A Study on Video Steganographic Techniques

ISSN (e): 2250 – 3005 || Volume, 05 || Issue, 10 ||October – 2015 ||
International Journal of Computational Engineering Research (IJCER)
www.ijceronline.com Open Access Journal Page 30
A Study on Video Steganographic Techniques
Syeda Musfia Nasreen , Gaurav Jalewal, Saurabh Sutradhar
I. Introduction
The premise from which to measure a secure video steganography system is to assume that the opponent knows
the system being employed, yet still cannot find any evidence of the hidden message. Video steganography
algorithm tries to replace the redundant bits of the cover medium by the bits of the secret medium. Now the
availability of those redundant bits to be inserted in the cover media depends on the quality of video or sound.
Military, industrial applications, copyright, intellectual property rights etc. are some of the most commonly used
applications of video steganography.
The advantages of using video stream as the cover file are to get extra security against the attacker because the
video file is much more complex that the image file. One more advantage of embedding the secret data to the
video is that the secret data is not recognized by the human eye as the change of a pixel color is negligible. In
video steganography, we can also very secretly hide data in audio files as it contains unused bits. We can store
secret data up to about four least significant bits in the audio file. So it is more beneficial to use video
steganography rather than other steganography methods when we need to store more amounts of secret data. [1]
1.1. Techniques of video steganography
There are various techniques of video steganography. The best technique is to hide the secret data without
reducing the quality of the cover video, so that it cannot be detected by naked eyes. The embedded video is
known as the “stego” video which is sent to the receiver side by the sender.[2]
Variety of video steganography techniques are used now days, to secure important information. Some much
known techniques are explained briefly in the following:
1.2. LSB (Least Significant Bit) method
LSB is said to be the best method for data protection because of its simplicity and commonly used approach. It
is the most easiest and effective way of embedding data. In LSB, the cover video’s pixel values are extracted
which are in bytes, then its LSB are substituted by the bits of the secret message that we will embed. Now since
we change only the lsb bits of the host video, it doesn’t gets distorted and almost looks alike as the original
video.[3]
Abstract
Data hiding techniques have taken important role with the rapid growth of intensive transfer of
multimedia content and secret communications. The method of Steganography is used to share the
data secretly and securely. It is the science of embedding secret information into the cover media
with the modification to the cover image, which cannot be easily identified by human eyes.
Steganography algorithms can be applied in audio, video and image file. Hiding secret information
in video file is known as video steganography. Video Steganography means hiding a secret message
that can be either a secret text message or an image within a larger one in such a way that just by
looking at it, an unwanted person cannot detect the presence of any hidden message. For hiding
secret information in the video, there are many Steganography techniques which are further
explained in this paper along with some of the research works done in some fields under video
steganography by some authors. The paper describes the progress in the field of video
Steganography and intends to give the comparison between its different uses and techniques.
Keywords: video steganography, LSB method, data hiding, embed, stego video, AVI, PSNR.

A Study On Video Steganographic…
1.3. Non-uniform rectangular partition
This method is for uncompressed videos. In non-uniform rectangular partition, data hiding is done by hiding an
uncompressed secret video file in the host video stream. But we have to make sure that both the secret as well as
the cover file should be of almost the same size. Each of the frames of both the secret as well as cover videos is
applied with image steganography with some technique. The secret video file will be hidden in the leftmost four
least significant bits of the frames of the host video. [3]
1.4. Compressed video steganography
This method is done entirely on the compressed domain. Data can be embedded in the block of I frame with
maximum scene change and in P and B block with maximum magnitude of motion vectors. The AVC encoding
technique yields the maximum compressing efficiency. [3]
1.5. Anti-forensics technique
Anti-forensic techniques are actions taken to destroy, hide and/or manipulate the data to attack the computer
forensics. Anti-forensic provides security by preventing unauthorized access, but can also be used for criminal
use also. Steganography is a kind of anti- forensic where we try to hide data under some host file.
Steganography along with anti- forensics makes the system more secure. [3]
1.6. Masking and filtering
Masking and filtering are used on 24 bits/pixel images and are applicable for both colored and gray scale
images. It is like watermarking over an image and doesn’t affect the quality of that image. Unlike other
steganography techniques, in data masking the secret message is so processed such that it appears similar to a
multimedia file. Data masking cannot easily be detected by traditional steganalysis.[3]
II. Related works
In 2009, Eltahir, L. M. Kiah, and B. B. Zaidan presented a high rate video streaming steganography system
based on least significant bit method.[22] The results of using this method on instant images saves up to 33.3%
of the image for data hiding which is an enhancement for LSB. The idea of the suggested method is by using 3-
3-2 approach, which uses the LSB of RGB (red, blue, green) colors in 24 bits image. The method here takes the
least 3 bits of red color, 3 bits of green color and only 2 bits from blue color because human vision system is
more sensitive to blue than red and green, to come up with 1 byte which is used for data hiding. So to make the
outcome image look almost the same as the original, the 3-3-2 approach is very efficient.
The result was found to be good and the size of data was substantial i.e. about 33.3% from the size of image.
They didn’t found any difference between two frames and their histograms, especially for human vision system.
[4]
In the year 2011 ShengDun Hu, KinTak U presented a video steganography system based on non-uniform
rectangular partition. This technique is used in uncompressed videos. In this method a secret video is hidden in
a cover video, both should be of almost the same size. In each frame of both the videos, a mechanism is applied
for hiding the video stream. The frame length of the cover video should be greater than or equal to the frame
length of the secret video, in order to hide the secret video in to the cover or host video. Each frame of secret
video is portioned in to non-uniform rectangular part which is encoded. The secret video stream is hidden in the
leftmost four least significant bits of each frame of the host video stream.
Results of using this technique showed no distortion, so no one will think that any kind of data is being hidden
in the frames. All the PSNR values of the frames were larger than 28db. [5]
In 2014, R. Shanthakumari and Dr.S. Malliga presented a paper on Video Steganography using LSB matching
revisited algorithm, where they have taken a video stream of AVI format. In the paper they have initially splitted
the cover video into frames. Now, the message can be embedded in multiple frames, therefore size of a message
does not matter in video steganography. After embedding the secret data in multiple frames of the cover video
stream, all the frames are then grouped together to form the stego video, which is then again will be splitted into
frames and data will be extracted in the receiver side.
The proposed method in this paper was found to have two problems which were low embedding rate and lack of
security. LSBMR algorithm has a low replacement rate and hence the Mean Square Error (MSE) is low, as a
result of which LSBMR is more secured than the LSB algorithm for data hiding. The PSNR value decreases on
increase of the embedding unit.[6]

In 2015, Vivek Kapoor and Akbar Mirza presented a paper on Enhanced LSB based Video Steganographic
System for Secure and Efficient Data Transmission. At first, they have divided the host video into frames, then
the text file to be embedded is compressed using the ZIP compressor and the bytes are generated. The advantage
of sending compressed data over the network is to minimize the payload and reduce extra burden of the
network. Then the color of each color pixels is calculated in RGB 24 bit format. Then chunks of bits are created
from the extracted bytes of the secret message. Then these chunks are embedded in the video frames. Text files
are then combined with video frames and the file is sent. The proposed method is designed for MPEG format;
however it can work with other video file formats also like AVI, 3GP by doing some modification in it. They
have calculated the Data Quality, Mean Squared Error(MSE) and Peak Signal to noise ratio(PSNR) of both the
original and the stego video and it was found that Mean Square Error and Peak Signal to Noise Ratio is low
enough that it cannot be noticed easily in the Steganalysis process. Then they have also compared their method
with LSB method and they found that the proposed method gives better MSE and PSNR values than the LSB
method. [7]
In 2013, Hemant Gupta, Dr. Setu Chaturvedi presented a video steganography through LSB based hybrid
approach. This method is used in AVI videos. The video is converted into 20 equal gray scale images. Data
hiding is done is done in the host video by using Single bit, two bit, three bit LSB substitution and after that
Advanced Encryption Standard Algorithm is applied. After processing the source video by using the data hiding
procedures, the encrypted AVI Video is sent by the sender and decryption is performed by the receiver. They
have found the PSNR and correlation factor between Original and embedded image for 1 bit LSB & 2 bit LSB
& 3 bit LSB Substitution and AES method. It is observed that PSNR value decreases and security increases with
the increase of LSB substitution bit. In this paper they have found no correlation relation between original image
and encrypted image for different frames.[8]
In 2013, Pooja Yadav, Nischol Mishra and Sanjeev Sarma presented a video steganography technique with
encryption and LSB substitution. In their technique, they had 2 video streams called Host and the Secret video
with same number of frames and equal frames per second (12 frames and 15 fps). A header of 8 bits is used for
representing the frame size of the hidden video and was appended in the beginning of each frame. After this the
appended header was encrypted along with the secret frames by using symmetric encryption. Using sequential
encoding, the secret video frames are encoded to the host video frames and then from the encoded frames the
secret video is generated. They used XOR transformation for encrypting the data with secret keys and
decrypting the secret message to retrieve the original information. For sequential encoding they used a pattern of
BGRRGBGR (Blue (B), green (G) and red (R)) to encode the message in the LSB. They used 2 AVI video files
and found the PSNR value of the stego video as 35 dB which was the same as the host video.
To find the result they have calculated the PSNR value frame by frame and maximum, minimum and average
PSNR value of total frames. Frame by frame comparison of the host and the embedded video stream shows that
there was no distortion in the stego video. According to the calculated PSNR values, it was observed that there
was much similarity between the host and the embedded video. The host video was found to be distortion less
and also the recovered video stream had also an acceptable quality. [9]
Ramadhan J. Mstafa and Khaled M. Elleithy, Senior Member, IEEE, Department of Computer Science and
Engineering, University of Bridgepor, proposed a highly secured method of video steganography by using
Hamming Code (7, 4). In their project, they used 9 video files as cover and 1 secret image which were to be
hidden. At first, they generated the frames from the video stream and then separated each frames into Y, U & V
components. Then by the use of a special key, all pixel positions of video are randomly ordered. A binary image
is used as the message which was converted to a 1-dimensional array and the position of the message is changed
by a key. Now, 4 bits of the message is encoded using Hamming Code (7, 4) encoding technique. Now, the
encoded data is XORed with the random values and the result is embedded in 1 pixel of Y, U and V
components. The pixels are then reordered in their original position and the final stego video was rebuilt from
the embedded frames. Similar steps are involved in the data extraction process.
The stego video has mostly the same quality as the original video because of the low modification on the host
video stream. The visual quality is measured by the PSNR and all the obtained experimental results have a
PSNR above 51 dBs. Using this method, attackers are not likely to be suspicious since they have a good visual
quality for stego videos. The algorithm is much secured because security has been satisfied by having more than
one key to embed and extract the secret message. [10]
In 2008, Bin Liu et al proposed a new steganography algorithm for compressed video Bit streams. In this
technique, the embedding and detecting data is done only in compression domain and no decompression is
needed here. The cover video was first compressed by eliminating temporal, spatial and statistical redundancies.

The video was then divided into several slow speed and single scene video sub-sequences. After the scene
detection process, they embedded the secret message in the video file without any distortion. Finally the
embedded video was tested for Steganalysis to check the presence of hidden data in the video. They constantly
adjusted the scale factor for manipulating the hidden data strength until the analyzer was unable to detect the
hidden message.
The PSNR value and the correlation value changes the magnitude of the stego-video, which says that the
perception quality and intraframe correlation of test video is little changed. There is no noticeable change in the
visual quality of the compressed video, also their system was found to highly secured as they were continuously
testing the video for Steganalysis.[11]
III. Comparative Analysis
This paper presented a background of Steganography and a comparative study of some Steganographic
techniques. There are two important parameters of evaluating all Steganography technique, first is
imperceptibility and the second is capacity. Imperceptibility means the embedded data must be imperceptible to
the observer and computer analysis. Capacity means maximum payload is required, i.e. maximum amount of
data that can be embedded into the cover image without losing the fidelity of the original image. The results of
surveying the papers in different techniques of video steganography showed that all the methods possess the
ability to hide data without noticing changes in their properties.
It was found that in [4], they have used the 3-3-2 approach along with LSB and the result was found to be good
and about 33.3% from the size of image can be used for data hiding. In other words, in the space of 5 images,
500 pages of data could be stored without resizing. Similarly, 1 second in certain video types contains
approximately 27 frames, which in turn creates a lot of room for hiding data. In [5], Results of using this
technique showed no visual distortion in the host file and even the quality of the new video generated can be
accepted for practical use. In [6], it has been known that in the LSB algorithm due to high replacement rate,
MSE value is high. So it lacks from security. In case of LSBMR algorithm due to low replacement rate, MSE
value is low which makes it secure when compared to LSB algorithm. In their method, intruder may not be able
to identify the presence of the secret message inside the frame. Also, the comparison with the original video
never gives the original secret message, which ensures additional security.
In [7], videos of different sizes and resolutions are tested for their method and they have got successful in
keeping the MSE and PSNR value low enough that it cannot be noticed easily in the Steganalysis process. They
have provided a comparison between the basic LSB method and their method gave better values of MSE and
PSNR than the LSB method. The average PSNR of the proposed LSB embedding technique (per pixel, RGB) to
the traditional layering technique in which embedding is done by layers of RGB. They found an improvement of
about 1.5 dB in their PSNR value when compared to the traditional LSB technique and also a lesser MSE which
means in detectability. In [8], the authors calculated PSNR value for different amount of LSB substitution. For 1
LSB substitution, the PSNR value was found between 45-50 for different no of frames. For 2 bit LSB
substitution, PSNR was found to be in the range of 40-45. And for 3 bit LSB substitution, PSNR value was
about 35. By the use of AES encryption, their method was more secured as compared to traditional LSB
techniques. In [9], their results showed that no visual distortion is there in the host video stream and even the
quality of the recovered secret video is also acceptable in practical. In [10], use of Hamming code makes the
technique highly efficient and more secured. The authors used more than 1 key and thus have obtained a high
level of security as compared to traditional steganographic methods like LSB substitution where only one XOR
encryption is used. In [11], unlike other steganographic technique, the authors have implemented a closed loop
feedback steganalysis to test their project’s immunity towards steganalysis. The complete project was done in
compressed domain hence avoiding decompression process.
With continuous advancements in technology it is expected that in the near future more efficient and advanced
techniques in steganalysis will emerge that will help law enforcement to better detect illicit materials transmitted
through the Internet.
IV. Conclusion
In the era of fast information interchange using internet and World Wide Web, video Steganography has become
essential tool for information security. This paper gave an overview of different video steganographic
techniques its major types and classification of steganography which have been proposed in the literature during
last few years.

References
[1] https://courses.cs.washington.edu/courses/csep590/06wi/finalprojects/chakraborty.doc
[2] Arup Kumar Bhaumik, Minkya choi, "Data hiding in video" IEEE International journal of database application,vol.2no.2 June
2009.pp.9-15
[3] https://edupediapublications.org/journals/index.php/ijr/article/view/678/309
[4] M. E. Eltahir, L. M. Kiah, and B. B. Zaidan, "High Rate Video Streaming Steganography," in Information Management and
Engineering, 2009. ICIME '09. International Conference on, 2009, pp. 550-553.
[5] ShengDun Hu, KinTak U," A Novel Video Steganography based on Non-uniform Rectangular Partition ",IEEE International
Conference on Computational Science and Engineering,pp 57-61, Aug.2011.
[6] R. Shanthakumari and Dr.S. Malliga,” Video Steganography using LSB matching revisited algorithm”, IOSR Journal of Computer
Engineering ,Volume 16, Issue 6, Ver. IV(Nov – Dec. 2014), PP 01-06
[7] Vivek Kapoor and Akbar Mirza, “An Enhanced LSB based Video Steganographic System for Secure and Efficient Data
Transmission”, International Journal of Computer Applications (0975 – 8887) Volume 121 – No.10, July 2015
[8] Hemant Gupta and Dr. Setu Chaturvedi,”video steganography through LSB based hybrid approach”, International Journal of
Engineering Research and Development, Volume 6, Issue 12 (May 2013), PP. 32-42
[9] Pooja Yadav, Nischol Mishra and Sanjeev Sarma, ”video steganography technique with encryption and LSB substitution”, 2013,
School Of Information Technology, Rajiv Gandhi Proudyogiki Vishwavidyalaya, Bhopal, India
[10] Ramadhan J. Mstafa and Khaled M. Elleithy, Senior Member, IEEE, Department of Computer Science and Engineering University of
Bridgeport Bridgeport, CT 06604, USA, “A Highly Secure Video Steganography using Hamming Code (7, 4) “
[11] Bin Liu, Fenlin Liu, Chunfang Yang and Yifeng Sun , “Secure Steganography in Compressed Video Bitstreams”,The Third
International Conference on Availability, Reliability and Security

A Study on Video Steganographic Techniques

More Related Content

What's hot (20)

Viewers also liked (15)

Similar to A Study on Video Steganographic Techniques (20)

Recently uploaded (20)

A Study on Video Steganographic Techniques