Conference PaperPDF Available

Image Spatial Domain Steganography: A study of Performance Evaluation Parameters

Authors:
  • Missan Oil Trainining Institute (MOTI)

Figures

Content may be subject to copyright.
Image Spatial Domain Steganography : A study of
Performance Evaluation Parameters
Mohanad Najm Abdulwahedand
Materials Department, University of
Technology, Baghdad, Iraq
mohanadnajmabdulwahed@gmail.com
*
Mustafa S. T.
School of Computing, Faculty of
Engineering, University Technology
Malaysia,
Basrah Oil Traning Institute,Ministry
of Oil , Iraq
tmimiymustafa@gmail.com
Mohd Shafry Mohd Rahim
School of Computing, Faculty of
Engineering, University Technology
Malaysia
shafry@utm.my
Abstract— In recent years, Steganography is deemed as a
distinctive field of study utilized for the safeguarding of data
from unallowed access. Steganography is signified as the skill
of concealing secret messages in obvious view within numerous
media sources like images, audio, texts, video, network
channel and others, for the purpose of not arousing any
concern. Meanwhile steganalysis is the art of using certain
attack strategies onto the the steganographic structure to
disclose the discrete data. This study explains clearly the
numerous evaluation factors according to image
steganographic algorithms. The efficacy of a steganographic
method is evaluated through three principle variables, which
are; embedding payload capacity, image quality measure and
security measure. The emphasis of this research on image
steganography that is the most famous method in the
steganographic field. Typically, the Least Significant Bit is the
main technique used, that is efficient, for the embedment of the
secret data. Additionally, the current research presents an
extensive knowledge-based Least Significant Bit (LSB)
information in numerous varying images formats. The entire
metrics that are displayed in this research are explained with
arithmetical equations, in addition to the discussions on several
significant trends carried out at the conclusion of this study.
Keywords— Steganography, Image Steganography (I.Stg.),
Least Significant Bit, Security, Evaluation Parameters
techniques.
I. INTRODUCTION
Steganography is a significant component of data
concealment, in which case secret information is concealed
within a cover file such as in image format for hiding the
secret data with no distortion within a cover [1]. Essentially,
there are six kinds of steganographic mechanisms: image,
text, audio, video, protocol, and DNA. Image steganography
possesses the capacity to conceal the secret information in a
cover image, and consequently, the quality of the cover
image will experience minor changes; however, the changes
are minute and not visible to the naked eye and is
undetected [2]. Image steganography methods are grouped
into two principal classes, known as; spatial domain (image-
domain )and frequency domain (transform domain) [3].
The spatial domain steganographic structure is applied
straight into the entire cover image pixels. Meanwhile, the
embedment of the secret data through the frequency
dominion steganographic method will be carried out
following certain reconstructions to change the image into
the frequency domain. The terminology of stego-image(S.I)
is derived from the hidden discrete data inside the original
cover image. There are numerous techniques which belong
to the spatial domain like LSB (matching and substitution),
Gray level modification GLV, and others [4]. The
frequency-domain techniques utilize diverse conversions
like Discrete Cosine Transform DCT, Discrete Fourier
Transform DFT and many more [4]. An alternative
steganographic method is known as the reversible
steganography. This system functions in the recovery of the
secret information of the S.I, which is embedded in the
cover likeness [5]. The Least Significant Bit (LSB)
interchange is the extremely famous I.stg. method. The LSB
interchange remains appropriate to the value of up to 4 LSB
levels, in order to obtain a bigger payload quantity, with
ease of execution. It is weak to certain attacks such as RS
analysis [6,7]. For the purpose of solving this problem, the
LSB can be applied in numerous differing ways. LSB
assemble can be produced by utilizing all of the LSB bits
inside the elements. The discrete data which is in binary
form can be masked at the lower manipulation of the LSB
bits assemble location for.
The three LSB bits are be examined in order to
improve the payload capacity with increased protection, in
which case two bits are solely utilized to conceal the secret
message inside two random selected bits, which is reliant
upon the secret data itself. This is termed as data-dependent
embedding [8-10]. The digital I.stg. consideration aspects
have essential significance to various image processing
executions and the evaluation of the techniques. The images
standard could experience lesser degradation at the location
that is utilized to conceal the secret data, whether it is able
to be seen by the human eye. Hence, algorithms are utilized
for the estimation of the image quality which could be
analyzed, and their quality is reported with payload capacity
and certainty, which is the aim of the consideration. There is
most reviewed material which has been published on
statistical consideration, that is used to differentiate the
cover image from the S.Is which were attained via varying
embedding methods [11 - 13].
In addition, past works have shown observable
appropriate favorable achievement on the controlled data
sets. However, there exists an inadequacy on the estimation
of the comparative studies on the recommended methods.
This may be caused by restricted amount of accessible
embedding techniques present or are associated with the
data set quality used for categorized.
The major provision of this study is to explain clearly
and extensively on the diverse evaluation parameters to
2019 IEEE 9th International Conference on System Engineering and Technology (ICSET 2019), 7 October 2019, Shah Alam, Malaysia
978-1-7281-0757-8©2019 IEEE
attain a great Image steganographic (I.Stg) achievement. In
addition, this study puts forward comprehensive information
on the LSB based I.Stg in varying differing resemblances
presentation.
The remainder of this study is arranged as the following.
Section 2 contains the clarity of the background of
steganography, an extensive description of image analysis,
in addition to the numerous varying LSB methods with
differing embedding bits was featured extensively.
Meanwhile, the digital I.Stg. performance consideration is
put forward in Section 3. Finally, Section 4. contains the
inference of this research.
II. STEGANOGRAPHY
/
BACKGROUND
Simmons in 1984 had initiated the introduction of the
concept of steganography framework [14], by giving an
example of the conveyance of secret message between two
inmates. The scenario is the hatching of a getaway plan for
sender and receiver who wishes to have a form of
communication from within the imprisonment area, however
the restriction to their communications came in the form of
messages that is compulsory to be examined by the third
party first; who will imprison them under solitary detention
on the smallest suspicion aroused of any transmission of
secret data. The steganography notion is typically modeled
by the challenges faced by the inmates as shown in Figure 1.
Sender conceals the secret message “SM” within a cover-
image “CI” through the utilization of an optional key
represented as "K". This procedure will produce the S.I "SI",
being transmitted via a public channel. As a result, the
process for the embedment of data embedding is illustrated
as:
Embedding scheme:
CI × SM × K SI, s = Emb(x,m,k) ………………. (1)
Extraction scheme:
"CI × K M, m = Ext (s, k)" ...…………….....…… .(2)
Fig 1. General block diagram of Steganography
As a result, based on Figure 1, the following definitions
are arrived at:
cover-image (CI): which is also termed as carrier image,
is used as a means and environment to hide the secret
information through the utilization of certain embedding
algorithms. S.I (SI); is the outcome obtained following the
embedment of the secret information (k), that is the objective
of the I.Steg system. Moreover, it is obligatory for the S.I to
possess similar cover image with no cover image quality
distortion. Secret Key (K); is utilized in the deciphering
process of the concealed information. Furthermore, the
Secret Message (SM); may exist in any form such as data,
file or image and many more, which may be hidden inside
the cover image. superintendent the warden may exercise a
passive or active disposition which is dependent upon the
type of measures that she executes upon any suspicion. A
passive superintendent would merely curb or disregard the
information, meanwhile, an active superintendent would
change the information to thwart the getaway plan. We pose
a supposition in our research that the superintendent is a
docile warden, who could not differentiate the cover
resemblance and the stego resemblance connection.
Therefore, steganalysis is the method which will assist the
superintendent to extricate the secret details from the S.I. She
will then examine for any deformities that arise from the
embedment process, and additional statistical artifacts
produced in the image after the embedment process, to
obtain a technique that will crack the steganographic
technique. It is appropriately be noted that the steganography
and steganalysis enhancement go hand-in-hand and happen
simultaneously.
A. Image Steganography (I.Stg.)
In recent times, Steganography has emerged as a
significant instrument in image format because the human
eyes are unable to concentrate on delicate picture details. A
slight alteration in the steganography of an image does not
have a noticeable image on the image. The image choice
possessing covert data is significant in the construction of
steganographic systems, however, the bottom colors with
consistent textured images are not appropriate for
steganography. There are numerous different I.Stg methods
with their principle objective of possessing high robust
nature, capacity and security.
From the various steganography compression methods,
there are only two kinds which are utilized in I.Stg which are
lossy compression and lossless compression. An exemplar of
lossless compression formats is the Huffman coding that
provides additional assurances [15]. Nearly the whole data
concealment methods target at the alteration of non-notable
data within the cover image. Among the most frequent
techniques of hiding secret information in a cover image, one
of it is the Least Significant Bit (LSB) injection procedure.
The positioning of the embedding details at the LSB of every
cover resemblance element is recommended by many easy
schemes [16,17]. Any changes on the LSB are not notable in
terms of impact on the quality of the resemblance according
to human perceptible, however, it is very vulnerable to
certain statistical attacks, that include cropping and
compression. The discrete data may be concealed by utilizing
the average significant bits within the cover image’s pixel. It
has been substantiated through experiments that the LSB
secret data length is predetermined with relatively high
accuracy.
B. Least Significant Bit (LSB) Approaches.
A secret information bit is the product of the
modification of the 8th bit of the LSB in a resemblance.
For the RGB element in a resemblance of 24-bits, each
pixel can keep 3 bits by adjusting a bit of every RGB
2019 IEEE 9th International Conference on System Engineering and Technology (ICSET 2019), 7 October 2019, Shah Alam, Malaysia
978-1-7281-0757-8©2019 IEEE
constituents as they individually are manifested by a byte.
A single resemblance of 800 × 600-pixel estimates can
cache 180,000 bytes or 1,440,000 bits of secretly injected
data in totality. It is the Least Significant Bit portion of
the cover image that is used for the embedment of data.
Figure 3 shows an example of the concealment of the
number 300 within the initial 8 bytes, in which 5 bits
solely need to be changed in the discrete data that had
undergone embedment process [18]. Thus, just 50% of
the image bits are required to be transformed for the
embedding of secret information with optimal cover size.
Approximately 256 potential intensities are present in
every individual primary color. Little differences happen
to colors intensity when the LSB is altered. The
adjustment cannot be seen by the human eye as the result
of not being sensitive to color development; hence, the
embedding of the secret information is positively carried
out [19].
Fig. 2. A general pictorial of LSB substitution
III. EVALUATION
OF
IMAGE
STEGANOGRAPHY
Predominantly, an ideal computation for performance or
steganographic systems effectiveness evaluation is lacking.
Nevertheless, it is a very crucial requirement to have an
evaluation plan. For the initiative of resolutions in :
A. Embedding Capacity (EC).
Hiding Capacity refers to the optimal/maximal hiding
volume or the bit-rate (depicted in Figure 4). The optimal
data quantity in bits, bytes, or kilobytes which can be
embedded within an image is termed as the utmost hiding
capacity for the image, meanwhile the quantity of bits per
pixel which can be embedded is termed as the bit-rate,
usually known as bits per byte (bpb) or bits per pixel.
(bpp). =



Fig. 3. Evaluation parameters criteria
Based on the fact that communications expenses in a
steganographic structure is reliant upon the utmost payload
capacity, hence, the payload capacity exhibits as the utmost
critical steganographic system feature. Take for example,
the assumption of a scenario where the gray image is
512×512; and that the maximal quantity of bits of this image
that can be hidden is 4,00,000 bits, this implicates the
image’s utmost hiding capacity is 4,00,000 bits or 1.5258
bpp.
B. Distortion Measurement
The S.I imperceptibility is guaranteed by an effective
steganographic procedure; signifying that it is a must that
the final image distortions should not be visible. For
example, the produced S.I by a BIM steganography method
for actual resemblances is indicated in Figure 5, and the S.Is
are indicated in Figure 5b, In these instances, the
S.Iimperceptibility was guaranteed as it is not possible for
any person to identify the the resemblance manipulations.
Any stego resemblances manipulation is only
noticeable/measurable through utilizing certain metrics,
such as the Mean Square Error (MSE), Root Mean Square
Error (RMSE), weighted PSNR, Structural Similarity Index
(SSIM),quality index, correlation, Manhattan Distance,
Kullback-Leibler Divergence (K-L divergence), Euclidean
Distance and Peak Signal-to-Noise Ratio (PSNR).
Fig.5 b. S.Iof a BIM method
Equation 3 is utilized to establish the MSE linking the
actual and the ultimate S.I[20], where

is the element
estimate of the actual resemblance i

row and

is the
element estimate of the S.Iat j

column; m = number of rows
in the digital resemblance and n = number of columns in the
digital resemblance. It is anticipated that the MSE linking
both resemblances ought to be as minimal as feasible; the
MSE is zero when the actual resemblance is similar to the
S.I.
(3)
With regards the RMSE, it normally has an affirmative
value, with an estimate of 0 indicating an ideal data fit
although it is not easy to attain in reality [21]. A further
down RMSE is often preferable to an inflated one. RMSE is
often utilized as an ideal statistical criterion for measuring
image deformation; its computation utilizes equation 4.
2019 IEEE 9th International Conference on System Engineering and Technology (ICSET 2019), 7 October 2019, Shah Alam, Malaysia
978-1-7281-0757-8©2019 IEEE
RMSE =
(4)
Peak Signal-to-Noise-Ratio (PSNR). The cover
resemblance modification that is utilized for bit embedment
is often related with cover image pixel values alterations.
Thus, the modifications must undergo an analysis because
of their direct impact on the resultant S.Iimperceptibility.
The determination of S.Iquality of S.Igenerally utilize
PSNR. Its computation is achieved through the calculation
of the MSE estimate linking both resemblance sets;
signifying that the PSNR gauges the degree of
S.Imodification. The calculation of PSNR utilizes
calculation 5 [20,22], where greater PSNR estimates
implicates a subsidiary modification. The PSNR estimates
of >40dB are generally suitable although estimates
fluctuating from 30 to 40 dB are admissible. Nevertheless,
PSNR estimates of <30 dB are not admissible because of the
inflated distortion level they signify. A pixel of pigment
image entails 3 bytes, that can be singly be accounted as
elements. The MSE and PSNR estimates can be evaluated
through the utilization of calculations 3 and 5.
PSNR = (5)
With WPSNR = Weighted Peak Signal-to-Noise-Ratio
that is alternative metric for gauging the images quality [36-
37]. This matriculation utilizes an alternative estimate
termed as Noise Visibility Function (NVF) that is utilized
for concealing the texture. The calculation of WPSNR
utilizes both NVF and MSE, with NVF estimate being
proximal to 0 and utmost up to 1. The WPSNR is
enumerated through the utilization of calculation 6.
WPSNR = (6)
where :
NVF (i, j) = (7)
With  (,)
= local resemblance variance is
determined by an element with coordinate (i, j); r
(calculated utilizing calculation 6), is a correlation estimate
that gauges resemblance amount for the actual and the stego
resemblances [20]. p = actual images’ average element
estimate, and q = S.I's average element estimate. The
association linking p and q is evaluated by the MATLAB
operation corr2(p,q), with corr2(p, q) possessing an utmost
estimate of 1 if p and q are similar images. Thus, the
correlation is utmost when the deformation is peripheral.
r= ∑∑p

−p∗


(q

−q)
∑∑(p

−p


)
∗(∑∑(q

−q


)
(8)
The determination of the S.Iqualities utilizes the general
image quality index (Q)[23,24]; Q is enumerated through
the utilization of calculation 9, with the utmost element (1)
attainable if p and q are identical resemblances.
Q =
(9)
Where = original images’(O.I) mean element estimate,
and = S.I's mean pixel estimate, in addition to
= O.I's
standard deviation,
= S.I’s standard deviation, and

is
the covariance. They are clarified in equations 10 -14,
accordingly.
= (10)
= (11)
(12)
(13)
(14)
An alternative metric utilized for image quality
measurement first expounded in [25] and implemented in
[26, 27] is the SSIM index. Here, the actual resemblance is
initially segmented to form B blocks of 8×8 elements each,
prior to evaluating the block average element value () and
standard deviation (
) for the actual and the average
element value () and standard deviation (
) for the S.I
The covariance (

) between the two resemblance sets is
also calculated. Finally, calculation 15 is utilized to evaluate
the SSIM, where
= constant for hindering variability when
(

+

) tends to 0. Moreover,
is utilized as the instability
hinderance constant when (
+
) is close to 0; the
estimate of
may be chosen as (K
L)
; for gray
resemblances, with L = 255 and K
≪1. The value of
may comparably be chosen as (K
L)
, where K
≪1. Note
that only when SSIM = Q, then
= 0, exhibiting it a
peculiar case of SSIM.
SSIM=
(
)(


)
(


)(


)
(15)
Evaluation of the SSIM inclusive of every images’ B
blocks takes place before computation of the mean SSIM
index for estimating the resemblances’ universal quality, as
in calculation 16.
MSSIM= (16)
Exploitation of the K-L divergence can be employed for
the determination of the disparity between the histograms of
the cover and the S.I[28 ,25]. Assuming that h1 and h2
separately constitute the cover resemblance and
S.Ihistograms, then, calculation 17 can be utilized in
evaluation of the K-L separation from c1 to c2 (say d1) and
2019 IEEE 9th International Conference on System Engineering and Technology (ICSET 2019), 7 October 2019, Shah Alam, Malaysia
978-1-7281-0757-8©2019 IEEE
from c2 to c1 (for example d2)) as figured by calculation 18
[46, 47]. The mean of d [(d1+d2)/2] is then regarded; where
there is similarity in the image sets and the identical
resemblance, the K-L divergence estimate will be 0.
d1= 1()∗log
()
()


(17)
d2 =2()∗log
()
()


(18)
The determination of the disparity between the actual
and S.I's histograms utilizes the Manhattan distance (MD)
[29]. Presume c1 and c2 as the cover resemblance and S.I's
histograms, accordingly, then, calculation 19 can be utilized
to compute the MD as the total of the complete contrasts of
their affliated constituents.
MD (c1,c2) = |1()−2()|


(19)
An alternative metric for determining the contrasts
linking the actual resemblance and S.I' shistograms is the
Euclidean distance (ED) [30]. The ED is gauged as the
square root of the sum of the squared differences. Presume
h1 and h2 as the cover resemblance and S.I's histograms,
accordingly, then the calculation 20 can be utilized to
compute the ED as the following:
ED (h1,h2) = (20)
A more superior performance is indicated by a
steganographic method in instances of the capability shown
of greater hiding capacity in the cover.
C. Security Evaluation
A method that demonstrates resistance to several
steganalytic attacks is associated with a secure
steganographic method. The assessment of the security of
any steganographic method is implemented through the
utilization of numerous steganalysis plans; for example, RS
investigation can be utilized to evaluate the certainty of an
LSB substitution-based method, meanwhile the certainty of a
PVD-based method can be checked according to the
investigation of the element difference histogram. The pixel
difference histogram(PDH) is a description in graphic form
[31] with the contrasts linking the elements in each set of two
succeeding elements are portrayed in the X-axis, meanwhile
the quantity of incidents is portrayed in the Y-axis. In the
event that the histogram consists of whatsoever steps that is
not desired, the steganography act can be detected. For
example, two varying elements contrast histograms are
indicated in Figure 6 representing two differing methods. In
Figure 6, the curves represented using solid lines exhibits the
element contrast histograms of the actual resemblances in
every graph, meanwhile the plotted lines symbolized by
dotted lines represented the S.I. The step effect in PDH for
Method 1 can obviously be seen, however this cannot be
seen for Method 2; thus, Method 2 is able to withstand PDH
evaluation
.
Fig. 6. RS figures of 2 methods [3]
IV. CONCLUSION
The performance evaluation of any novel steganographic
formula depends on the volume to conceal, certainty and
deformation computation is compulsory. The current
research explicates the entire metrics of the measurement of
distortion through the use of mathematical functions. A pair
of steganalysis tools, which are; RS evaluation and PDH
investigation were besides explained extensively, for the
steganographic algorithms certainty assessment measurement
and others are deliberate, using specifications that anticipate
your paper as one part of the entire proceedings, and not as
an independent document. Please do not revise any of the
current designations.
REFERENCES
[1] Meng, R., Zhou, Z., Cui, Q., Sun, X., & Yuan, C. (2019). A Novel
Steganography Scheme Combining Coverless Information Hiding and
Steganography.
[2] Hashim, M. M., Rahim, M. S. M., Johi, F. A., Taha, M. S., & Hamad,
H. S. (2018). Performance evaluation measurement of image
steganography techniques with analysis of LSB based on variation
image formats. International Journal of Engineering &
Technology, 7(4), 3505-3514.
[3] Ansari, A. S., Mohammadi, M. S., & Parvez, M. T. (2019). A
Comparative Study of Recent Steganography Techniques for Multiple
Image Formats. International Journal of Computer Network and
Information Security, 11(1), 11.
[4] Hussain, M., Wahab, A. W. A., Idris, Y. I. B., Ho, A. T., & Jung, K. H.
(2018). Image steganography in spatial domain: A survey. Signal
Processing: Image Communication, 65, 46-66.
[5] Hou, D., Zhang, W., Yang, Y., & Yu, N. (2018). Reversible data hiding
under inconsistent distortion metrics. IEEE Transactions on Image
Processing, 27(10), 5087-5099.
[6] Fridrich, J., Goljan, M., & Du, R. (2001). Detecting LSB steganography
in color, and gray-scale images. IEEE multimedia, 8(4), 22-28.
[7] Swain, G., & Lenka, S. K. (2015). A novel steganography technique by
mapping words with LSB array. International Journal of Signal and
Imaging Systems Engineering, 8(1-2), 115-122.
[8] Swain, G., & Lenka, S. K. (2011, December). LSB array based image
steganography technique by exploring the four least significant bits.
In International Conference on Computing and Communication
Systems (pp. 479-488). Springer, Berlin, Heidelberg.
[9] Swain, G., & Lenka, S. K. (2012). A technique for secret
communication using a new block cipher with dynamic
steganography. International Journal of Security and Its
Applications, 6(2), 1-12.
[10] Swain, G., & Lenka, S. K. (2012). A robust image steganography
technique using dynamic embedding with two least significant bits.
In Advanced Materials Research (Vol. 403, pp. 835-841). Trans Tech
Publications.
[11] Kadhim, I. J., Premaratne, P., Vial, P. J., & Halloran, B. (2019).
Comprehensive survey of image steganography: Techniques,
Evaluations, and trends in future research. Neurocomputing, 335, 299-
326.
[12] aini, M., & Chhikara, R. (2019). Performance Evaluation of Features
Extracted from DWT Domain. In Software Engineering (pp. 257-265).
Springer, Singapore
2019 IEEE 9th International Conference on System Engineering and Technology (ICSET 2019), 7 October 2019, Shah Alam, Malaysia
978-1-7281-0757-8©2019 IEEE
[13] Murinto, M., Astuti, N. R. D. P., & Mardhia, M. M. (2019). Multilevel
thresholding hyperspectral image segmentation based on independent
component analysis and swarm optimization methods. International
Journal of Advances in Intelligent Informatics, 5(1), 66-75.
[14] Simmons, G. J. (1984). The prisoners’ problem and the subliminal
channel. In Advances in Cryptology (pp. 51-67). Springer, Boston,
MA. https://doi.org/10.1007/978-1-4684-4730-9_5.
[15] Knuth, D. E. (1985). Dynamic huffman coding. Journal of algo-
rithms, 6(2), 163-180. https://doi.org/10.1016/0196-6774(85)90036-7.
[16] T. Sharp, \Hide 2.1, 2001," http://www.sharpthoughts.org
[17] G. Pulcini, \Stegotif,” http:// www.geocities.com/ SiliconValley /9210
/gfree.html
[18] Zhang, Y., Qin, C., Zhang, W., Liu, F., & Luo, X. (2018). On the
fault-tolerant performance for a class of robust image
steganography. Signal Processing, 146, 99-111.
[19] Li, B., Li, Z., Zhou, S., Tan, S., & Zhang, X. (2017). New steganalytic
features for spatial image steganography based on derivative filters and
threshold LBP operator. IEEE Transactions on Information Forensics
and Security, 13(5), 1242-1257.
[20] Swain, Gandharba, and Saroj Kumar Lenka. "Classification of image
steganography techniques in spatial domain: a study." International
Journal of Computer Science & Engineering Technology 5, no. 3
(2014): 219-232.
[21] Chai, Tianfeng, and Roland R. Draxler. "Root mean square error
(RMSE) or mean absolute error (MAE)?–Arguments against avoiding
RMSE in the literature." Geoscientific model development 7, no. 3
(2014): 1247-1250.
[22] Moon, Sunil K., and Rajeshree D. Raut. "Information security model
using data embedding technique for enhancing perceptibility and
robustness." International Journal of Electronic Security and Digital
Forensics 11, no. 1 (2019): 70-95.
[23] Wang, Zhou, and Alan C. Bovik. "A universal image quality
index." IEEE signal processing letters 9, no. 3 (2002): 81-84.
[24] Peksinski, Jakub, Grzegorz Mikolajczak, and Janusz P. Kowalski.
"Recognising the user's face based on algorithms using the analysis of
Q measure indication for their operation." In 2017 40th International
Conference on Telecommunications and Signal Processing (TSP), pp.
573-577. IEEE, 2017.
[25] Wang, Zhou, Alan C. Bovik, Hamid R. Sheikh, and Eero P.
Simoncelli. "Image quality assessment: from error visibility to
structural similarity." IEEE transactions on image processing13, no. 4
(2004): 600-612.
[26] oyal, Megha, Yashpal Lather, and Vivek Lather. "Analytical relation
& comparison of PSNR and SSIM on babbon image and human eye
perception using matlab." International Journal of Advanced Research
in Engineering and Applied Sciences 4, no. 5 (2015): 108-119.
[27] Banerjee, Indradip, Souvik Bhattacharyya, and Gautam Sanyal.
"Hiding & analyzing data in image using extended PMM." Procedia
Technology 10 (2013): 157-166
[28] Joo, Jeong-Chun, Hae-Yeoun Lee, and Heung-Kyu Lee. "Improved
steganographic method preserving pixel-value differencing histogram
with modulus function." EURASIP Journal on Advances in Signal
Processing 2010, no. 1 (2010): 249826.
[29] Chen, Siyi, and Zhiguo Qu. "Novel Quantum Video Steganography
and Authentication Protocol with Large Payload." International
Journal of Theoretical Physics 57, no. 12 (2018): 3689-3701
[30] Hasnat, Abul, Santanu Halder, D. Bhattacharjee, M. Nasipuri, and D.
K. Basu. "Comparative study of distance metrics for finding skin color
similarity of two color facial images." In National Conference on
Advancement of Computing in Engineering Research (ACER 13).
2013.
[31] Taha, Mustafa Sabah, et al. "Wireless body area network
revisited." International Journal of Engineering & Technology 7.4
(2018): 3494-3504.
2019 IEEE 9th International Conference on System Engineering and Technology (ICSET 2019), 7 October 2019, Shah Alam, Malaysia
978-1-7281-0757-8©2019 IEEE
... Authors of [31] defined MSE as a metric that should be able to measure steganography manipulation, which occurred during the embedding procedure on the cover image to produce the stego image. In this study, the BIM steganography technique was used on an image and the following equation (2.4) was utilized: ...
... where m is the number of rows, n is the number of columns, i th row, q ij is the element estimate of the stego image at i th column, and p ij is the element estimate of the actual resemblance at the j th column [31]. ...
... Starting with, the authors of [28] defined the R measurement as the calculation of how closely the cover image and the stego image match one another. This definition was supported in [31], regarding their use of the measurement in 2D ...
Thesis
Full-text available
Nowadays, technology’s effect is evident in all fields of business and daily life. With the surge of information being shared worldwide the need for protection and security has become a real issue that digital rights management (DRM) solves. DRM techniques are used to establish ownership over shared data, protecting the intellectual property of owners. This thesis focuses on a certain DRM technique called steganography which is the safe sharing of said information from unauthorized access or attacks. The number sequence followed by the LSB substitution is the Lucas number sequence, which is responsible for choosing the image pixels. The performance of the proposed scheme is evaluated using the following evaluation metrics: image fidelity (IF), structural similarity index (SSIM), mean square error (MSE), performance evaluation metrics (PSNR), R measurement, normalized cross correlation, (NCC), entropy, and histogram analysis. These metrics are used to compare the proposed scheme with other literature and to evaluate its performance.
... Each primary colour has a total of roughly 256 possible intensities. When the LSB is changed, the intensity of the colours barely changes [30]. As shown in Fig. 2 with the PVD approach are noteworthy in the histogram analysis that indicates the presence of a secret message [31]. ...
... Following some reconstructions to convert the image into the frequency domain, the secret data will be embedded using the frequency domain steganographic approach. Under the umbrella of the frequency domain, variant diversions like Discrete Cosine Transform DCT, and Discrete Fourier Transform DFT are used in [30]. A finite sequence of data points is expressed using a discrete cosine transform (DCT) as the sum of cosine functions fluctuating at various frequencies [32]. ...
Thesis
Full-text available
Since images are tremendously common over the 5G telecommunications network, experiencing safe and secure transmission of such type of data has been the concern for many users for a while. To allow a doctor in the health care sector or an operator on cloud computing services be almost sure that their secret information with highest possibility will not be intruded, this thesis is intended to contribute security solution for that concern. A cryptography layer with logistic tan map encryption will be performed on the image designated to be secret, for example a QR code image containing personal information. Then a steganography layer will be executed with LSB embedding in a conventional image to hide the very beginning secret image. This model was implemented and tested with three different sizes of the image up to 64×64 image size which equates to 4,096 bytes in a 512 × 512 cover image. Low mean squared error (MSE), high peak signal to noise ratio (PSNR) were amongst the numerical results. In addition to that, structural similarity index metric (SSIM), R measurement and others will be discussed and compared with others in the literature to test the efficiency of the model between the image processing community.
... Image steganography can be classified into two classes: Spatial domain and frequency domain. The cover image's whole set of pixels are the direct subject of the spatial domain steganographic structure [12]. ...
... In [12]:= GrouppingImage = Partition [StegoBits,8]; ...
Thesis
Full-text available
Nowadays, the sheer amount of data available at all times online is astounding. The data is usually easily accessible to everyone, including unauthorized parties, leading to security concerns. Proper data security measures must be adopted to prevent data breaches. Steganography is the art of hiding information within an image, video, or audio file while maintaining the file’s original format. This paper proposes a scheme to embed concealed text within images, using pixel value manipulation and encryption techniques. Least Significant Bit (LSB) embedding is used alongside the Polite Number sequence. Comparing the results to common methodologies from the literature, the numerical results were great.
... After carrying out analysis and applying the performance evaluation metrics, where they calculated PSNR and MSE, it was found that this scheme met the capacity, perceptibility, and robustness criteria. The PSNR value also illustrated that the decline in quality between the stego and cover image is not noticeable by the human eye.The authors of[36] conducted a study that presented a vast knowledge of the LSB of image steganography. The authors explained and discussed the performance evaluation metrics that are deployed to analyze and evaluate image steganography; also, they illustrated the mathematical equations for each metric. ...
Thesis
Full-text available
Nowadays, it’s becoming easier and it’s more common for people to steal others’ work and claim it as their own, leaving owners of the work vulnerable to exploitation and theft of their hard work. As a result, digital rights management (DRM) is becoming an important field to discover, more specifically steganography; as it is a method to employ that protects digital content from being stolen, shared, or copied by other people. This thesis proposes a steganographic scheme on 2D images using the Least Significant Bit (LSB) embedding approach that applies the square number sequence. This scheme is evaluated by using a number of performance metrics which are the mean square error (MSE), peak signal to noise ratio (PSNR), entropy, structural similarity index metric (SSIM), R measurement (R), normalized cross correlation (NCC), image fidelity (IF), and histogram analysis. Moreover, a comparison between the proposed scheme and other proposed works from the literature is performed to show that the proposed scheme has good performance that is similar to its counterparts from the literature.
... Steganography is a Greek word that means "covered writing" with "stegano" referring to "cover or concealment" and "graphic" referring to "writing" [3]. When employing images as cover material, you can choose between using the spatial domain or the transform domain [4]. The spatial domain strategies are straightforward to comprehend and apply. ...
Article
Steganography is a vital technique for transferring confidential information via an insecure network. In addition, digital images are used as a cover to communicate sensitive information. The Least Significant Bit (LSB) method is one of the simplest ways to insert secret data into a cover image. In this paper, the secret text is compressed twice by an Arithmetic coding algorithm, and the resulting secret bits are hidden in the cover pixels of the image corresponding to the pixels of each of the following three methods, one of three methods is used in each experiment: The first method, the edges of the image are modified to increase the number of edges, in the second method the lighter-colored regions are selected, and in the third method, the two methods are combined together to increase security and keep the secret message unrecognized. Hiding in each of the previous methods is done by using the LSB technique in the last 2-bit. The correction approach is used to increase the stego image's imperceptibility. The experimental results show that with an average message size of 29.8 kb, the average Peak Signal-to-Noise Ratio (PSNR) for the second proposed (Light regions) method equals 62.76 dB and for the third proposed (Edge and region) method equals 62.72 dB, which is a reasonable result when compared to other steganographic techniques.
... Steganography is a procedure of making the presence of hidden information in a host signal cannot be detected, such as text, video, and images. Steganography algorithms hide information in the spatial domain [1,2,3], the frequency domain [4,5,6] and also hide information using both domains [8,9] to make it more robust against attacks. Digital images are mostly used as host due to the redundancy of data, and the frequency of use on the internet. ...
Conference Paper
The security of data is considered as important issue in any organization. The highly level of security is required to prevent data attacked, damaged or stolen. The online data transfer is rapidly on demand to reduce the costs and time for the organization. This study aims to discuss security approaches that could be adopted in order to secure data transfer online. The literature in the security domain is discussed to determine the most suitable security techniques for the transfer data depends on the data nature, i.e. textual, video, images, or audio data. The discuss shows that techniques such as encrypted data are suitable to safe the textual transfer data. The chaotic map and steganography techniques are suitable to protect the images, video, and audio data. Steganography works to hiding the data in other plain media such as images. The chaotic map works on remapping the data pixel map in order to maximize the complexity of handling the original data by the attackers. It is recommended to offer integration between the security techniques to improve data transfer protection. For example, the data encryption would be supported by other techniques such as a chaotic map, and the steganography can be supported by techniques such as chaotic map. This paper offers important evidences about the security techniques that would be adopted which depends on the nature of data transfer as well as the integration between the techniques to enhance the protection of data transfer. The users or organizations can adopt security techniques according to their data in the environment of working. In the future, the privacy techniques of data transfer would be investigated to prevent the illegal accessing of the data.
Article
Full-text available
Rapid growth of wireless body area networks (WBANs) technology allowed the fast and secured acquisition as well as exchange of vast amount of data information in diversified fields. WBANs intend to simplify and improve the speed, accuracy, and reliability of com-munica-tions from sensors (interior motors) placed on and/or close to the human body, reducing the healthcare cost remarkably. However, the security of sensitive data transfer using WBANs and subsequent protection from adversaries attack is a major issue. Depending on the types of applications, small and high sensitive sensors having several nodes obtained from invasive/non-invasive micro-and nano-technology can be installed on the human body to capture useful information. Lately, the use of micro-electro-mechanical systems (MEMS) and integrated circuits in wireless communications (WCs) became widespread because of their low-power operation, intelligence , accuracy, and miniaturi-zation. IEEE 802.15.6 and 802.15.4j standards have already been set to specifically regulate the medical networks and WBANs. In this view, present communication provides an all-inclusive overview of the past development, recent progress, challenges and future trends of security technology related to WBANs.
Article
Full-text available
Rapid growth of wireless body area networks (WBANs) technology allowed the fast and secured acquisition as well as exchange of vast amount of data information in diversified fields. WBANs intend to simplify and improve the speed, accuracy, and reliability of communica-tions from sensors (interior motors) placed on and/or close to the human body, reducing the healthcare cost remarkably. However, the secu-rity of sensitive data transfer using WBANs and subsequent protection from adversaries attack is a major issue. Depending on the types of applications, small and high sensitive sensors having several nodes obtained from invasive/non-invasive micro- and nanotechnology can be installed on the human body to capture useful information. Lately, the use of micro-electro-mechanical systems (MEMS) and integrated circuits in wireless communications (WCs) became widespread because of their low-power operation, intelligence, accuracy, and miniaturi-zation. IEEE 802.15.6 and 802.15.4j standards have already been set to specifically regulate the medical networks and WBANs. In this view, present communication provides an all-inclusive overview of the past development, recent progress, challenges and future trends of security technology related to WBANs.
Article
Full-text available
High dimensional problems are often encountered in studies related to hyperspectral data. One of the challenges that arise is how to find representations that are accurate so that important structures can be clearly easily. This study aims to process segmentation of hyperspectral image by using swarm optimization techniques. This experiments use Aviris Indian Pines hyperspectral image dataset that consist of 103 bands. The method used for segmentation image is particle swarm optimization (PSO), Darwinian particle swarm optimization (DPSO) and fractional order Darwinian particle swarm optimization (FODPSO). Before process segmentation image, the dimension of the hyperspectral image data set are first reduced by using independent component analysis (ICA) technique to get first independent component. The experimental show that FODPSO method is better than PSO and DPSO, in terms of the average CPU processing time and best fitness value. The PSNR and SSIM values when using FODPSO are better than the other two swarm optimization method. It can be concluded that FODPSO method is better in order to obtain better segmentation results compared to the previous method.
Article
Full-text available
Recently, Steganography is an outstanding research area which used for data protection from unauthorized access. Steganography is defined as the art and science of covert information in plain sight in various media sources such as text, images, audio, video, network channel etc. so, as to not stimulate any suspicion; while steganalysis is the science of attacking the steganographic system to reveal the secret message. This research clarifies the diverse showing the evaluation factors based on image steganographic algorithms. The effectiveness of a steganographic is rated to three main parameters, payload capacity, image quality measure and security measure. This study is focused on image steganographic which is most popular in in steganographic branches. Generally, the Least significant bit is major efficient approach utilized to embed the secret message. In addition, this paper has more detail knowledge based on Least significant bit LSB within various Images formats. All metrics are illustrated in this study with arithmetical equations while some important trends are discussed also at the end of the paper.
Article
Full-text available
Recently, Steganography is an outstanding research area which used for data protection from unauthorized access. Steganography is de-fined as the art and science of covert information in plain sight in various media sources such as text, images, audio, video, network channel etc. so, as to not stimulate any suspicion; while steganalysis is the science of attacking the steganographic system to reveal the secret message. This research clarifies the diverse showing the evaluation factors based on image steganographic algorithms. The effec-tiveness of a ste-ganographic is rated to three main parameters, payload capacity, image quality measure and security measure. This study is focused on im-age steganographic which is most popular in in steganographic branches. Generally, the Least significant bit is major efficient approach utilized to embed the secret message. In addition, this paper has more detail knowledge based on Least signifi-cant bit LSB within various Images formats. All metrics are illustrated in this study with arithmetical equations while some important trends are discussed also at the end of the paper.
Article
Full-text available
This paper proposed a secure authenticated quantum video steganography protocol with large capacity. The new protocol can embed secret quantum information into carrier quantum video, and expand the embedding capacity to a large extent. It also manages to accomplish quantum information steganography process based on unique features of video as well as authentication mechanism for better security. Finally, the simulation experiment proves that the new protocol not only has good performance on imperceptibility and security, but also owns a large capacity.
Article
At present, the coverless information hiding has been developed. However, due to the limited mapping relationship between secret information and feature selection, it is challenging to further enhance the hiding capacity of coverless information hiding. At the same time, the steganography algorithm based on object detection only hides secret information in foreground objects, which contribute to the steganography capacity is reduced. Since object recognition contains multiple objects and location, secret information can be mapped to object categories, the relationship of location and so on. Therefore, this paper proposes a new steganography algorithm based on object detection and relationship mapping, which integrates coverless information hiding and steganography. In this method, the coverless information hiding is realized by mapping the object type, color and secret information in object detection method. At the same time, the object detection method is used to find the safe area to hide secret messages. The proposed algorithm can not only improve the steganographic capacity of the two information hiding methods but also make the coverless information hiding more secure and robust.
Article
Information concealing using steganography is simple, but to maintain its security, perceptibility, robustness, embedding capacity and better recovery of both covers as well as secret data are the major issues. This paper is focused on the improvement in all these major issues. The proposed technique embedded the image and audio as secret data into the randomly selected frames of the video using multi frame exploiting modification direction (MFEMD) algorithm. Hence, it is very difficult to understand in which part of video, data is hidden. At the receiver end, we have used the forensic tool for authentication to improve the data security. Furthermore the obtained simulation results are found to be better than any other existing methods in terms of good visual recovery of both original video and secret data, embedding capacity, security of hidden secret data. Different types of attacks are applied on stego video during transmission like visual, chi-square, histogram, etc. to improve the perceptibility and robustness of secret data.
Article
Storing and communicating secret and/or private information has become part of our daily life whether it is for our employment or personal well-being. Therefore, secure storage and transmission of the secret information have received the undivided attention of many researchers. The techniques for hiding confidential data in inconspicuous digital media such as video, audio, and image are collectively termed as Steganography. Among various media types used, the popularity and availability of digital images are high and in this research work and hence, our focus is on implementing digital image steganography. The main challenge in designing a steganographic system is to maintain a fair trade-off between robustness, security, imperceptibility and higher bit embedding rate. This research article provides a thorough review of existing types of image steganography and the recent contributions in each category in multiple modalities. The article also provides a complete overview of image steganography including general operation, requirements, different aspects, different types and their performance evaluations. Different performance analysis measures for evaluating steganographic system are also discussed here. Moreover, we also discuss the strategy to select different cover media for different applications and a few state-of-the-art steganalysis systems.