Content uploaded by Ramadhan J. Mstafa
Author content
All content in this area was uploaded by Ramadhan J. Mstafa on Aug 11, 2016
Content may be subject to copyright.
Content uploaded by Khaled Elleithy
Author content
All content in this area was uploaded by Khaled Elleithy on May 17, 2015
Content may be subject to copyright.
2015 ASEE Northeast Section Conference
1
© American Society for Engineering Education, 2015
An Efficient Video Steganography Algorithm Based on BCH Codes
Ramadhan J. Mstafa and Khaled M. Elleithy
Department of Computer Science and Engineering
University of Bridgeport
Bridgeport, CT 06604, USA
rmstafa@my.bridgeport.edu elleithy@bridgeport.edu
Abstract
In this paper, in order to improve the security and efficiency of the steganography algorithm, we
propose an efficient video steganography algorithm based on the binary BCH codes. First the
pixels’ positions of the video frames’ components are randomly permuted by using a private key.
Moreover, the bits’ positions of the secret message are also permuted using the same private key.
Then, the secret message is encoded by applying BCH codes (n, k, t), and XORed with random
numbers before the embedding process in order to protect the message from being read. The
selected embedding area in each Y, U, and V frame components is randomly chosen, and will
differ from frame to frame. The embedding process is achieved by hiding each of the encoded
blocks into the 3-2-2 least significant bit (LSB) of the selected YUV pixels. Experimental results
have demonstrated that the proposed algorithm have a high embedding efficiency, high
embedding payload, and resistant against hackers.
Keywords
Video Steganography, BCH Codes, Linear Block Code, Embedding Efficiency, Embedding
Payload.
Introduction
Due to technological advances and the speed of the Internet, people are concerned that their
personal information will be stolen by hackers. In today’s society, many data hiding algorithms
and steganographic algorithms have been introduced in order to protect valuable information.
Steganography is one of the methods that protects and hides valuable data from unauthorized
people without hackers having any suspicion of the data’s existence. The Human Visual System
(HVS) cannot recognize a slight change that occurs in the cover data such as audio, image and
video1,2. Unfortunately, many strong steganography analyzing tools have been provided to
unauthorized users in order for them to retrieve valuable secret data previously embedded in
cover objects. The weakness of some steganography algorithms occur through steganalytical
detectors because of the lack of security and embedding efficiency in these algorithms.
The embedding efficiency and the embedding payload are two important factors that every
successful steganography system should take into consideration3. First, the steganography
scheme that has a high embedding efficiency translates to a good visual quality of stego data and
a less amount of host (carrier) data are going to be changed4. Any obvious distortion to the
viewers will increase the probability of the attacker's suspicion, and the secret information can be
easily detected by some of the steganalysis tools5. These kinds of schemes are difficult to detect
2015 ASEE Northeast Section Conference
2
© American Society for Engineering Education, 2015
by the steganalytical detectors. The security of the steganography scheme depends directly on the
embedding efficiency6. Second, a high embedding payload means that the capacity of secret
information to be hidden inside host data is large. These two factors, embedding efficiency and
embedding payload, contradict one another. Once, the data embedding efficiency is increased,
the data embedding payload is decreased. These two factors will change depending on the users’
requirements and the type of steganography scheme4,7. The remainder of the paper is organized
as follows: Section 2 presents some related work. Section 3 introduces an overview of the Linear
Block Code and BCH codes, and then presents the proposed steganography algorithm. Section 4
presents experimental results and discussion. Section 5 provides the conclusion.
Related Work
Feng et al. proposed a novel of a video steganography scheme based on motion vectors as carrier
data in order to embed the secret message in H.264 video compression processing. The algorithm
also uses the principle of linear block codes to reduce motion vectors’ modification rate. The
algorithm has a good visual quality of stego data, which is proved by the low modification rate of
motion vectors. The Peak Signal to Noise Ratio that was obtained in both Flower and Foreman
videos are more than 37 dB8. Hao et al. proposed a novel video steganography method based on
a motion vector by using matrix encoding. A motion vector component that has high amplitude
among both horizontal and vertical components is chosen to embed the secret message. The HVS
can identify the change that occurs when the object is moving slowly. However, if the same
object moves quickly, the HVS will not be able to recognize the change that occurs. Motion
vectors with large sizes are selected for embedding the secret messages. The macro blocks that
are moving quickly will generate motion vectors with large amplitudes. The direction of macro
blocks depends on the motion vectors’ components. For example, if the vertical component is
equal to zero, then the macro block direction is moving vertically. The visual quality of the tested
videos that were obtained is more than 36 dB9. Rongyue et al. proposed an efficient BCH coding
for steganography which is embedding the secret information inside a block of cover data by
changing some coefficients. Authors have improved the computational time of the system and
the complexity becomes low because of the system’s linearity10. Liu et al. proposed a robust
data hiding scheme in an H.264 compressed video stream, where they have prevented a drift of
intra-frame distortion. To give the system more robustness, the authors have encoded the
message using BCH codes before the embedding process. The host data is the DCT coefficients
of the luminance Intra frame component. The obtained results have a high visual quality and
robustness against hackers11.
The Proposed Steganography Algorithm
The proposed algorithm uses an uncompressed video stream which is based on the frames as still
images. The video sequences are divided into frames, and each frame’s color space is converted
to YCbCr. The reason for using YCbCr color space is to remove the correlation between Red,
Green, and Blue colors. The luminance (Y) component represents the brightness data, which the
human eye is more sensitive to than the other color components. As a result, the chrominance
(CbCr) components can be subsampled in the video stream and some information might be
discarded.
2015 ASEE Northeast Section Conference
3
© American Society for Engineering Education, 2015
A. Linear block codes
Any specific block code is defined as a linear block code if the sum of any two codewords equals
a new codeword. Furthermore, a binary linear block code includes a linear block code that
contains a block of binary bits. A binary linear block code (n, k) consists of columns and
rows in a linear code array. This code is an n-dimensional vector space and a k-
dimensional subspace. N refers to the length of the
codeword and K refers to the number of symbols in each codeword. In the standard array, it is
not possible for two equal vectors to exist in the same row. For example, if C is a (n, k) code on
the Galois Field GF (2), then:
Each coset has vectors.
All F vectors of length n belong to coset of C.
If C+Z is a coset of C and F as belonging to (C+Z), then C+Z=C+F.
B. BCH codes (7, 4, 1)
Bose, Chaudhuri, and Hocquenghem invented the BCH encoder. It is one of the most powerful
random cyclic code methods, which can be used for detecting and correcting errors in a block of
data. The BCH codes are different from the Hamming codes because BCH can correct more than
one bit. A binary BCH (n, k, t) can correct errors of a maximum t bits for codewords of the
length n and message length k. Encoded codewords and
messages can both be interpreted as polynomials, where
and. When m and t are any positive integers where
and, there will be a binary BCH codes with the following properties:
Block codeword length
Message length k
Maximum correctable error bits t
Minimum distance
Parity check bits
The BCH codes inventors decided that the generator polynomial g(x) will be the polynomial of
the lowest degree in the Galois field GF (2), with as roots on the condition that
is a primitive of When is a minimal polynomial of where ,
then the least common multiple (LCM) of 2t minimal polynomials will be the generator
polynomial g(x). In this paper, the BCH codes (7, 4, 1) is used. The parity-check matrix H of the
BCH codes12,13 is described as follows:
(1)
2015 ASEE Northeast Section Conference
4
© American Society for Engineering Education, 2015
C. Data embedding stage
Data embedding is the process of hiding a secret message inside cover videos. This process
converts the video stream into frames. Each frame separates into the Y, U and V components of
color space. For security purposes, the pixels’ positions of Y, U, and V components are permuted
by using a special key (Key1). Also, characters of the secret message are converted into an array
of binary bits. In order to change the bits’ positions of the secret message, the entire bits’
positions within the array are permuted using Key1. After permutation, the array is divided into
4-bit blocks. Then, each block is encoded by the BCH (7, 4, 1) encoder. The outcome of the 7-bit
encoded block (consists of 4-bit message and 3-bit parity) is XORed with the 7-bit number.
These numbers are randomly generated by using Key2. In order to select the locations for hiding
the secret message into the frame components, Key2 is utilized. In other words, Key2 chooses
random rows and columns for data embedding in each disordered Y, U, and V component. The
embedding process is achieved by hiding each of encoded blocks into the 3-2-2 LSB of the
selected YUV pixels. The pixels of the YUV components will be repositioned in order to the
original frame pixel positions to produce the stego frames. Finally, the stego video is constructed
from these stego frames. The block diagrams of the data embedding stage and the data extracting
stage are illustrated in Fig. 1 and Fig. 2, respectively.
Fig. 1: Block diagram for data embedding stage.
2015 ASEE Northeast Section Conference
5
© American Society for Engineering Education, 2015
D. Data extracting stage
Data extracting is the process of retrieving the secret message from the stego videos. This
process is achieved by converting the distorted videos into frames. Then, each frame is
partitioned into Y, U and V components. In every Y, U, and V component, the pixels’ positions
are permuted by using Key1. The process of extracting the secret message from YUV
components is accomplished by taking out 3-2-2 LSB in each selected pixel. The obtained
message block will be XORed with the 7-bit number that is generated by using Key2. The
outcomes of 7 bits are decoded by the BCH (7, 4, 1) decoder in order to produce 4-bit blocks.
These blocks are stored into a binary array. Since the message of entire bits is permutated prior
to the data embedding process, the permutation process of the entire binary array to the original
bits order will be performed again. Then, the binary array of bits will be converted into the
characters of the secret message. The purpose of using two keys and the XOR operation is to
improve the security and robustness of the proposed algorithm. These keys are only shared
between sender and receiver, and used in both the data embedding and extracting processes.
Fig. 2: Block diagram for data extracting stage.
2015 ASEE Northeast Section Conference
6
© American Society for Engineering Education, 2015
Experimental Results and Discussion
A database of eight standard of Common Interchange Format (CIF) video sequences is used,
with the video resolutions equaled to (288 x 352), and the format represented by 4:2:0 YUV.
Video sequences are equal in length to 300 frames. A text file consisting of alphabet characters is
used as a secret message. This work is implemented using MATLAB program to test the
proposed algorithm’s performance. The Peak Signal to Noise Ratio (PSNR) is a visual quality
measurement which is used to compute the difference between the original and the stego video
frames. PSNR is calculated by the following equation:
And Mean Square Error (MSE) is calculated as follows:
Where O and S denote the original and stego YUV frame components, respectively, and m and n
are the video resolutions.
Fig. 3 illustrates the PSNRs of 300 stego frames for the Hall video. In Fig. 4, the PSNRs of 300
Stefan stego video frames are shown. Fig. 5 illustrates the PSNRs of 300 stego frames for the
Foreman video. By using our proposed algorithm, the obtained visual quality is similar to the
original videos’ visual quality. In general, PSNRs are greater than 55 dB, and the V component
has a better visual quality among the three components. The reason of V component has a better
visual quality among other components is because V component has a longest wavelength.
Fig. 3: PSNR of 300 stego frames for the Hall video.
2015 ASEE Northeast Section Conference
7
© American Society for Engineering Education, 2015
In Table 1, the PSNR for eight video sequences is shown for each Y, U, and V component,
separately. The visual quality of the stego videos is the same as the original videos’ visual
quality because all PSNR values are greater than 55 dB.
TABLE 1 THE AVERAGES OF PSNRY, PSNRU, AND PSNRV FOR
ALL VIDEOS
Sequences
Frame No.
PSNRY
PSNRU
PSNRV
Hall
1-100
55.401
56.34
56.52
101-200
55.357
56.324
56.649
201-300
55.317
56.459
56.638
Stefan
1-100
55.381
56.388
57.031
101-200
55.359
56.378
57.058
201-300
55.359
56.343
57.028
Coastguard
1-100
55.335
56.064
56.454
101-200
55.324
56.082
56.406
201-300
55.323
56.055
56.395
Mobile
1-100
55.321
56.546
56.667
101-200
55.306
56.517
56.608
201-300
55.285
56.456
56.57
Foreman
1-100
55.287
56.484
56.621
101-200
55.297
56.479
56.616
201-300
55.285
56.463
56.605
Container
1-100
55.362
56.527
56.665
101-200
55.329
56.476
56.674
201-300
55.334
56.464
56.672
News
1-100
55.527
56.567
56.381
101-200
55.512
56.539
56.369
201-300
55.498
56.535
56.364
Akiyo
1-100
55.385
56.482
56.561
101-200
55.31
56.511
56.564
201-300
55.348
56.481
56.504
2015 ASEE Northeast Section Conference
8
© American Society for Engineering Education, 2015
Fig. 4: PSNR of 300 stego frames for the Stefan video.
Fig. 5: PSNR of 300 stego frames for the Foreman video.
Fig. 6 shows the comparison of the visual quality between eight stego videos. The PSNR of each
component, Y, U, and V is separately calculated, in which the average equals 300 frames per
video. The values of PSNRs are between 55 and 57 dBs, which are considered excellent visual
quality results.
2015 ASEE Northeast Section Conference
9
© American Society for Engineering Education, 2015
Fig. 6: Comparison between the averages of the PSNRY, PSNRU, and PSNRV components for
eight video sequences
Conclusion
In this paper, an efficient video steganography based on the BCH codes concepts has been
proposed. The proposed steganography algorithm utilized frames as still images. It divides the
video stream into frames, and then converts the frames to the YUV format. This algorithm is
considered a high embedding efficiency algorithm due to the low modification on the cover data
that translates into perfect visual quality in the stego videos. The visual quality is measured by
the PSNR metric, and all the obtained experimental results have a PSNR above 55 dB. By
achieving a good visual quality for stego videos, hackers will have difficulty retrieving secret
messages. The security of our proposed algorithm has been satisfied by having more than one
key to embed and extract the secret message. In addition to the secret keys that we have used, we
also encoded and decoded the message with the BCH codes (7, 4, 1).
Experimental results prove both a high embedding efficiency and a high embedding payload of
the proposed algorithm exist. The visual qualities of the stego videos are the same as the original
video visual qualities. The PSNR of stego videos is above 55 dB. In each video frame, the
embedding capacity is 246 Kbits, and can increase up to 405 Kbits without any noticeable
degradation in the visual quality.
References
1. Yuh-Ming, H. and J. Pei-Wun. Two improved data hiding schemes. in Image and Signal Processing
(CISP), 2011 4th International Congress on. 2011.
2. Mstafa, R.J. and C. Bach. Information Hiding in Images Using Steganography Techniques. in American
Society for Engineering Education (ASEE Zone 1), 2013 Zone 1 Conference. 2013.
3. Mstafa, R.J. and K.M. Elleithy. A highly secure video steganography using Hamming code (7, 4). in
Systems, Applications and Technology Conference (LISAT), 2014 IEEE Long Island. 2014.
2015 ASEE Northeast Section Conference
10
© American Society for Engineering Education, 2015
4. Chin-Chen, C., T.D. Kieu, and C. Yung-Chen. A High Payload Steganographic Scheme Based on (7, 4)
Hamming Code for Digital Images. in Electronic Commerce and Security, 2008 International Symposium
on. 2008.
5. Guangjie, L., et al. An Adaptive Matrix Embedding for Image Steganography. in Multimedia Information
Networking and Security (MINES), 2011 Third International Conference on. 2011.
6. Jyun-Jie, W., et al. An embedding strategy for large payload using convolutional embedding codes. in ITS
Telecommunications (ITST), 2012 12th International Conference on. 2012.
7. Mstafa, R.J. and K.M. Elleithy. A Novel Video Steganography Algorithm in the Wavelet Domain Based on
the KLT Tracking Algorithm and BCH Codes. in Systems, Applications and Technology Conference
(LISAT), 2015 IEEE Long Island. 2015.
8. Feng, P., et al. Video steganography using motion vector and linear block codes. in Software Engineering
and Service Sciences (ICSESS), 2010 IEEE International Conference on. 2010.
9. Hao, B., L.-Y. Zhao, and W.-D. Zhong. A novel steganography algorithm based on motion vector and
matrix encoding. in Communication Software and Networks (ICCSN), 2011 IEEE 3rd International
Conference on. 2011.
10. Rongyue, Z., et al., An Efficient Embedder for BCH Coding for Steganography. Information Theory, IEEE
Transactions on, 2012. 58(12): p. 7272-7279.
11. Liu, Y., et al., A Robust Data Hiding Algorithm for H. 264/AVC Video Streams. Journal of Systems and
Software, 2013.
12. Hoyoung, Y., et al., Area-Efficient Multimode Encoding Architecture for Long BCH Codes. Circuits and
Systems II: Express Briefs, IEEE Transactions on, 2013. 60(12): p. 872-876.
13. Mstafa, R.J. and K.M. Elleithy. A High Payload Video Steganography Algorithm in DWT Domain Based
on BCH Codes (15, 11). in Wireless Telecommunications Symposium (WTS), 2015. 2015.
Ramadhan J. Mstafa
Ramadhan J. Mstafa is originally from Duhok, Kurdistan Region, Iraq. He is pursuing his PhD
degree in Computer Science and Engineering at the University of Bridgeport, Bridgeport,
Connecticut, USA. He received his Bachelor’s degree in Computer Science from the University
of Salahaddin, Erbil, Iraq. Mr. Mstafa received his Master’s degree in Computer Science from
University of Duhok, Duhok, Iraq. He is IEEE Student Member. His research areas of interest
include image processing, mobile communication, security, and steganography.
Prof. Khaled M. Elleithy
Dr. Elleithy is the Associate Vice President of Graduate Studies and Research at the University
of Bridgeport. He is a professor of Computer Science and Engineering. He has research interests
are in the areas of wireless sensor networks, mobile communications, network security, quantum
computing, and formal approaches for design and verification. He has published more than three
hundred research papers in international journals and conferences in his areas of expertise. Dr.
Elleithy has more than 25 years of teaching experience. His teaching evaluations are
distinguished in all the universities he joined. He supervised hundreds of senior projects, MS
theses and Ph.D. dissertations. He supervised several Ph.D. students. He developed and
introduced many new undergraduate/graduate courses. He also developed new teaching /
research laboratories in his area of expertise. Dr. Elleithy is the editor or co-editor for 12 books
by Springer. He is a member of technical program committees of many international conferences
as recognition of his research qualifications. He served as a guest editor for several International
Journals. He was the chairman for the International Conference on Industrial Electronics,
Technology & Automation, IETA 2001, 19-21 December 2001, Cairo – Egypt. Also, he is the
General Chair of the 2005-2013 International Joint Conferences on Computer, Information, and
Systems Sciences, and Engineering virtual conferences.
