality
of stego image and successfully achieves reversibility.
A reversible data hiding scheme that use the histogram
shifting method based on DCT coefficients was proposed
[7]. Cover images are partitioned into several different
frequencies, and the high-frequency parts are used for
embedding the secret data. For hiding secret data, this
method of histogram shifting shifts the positive
coefficients around zero to the right and the negative
coefficients around zero to the lef . It improves the hiding
capacity and quality of the stego-images. On reversing
the frequency domain stego-image back to the spatial
domain image may cause underflow and overflow
problems.
Wavelets transform (WT) converts spatial domain
information to the frequency domain information.
Wavelets are used in the image steganographic model
because the wavelet transform clearly partitions the highfrequency and low-frequency information on a pixel by
pixel basis. Many practical tests propose to use the
Wavelet transform domain 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 discrete wavelet transformation (HDWT)based reversible data hiding method was proposed in
2009 [8]. In this method a spatial domain image is
transformed into a HDWT-based frequency domain
image and then the high frequency coefficients are used
to embed the secret data. This method provides a high
hiding capacity and a good stego-image quality.
In the recent year DWT based algorithm for image
data hiding has been proposed that uses CH band of
cover image for hiding the secret message. Vijay kumar
[9] proposed an algorithm in which secret message is
embed in different bands of cover image. PSNR has been
used to measure the quality of stegano image and it gives
better PSNR by replacing error block with diagonal detail
coefficients (CD) as compare to other coefficients.
A new image steganography technique based on
Integer Wavelet Transform (IWT) and Munkres'
assignment algorithm was introduced [10]. IWT converts
spatial domain information to the frequency domain
information. For embedding secret data, assignment
algorithm is used for best matching between blocks.
Stego image is subjected to various types of image
processing attacks and it shows high robustness against
these attacks.

Figure 1: Basic digital Steganography Encoder

III.

STEGANOGRAPHY IN IMAGE SPATIAL DOMAIN

Steganography techniques that modify the cover image

and the secret image in the spatial domain are known as
spatial domain methods. It involves encoding at the LSBs
level.
Least Significant Bit Substitution (LSB) [3] is the
most commonly used stenographic technique. The basic
concept of Least Significant Bit Substitution includes the
embedding of the secret data at the bits which having
minimum weighting so that it will not affect the value of
original pixel.
A new steganographic method to hide a secret
message into a gray -valued cover image was proposed
[4]. For embedding a secret message, a cover image is
partitioned into non-overlapping blocks of two
consecutive pixels. In each block, a difference value is
calculated from the values of the two pixels. Then that
difference value is replaced by a new value to embed the
value of the secret message. This method produces a
more imperceptible result than those obtained from
simple least-significant-bit substitution methods. The
embedded secret message can be extracted from the
resulting stego-image without referencing the original
cover image.
Iuon-Chang Lin [5] proposed a Data hiding scheme
with distortion tolerance which uses spatial domain for
hiding data. This method provides distortion tolerance
and gives better quality of processed image. This scheme
provides effective results than other schemes in terms of
distortion tolerance.
As LSB insertion is simpler and good for
steganography, we can try to improve one of its major
drawbacks: the ease of extraction. We dont want that an
eavesdropper be able to read everything we are sending.
IV.

STEGANOGRAPHY IN FREQUENCY DOMAIN

The need for enhanced security, has led to the

development of other algorithms. LSB technique has
weak resistance to attacks. So to overcome this
shortcoming, researchers found a better way for hiding
information in areas of the image that are less exposed to
compression, cropping, and image processing.

708

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 3, Issue 1, January 2013)
Prabakaran G. [11] proposed a steganography
approach for hiding a large-size secret image into a
small-size cover image. Arnold transformation is
performed to scrambles the secret image. Both secret and
cover images are decomposed using discrete wavelet
transform (DWT) and followed by Alpha blending
operation. Discrete wavelet coefficients are used for
hiding the data in order to maximize the hiding capacity.
This DWT based approach provides high security and
certain robustness.
V.

As a performance measure for image distortion due to

embedding, the well-known peak-signal-to noise ratio
(PSNR), which is categorized under difference distortion
metrics, can be applied to stego images .

Where MSE denotes the mean square error, which is

given as

ADAPTIVE S TEGANOGRAPHY

Adaptive steganography is a special case of the spatial

and transform techniques. Moreover, it is introduced as
statistics-aware embedding and masking. Global
statistical characteristics of the image are basically used
before any attempt to deal with its frequency transformed
coefficients. These statistics decide what changes can be
made. A random adaptive selection of pixels actually
characterizes this method, relying on the cover image and
the selection of pixels in a block with a large standard
deviation (STD). The latter is intended to avoid areas of
uniform color, such as smooth areas. This technique is
known for exploiting images with existing or deliberately
added noise and with images that show color complexity
[12-15].
An adaptive least-significant bit (LSB) steganographic
method was proposed [16]. This method includes pixelvalue differencing (PVD) which uses the difference value
of two consecutive pixels to estimate how many secret
bits will be embedded into the two pixels. The PVD
approach is used to differentiate the smooth and edge
areas. A k-bit LSB substitution method is used for hiding
data in the pixels located in the edge areas. The range of
difference values is adaptively divided into three
different levels that are lower level, middle level, and
higher level. This method results in larger payload
capacity and high image quality.
A new method which makes use of more surrounding
pixels around a target pixel to find the most appropriate
capacity value in order to improve imperceptibility was
introduced [17]. As compare to other steganographic
techniques which use either three or four adjacent pixels
around a target pixel, this technique is able to utilize all
eight adjacent neighbours, which improves the
imperceptibility value.
VI.

Where and are the image coordinates,

and
are the dimensions of the image,
is the generated
stego-image and
is the cover image. Also
represents the maximum value in the image.
The other Image quality parameters are normalized
cross correlation, average difference, and maximum
difference [11].
Normalized cross correlation (NCC) is defined as in
(1)
(1)
Average difference (AD) is defined as in (2)
(2)
Maximum difference (MD) is defined as in (3)
(3)
The original cover image sized
image sized
, and the

and

and the stego

are pixel

located at the row and the

column of images and
, respectively .
Spatial domain techniques have large payload but
often offset the statistical properties of the image. It is not
robust against lossy compression, image filters, rotation,
cropping and translation noise. As LSB insertion is
simpler and good for steganography, we can try to
improve one of its major drawbacks: the ease of
extraction. We dont want that an eavesdropper be able to
read everything we are sending.
DCT based domain techniques are less prone to
attacks than the spatial domain methods at the expense of
capacity. It is not robust against rotation, cropping and
translation. Embedding in the DWT domain gives better
result [18].

ANALYSIS

Three common requirements, security, capacity, and

imperceptibility, may be used to rate the performance of
steganographic techniques.

709

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 3, Issue 1, January 2013)
VII.

[6 ] C.-C. Chang et al., Reversible hiding in DCT-based compressed

images, Information Sciences 177 (2007) 27682786.
[7 ] Yih-Kai Lin, High capacity reversible data hiding scheme based
upon discrete cosine Transformation, The Journal of Systems
and Software 85 (2012) 2395 2404.
[8 ] Y.-K. Chan et al, A HDWT-based reversible data hiding
method, The Journal of Systems and Software 82 (2009) 411
421.
[9 ] Vijay Kumar and Dinesh Kumar, Performance
Evaluation of
DWT Based Image Steganography, 2010 IEEE 2nd International
Advance Computing Conference.
[10 ] N.Raftari, Digital Image Steganography
Based on Integer
Wavelet Transform and Assignment Algorithm, 2012 Sixth Asia
Modelling Symposium.
[11 ] Prabakaran. G and Bhavani.R, A Modified Secure Digital Image
Steganography Based on Discrete Wavelet Transform, 2012
International Conference on Computing, Electronics and
Electrical Technologies [ICCEET].
[12 ] E. Franz and A. Schneidewind., Adaptive steganography based
on dithering, in Proc. Of the 2004 workshop on Multimedia and
Security, 2004. pp. 56-62.
[13 ] R. Bohm and A. Westfeld, Breaking cauchy model-based JPEG
steganography with first order statistics, In Proc. of
ESORICS'2004, 2004. pp. 125-140.
[14 ] A.M. Fard, M. Akbarzadeh-R., and F. Varasteh-A., A new
genetic algorithm approach for secure JPEG steganography, in
Proc. of IEEE International Conference on Engineering of
Intelligent Systems ICEIS, 2006. pp. 216-219.
[15 ] A. Shaddad, J. Condell, K. Curran, and P. Mckevtt.,Biometric
inspired digital image steganography, In Proc. of the 15th
Annual IEEE International Conference and Workshop on the
Engineering of Computer Based Systems, 2008. Pp. 159-168.
[16 ] Yang et al., Adaptive data hiding in edge areas of images with
spatial LSB domain systems, IEEE transactions on information
forensics and security, vol. 3, no. 3, september 2008.
[17 ] Masoud Afrakhteh and Subariah Ibrahim, Adaptive
Steganography Scheme Using More Surrounding Pixels, IEEE
International Conference On Computer Design And Appliations
(2010).
[18 ] Cheddad, J. Condell, K. Curran and P.M. Kevitt. (2010), Digital
image steganography: survey and analysis of current methods,
Signal Processing Journal. [On line]. 90(3), pp.727-752.
Available: http://www.abbascheddad.net/Survey.pdf [Aug. 2011].

CONCLUSION

This paper reviewed the main steganographic

techniques. Each of these techniques tries to satisfy the
three most important factors of steganographic design
(imperceptibility or undetectability, capacity, and
robustness). LSB techniques in a spatial domain have a
high payload capacity, but they often fail to prevent
statistical attacks and are thus easily detected.
The promising techniques such as DCT, DWT and the
adaptive steganography are not prone to attacks,
especially when the hidden message is small. They
modify the coefficients in the transform domain, thus
results in minimum image distortion. Generally, such
techniques tend to have a lower payload when they are
compared to the spatial domain algorithms. The
experiments on the discrete cosine transform (DCT)
coefficients have introduced some promising results and
then they have diverted the researchers attention towards
JPEG images. Working at some level like that of DCT
turns steganography much more powerful and less prone
to statistical attacks. Embedding in the DWT domain
reveals a sort of constructive results and outperforms
DCT embedding.
REFERENCES
[1 ] Anderson, R.J. & Petitcolas, F.A.P., On the limits of
steganography, IEEE Journal of selected Areas in
Communications, May 1998.
[2 ] S.B. Sadkhan., Cryptography: Current status and future trends,
in Proc. IEEE Conference on Information & Communication
Technologies, 2004, pp. 417-418.
[3 ] Chan, C.K., Chang, L.M., Hiding data in image by simple LSB
substitution, Pattern Recognition, vol 37, pp.469-471, 2003.
[4 ] Da-Chun Wu, Wen-Hsiang Tsai, A steganographic method for
images by pixel-value differencing, Pattern Recognition Letters
24 (2003) 16131626.
[5 ] I.-C. Lin et al, Hiding data in spatial domain images with
distortion tolerance, Computer Standards & Interfaces 31 (2009)
458464.

710