Escolar Documentos
Profissional Documentos
Cultura Documentos
Abstract- This paper provides a comparison of some of the replaces least significant bits plane of the image by a secret
steganalysis methods proposed in the literature, and using message or its encrypted version [8, 9]. The hiding capacity
these comparison results, a global steganalysis methodology is of this technique is directly related with the image size.
proposed. The secret message detection capacities of these Principal advantages of the LSB embedding technique are
steganalysis methods are evaluated using stegoimages
simple implementation, imperceptibility of hidden message
generated by typical data hiding algorithms. The evaluation of
steganalysis methods is realized in terms of false negative and to human visual system and high embedding rate of secret
false positive error rates using 100 images. There isn’t any data. Some public domain tools, such as S-tools, Invisible
steganalysis that can detect presence of secret message in all Secret and J-Steg, use this technique. However the LSB
type of stegoimages. Therefore, to realize a reliable analysis embedding technique is generally vulnerable to statistical
about a suspicious image, several steganalysis methods must be analysis.
efficiently combined. In this paper, some considerations about Frequency domain embedding methods have some
steganalysis are provided using the results obtained of the advantages and disadvantages respect to spatial domain’s
comparison of steganalysis methods. one. Principal advantage is robustness against simple
statistical analysis such as those mentioned in [1] and [3],
Keywords-steganalysis; steganography; false negative error;
stegoimage; data hiding. and the principal disadvantage is the limitation of the
amount of embedded data. Recently, some steganalysis
methods that try to detect the presence of secret information
I. ITRODUCTION in stegoimages generated by frequency domain embedding
method have been proposed [6]. Many of these methods
Steganography is a science or art of secret extract some relevant features from an image in frequency
communication and recently digital steganography has domain, and using these features, stegoimages are
become a hot research issue, due to the wide use of Internet discriminated from natural images. The extracted features
as popular communication media. The goal of digital from the image are fed to a classifier such as neural
steganography is to conceal covert message in digital networks, Bayesian classifier, Support Vector Machine, etc.,
material in totally innocent manner. Even though digital to determine if the image has secret message or not. Each
images, audio files, video data and all types of digital files steganalysis method shows its efficiency for some types of
can be considered as a cover material to conceal secret steganography or some types of images, however there isn’t
information, in this paper, we consider only digital images a reliable comparison between various steganalysis methods
as cover material. After hiding a secret message into the from false negative and false positive error rates points of
cover image, we get an image with secret message: so-called view.
stegoimage, which is transmitted to a receptor via popular In this paper we present a comparison between three
communication channels or put on some Internet web-site. steganalysis methods using 100 natural images and 500
To design useful steganography algorithm, it is very stegoimages generated by five steganography methods: two
important that the stegoimage does not have any visual LSB embedding methods, two Discrete Cosine Transform
artifact and it is statistically similar to natural images. If a (DCT) domain embedding methods [10,11] and Bit Plane
third party or observer has some suspicion over the Complexity Segmentation Steganography (BPCS) method
stegoimage, steganography algorithm becomes useless [1]. [4], [12,13]. Comparison results suggest us some
During the last decade, many steganographic algorithms for consideration about a global steganalysis methodology.
digital images have been proposed [2-4]. On the other hand, The remainder of this paper is organized as follows.
many steganalysis methods, whose goal is to analyze an Section II briefly describes the steganography methods used
image to detect the presence of a hidden message in it, are to generate stegoimages. In Section III, three steganalysis
also proposed [5-7]. The image steganography algorithm methods are described. Section IV shows comparison
can be classified in two classes by its embedding domain: results of Steganalysis methods using the above mentioned
spatial domain embedding method and frequency domain steganography algorithms and the some considerations
embedding method. LSB embedding technique is one of the about efficient steganalysis methodology are provided.
popular spatial domain embedding techniques, which Finally, the conclusions are summarized in Section V.
Authorized licensed use limited to: M H Saboo Siddik College of Engineering. Downloaded on August 09,2010 at 09:16:26 UTC from IEEE Xplore. Restrictions apply.
stegoimage, therefore the histogram distribution only can
II. STEGANOGRAPHY METHODS not be used for this purpose. Zhang et al. proposed the
difference image histogram method [15], that generates a
In this section five data hiding methods that are used to difference image D, calculating the difference value
generate stegoimages are described. between two adjacent pixels of the image as given by
103
Authorized licensed use limited to: M H Saboo Siddik College of Engineering. Downloaded on August 09,2010 at 09:16:26 UTC from IEEE Xplore. Restrictions apply.
This method can be applied directly to grayscale images, but 1) De-correlation of wavelet transform: The
for color images, some considerations must be taken histograms of all wavelet subbands only reflect the
account. statistical distribution of coefficients in the subband, but it
doesn’t reflect the correlation of the coefficients within this
B. Closest Color Pair Method subband. The wavelet transform is well known for its
Fridrich et al. observed that the number of close color capability of multi-resolution decomposition and
pairs is increased considerably when an image has a secret coefficients de-correlation. It is known that for discrete
message embedded in its LSB plane, and they proposed a wavelet transform, different high frequency subbands within
steganalysis method for LSB embedding technique using a one level will be uncorrelated to each other. The features
ratio between the number of closest color pair and all pairs extracted from one high frequency subband are thus
of colors of the image [5]. In this method, the number of all uncorrelated to that extracted from another high frequency
color pairs existing in an image and the number of close subband at the same level. Therefore, features from different
color pairs within all existing color pairs are computed. dimensions most likely uncorrelated to each other. From this
Then on purpose, a LSB embedding steganography point of view, this multi-dimensional feature vector will be
algorithm is applied to the image, and also number of suitable to represent the image for steganalysis purpose.
existing color pairs and close color pairs are computed. The
condition of close color pair 2) Characteristic Function and Its Statistical
( C1 = [ R1 , G1 , B1 ] , C2 = [ R2 , G2 , B2 ]) is given by Moments: The data hiding process can be modelled as an
additive signal, which is independent to the cover-image;
this signal is added to the cover media, as proposed in [16].
( R1 − R2 )2 + ( G1 − G2 )2 + ( B1 − B2 )2 ≤ 3 (5) It is well known that the effect of the additive signal on the
image is equivalent to a convolution of two probability
The ratios R and R’ between the number of close color pair density functions (PDFs). According to [17], one
and the number of all color pairs is calculated for the image interpretation of the characteristic function (CF) is that it is
under analysis and its steganography version as complex conjugate of Fourier transform of the PDF. We can
consider the PDF as the normalized version of a histogram,
(6) in this case the image histogram and the subbands
P P' coefficients histogram. The CF is defined as:
R= R' =
⎛U ⎞ ⎛U ' ⎞
⎜⎜ ⎟⎟ ⎜⎜ ⎟⎟
⎝2 ⎠ ⎝2 ⎠ N −1 −
2πjxfi
H ( f i ) = DFT (h[x ]) = ∑ h[x ]e N
(8)
where P and P’ are the number of close color pairs, and U x =0
and U’ are number of all color pairs in the image and its
steganography version, respectively. If (7) is satisfied, the where H(fi) is the equal to CF, DFT(h[x]) is the Discrete
image can be considered as natural image, otherwise the Fourier Transform of the histogram, N is the total number of
image contains some secret message. points in the horizontal axis of the histogram, fi is frequency
component.
R Owing to the de-correlation capability of the discrete
≥ Th (7) wavelet transform (DWT), the coefficients of different
R'
subbands at the level, we can assume independence to each
Fridrich et al. proposed, after exhaustive proves, that a other. Therefore, the features generated from different
suitable threshold value Th is 1.1. wavelet subbands at the same level are independent to each
other as well. This property is desirable for steganalysis.
C. Features Extraction Method Therefore, in [18] is proposed to use the statistical
Recently the DCT domain embedding methods were moments of the characteristic functions of wavelet subbands
proposed instead of LSB embedding method. The as futures for steganalysis. The n-th statistical moment of a
efficiency of two steganalysis methods mentioned above for CF is defined as follows
LSB embedding method is shown, but there are not
N /2
information about the efficiency to detect the DCT domain-
based embedding method. The feature extraction method ∑f
j =1
j
n
H( fj)
proposed by [6] extracts 39 features, which are used to Mn = N /2
(9)
classify images as natural images or stegoimages.
In this section, we describe the features based on ∑ H( f
j =1
j )
statistical moments of wavelets characteristic function for
the steganalysis. where ׀H(fi) ׀is the magnitude of the CF component at
frequency fi, N is the total number of points in the horizontal
104
Authorized licensed use limited to: M H Saboo Siddik College of Engineering. Downloaded on August 09,2010 at 09:16:26 UTC from IEEE Xplore. Restrictions apply.
axis of the histogram. The zero frequency component of the steganalysis algorithm detects the presence of some secret
CF is excluded from the calculation of moments because it message, when the image does not contain any message.
represents only the summation of all components in the The meanings of symbols used in the table I are shown
discrete histogram. For an image, the zero frequency by table II.
component is the total number of pixel, while for a wavelet
subband, it is the total number of the coefficients in the
subband. In either case, it does not change during the data Table I False Negative And False Positive Error Rate (%)
hiding process. In order to get the feature, the subbands
decomposed by Haar wavelets until three levels were used. Data Hiding
DH CC FE
Therefore, there are 12 subbands, denoted by LL1,HL1, Methods
LH1, HH1, LL2, HL2, LH2, HH2, LL3, HL3, LH3, HH3. FN FN FN
The first three moments for each of subbands and the test S-tool 0 0 2
image, denoted by LL0, result a vector with 39 features.
IS 0 0 2
105
Authorized licensed use limited to: M H Saboo Siddik College of Engineering. Downloaded on August 09,2010 at 09:16:26 UTC from IEEE Xplore. Restrictions apply.
steganalysis method can be applied to the image to reduce [15] T. Zhang, X. Ping,"A new approach to reliable detection of LSB
global false negative error rates. steganography in natural images" Signal Processing, vol. 83, no. 10,
pp. 2085-2093, 2003.
[16] K. Sullivan, U. Madhow, S. Chandrasekaran, and B. S. Manjunath,
V. CONCLUSION “Steganalysis of Spread Spectrum Data Hiding Exploiting Cover
Memory," the International Society for Optical Engineering,
In this paper, a performance comparison results of three Electronic Imaging, San Jose, CA, USA, 2005.
[17] A. Leon-Garcia, Probability and random processes for electrical
steganalysis methods, using 100 natural images and 500 engineering, 2nd edition, reading, MA: Addison-Wesley Publishing
stegoimage generated by five steganography methods, are Company, 1994.
presented. To perform a reliable analysis about a suspicious [18]G. Xuan, Y. Q. Shi, D. Zou, J. Gao, C. Yang, Z. Zhabg, P. Chai, W.
image, several steganalysis methods must be combined. Chen, C. Chen, “Steganalysis based on multiple features formed by
statistical moments of wavelet characteristic Functions," IH 2005
The comparison results suggest us some steganalysis LNCS 3727, pp. 262-277, Springer-Verlag Berlin Heidelberg 2005.
methodology that can be followed. DH and CC steganalysis
methods can be used to detect the presence of the hidden
message of stegoimages generated by LSB embedding
steganography, such as S-tools and Invisible Secret. About
the images considered as natural images, after DH and CC
steganalysis, furthermore FE steganalysis method is applied
to detect stegoimages generated by DCT embedding
methods or BPCS method. Also more reliable steganalysis
method for frequency domain embedding steganography
must be developed and analyzed.
VI. REFERENCES
106
Authorized licensed use limited to: M H Saboo Siddik College of Engineering. Downloaded on August 09,2010 at 09:16:26 UTC from IEEE Xplore. Restrictions apply.