International Journal of Computer Science Engineering

and Information Technology Research (IJCSEITR)

ISSN(P): 2249-6831; ISSN(E): 2249-7943
Vol. 4, Issue 6, Dec 2014, 23-34
TJPRC Pvt. Ltd.

HIGHLY SECURE IMAGES STEGANOGRAPHY TECHNIQUES BASED ON LSB, X-BOX

MAPPING AND HUFFMAN ENCODING
KANZARIYA NITIN K1, NIMAVAT ASHISH V2 & JADEJA VIJAYSINH K3
1,3

Assistant Professor, Department of Information Technology, C. U. Shah College of Engineering & Technology,
Wadhwan City, India
2

Assistant Professor, Department of Computer Science, C. U. Shah College of Engneering & Technology,
Wadhwan City, India

ABSTRACT
In advanced correspondence, everyone needs high insurance from unapproved client. For information security,
numerous information concealing strategies are accessible that fundamental part is to secure our private/important data
from unapproved client. Steganography is an information concealing procedure that conceals data in such a route, to the
point that nobody can without much of a stretch discover that presence of the data stowed away. LSB is a spatial Domain
method. This paper presents a novel image Steganography method using X-Box Mapping and Huffman Encoding. Here
two 8 bit gray scale images used of different sizes are utilized as cover image and secret image separately. Basically secret
image is not directly inserted into cover image; first to applied Huffman Encoding on the secret image to increase security.
Here we have utilized a few remarkable X-boxes with sixteen separate values. In this calculation, we have utilized four
exceptional X-boxes with sixteen separate values and each one worth is mapped to the four bits of Huffman code stream of
secret image. At that point utilizing LSB replacement, mapped bits are reinstated with Cover Image. The trial consequence
indicates that the calculation has a high installing limit and great imperceptibility. These techniques to provide improve
PSNR of stego image compare to other existing Steganography approaches. Additionally, these methods provide high
Security so, no one can easily extract secret message from stego image without knowing this method and Huffman Table.

KEYWORDS: Steganography, LSB, X Box, Huffman Encoding

INTRODUCTION
In computerized communication, everybody can undoubtedly exchange advanced information starting with one
side then onto the next. In any case numerous unapproved clients who are viewing you and them generally attempt to get to
your private information to abuse [1] [2] [3]. So at whatever point you exchange your private information (messages) to
your nearby companions, or business mysteries with your accomplices, you must take care about your secret data. Be that
as it may we live in a frail world where unwanted gathering can get to your individual data (like messages or particular
reports) and regularly utilize it. [1][2] So we oblige high security of private information. Today such a large number of data
security procedures accessible like cryptography, Steganography and so on.
The word Steganography is a Greek Word and consists of two words such as stego means covered (hidden) and
graphy means writing or drawing. Consequently, Importance of Steganography is concealed (cover) written work. Secret
data is implanted in such a path, to the point that the presence of the Secret data is hidden. Steganography technique has
recently become significant in number of applications such as, Digital audio, video, and images [3][4][5][6].
www.tjprc.org

editor@tjprc.org

24

Kanzariya Nitin K, Nimavat Ashish V & Jadeja Vijaysinh K

Cryptography and Steganography both are information security methods that ensure Secret data from unapproved
access. The fundamental reason for cryptography and Steganography is to give secret and covert communication [1][2][9].
Anyhow, there is distinction between them. Cryptography gives Secret communication yet it is noticeable to other.
Cryptography changes over the substance of a Secret message starting with one structure then onto the next structure which
is not effortlessly seen by any unapproved access. Steganography gives hidden communication that means it hides the
presence of secret information from unapproved access. In cryptography, attacker can see the Secret data. While
Steganography it requires attacker to locate that Steganography has been utilized and inserted message could be extracted.
[1][2][12][13].
To increase the security of information, we can utilize mixture system that is combination of cryptography with
Steganography. To begin with message might be encrypted and after that could be made hidden utilizing Steganography.
Hybrid method increases the security of secret message. Even if an enemy extract the hidden message from the stegoobject, he would still require the cryptographic decipher (key) to get original secret message [7] [13].
Steganography have numerous diverse sorts that could be arranged relying on the cover medium utilized [1][2][4].
Many advanced media could be utilized as a cover medium, for example such as text, image or sound or video. So
Steganography may be defining three different areas:

Text Steganography

Image Steganography

Audio/video Steganography
Out of all image Steganography is most well known strategy on the grounds that images are the most famous

cover items utilized as a part of advanced communication. Second important point is that numerous type of image format
exist, for example, jpg, png, bmp, gif.
In image processing, Image Steganography might be ordered into principle two area strategies. To start with is
spatial domain Steganography and second is Transform domain Steganography.
Spatial domain Steganography is an exceptionally prevalent and simple procedure. In this strategy we can
straightforwardly insert the Secret data into cover image. LSB is spatial domain and is a straightforward technique to
implanting data in cover image [1][2][6][10][14]. In LSB technique, the bits of the data straightforwardly reinstated with
least significant bit plane of the cover image. In any case straightforward LSB technique measurement not gives much
security compare to frequency domain.
Second name of Transform methods is known as frequency domain. Transform domain can have numerous sorts,
for example, the Discrete Cosine Transform (DCT), Discrete Fourier Transform (DFT), or Wavelet Transform
[1][2][6][10][12]. Transform domain is superior to spatial domain strategy. In this procedure, first image is transformed
that is known as transformed image and afterward the Secret data are implanted in the converted image. In this methods
Secret data in more critical ranges of the cover image that is it more robust [4][15][17].
This paper present a novel image Steganography method based LSB, X box mapping and Huffman Encoding. For
testing of algorithm, here main two 8 bit gray scale images are used. First image used as cover image and second image is
secret image. Firstly Huffman Encoding is performed over the secret image before embedding. Then generated compressed
Impact Factor (JCC): 6.8785

Index Copernicus Value (ICV): 3.0

Highly Secure Images Steganography Techniques Based on LSB, X-Box Mapping and Huffman Encoding

25

secret information embedded into cover image using X box mapping concept.

BACKGROUND/EXISTING METHODS
One of the earliest discussions on digital image Steganography by Charles Kurak and John McHugh explains
image downgrading through contamination. They have replaced the low order bits of one image with high order bits of
another image. Further they have shown that contaminant (hidden image) can be extracted from the contaminated (carrier)
image [19].It has provided the base for image Steganography and such method is known as least significant bits
replacement.
The simplest form of LSB replacement method uses least significant bit of all pixels in carrier image to embed the
secret information. It provides high embedding capacity but knowledge of such algorithms possible utilization let the
person to extract hidden content using a simple image processing program [1][2][19]. To avoid this unusual behavior
random and selective pixels, called potential locations, should be used for data embedding. The random distribution of
secret information in LSB plane of carrier image can avoid blind steganalysis of stego image and thus increases security.
Transform domain is also known as frequency domain, here image is first transformed using transformation such
as the Discrete Cosine Transform (DCT) [1], [2],[10],[12] Discrete Fourier Transform (DFT) [1],[2],[10] or Discrete
Wavelet Transform (DWT) [4],[7],[8],[15] and then the message is embedded in transformed image. These techniques hide
secret message in more significant regions of the cover image, making it more robust but provides less embedding
capacity.
Image Steganography based on LSB using X-box mapping in which many several Xboxes used with unique
data[4]. In embedding, Steganography algorithm used four unique X-boxes with sixteen different values (represented by 4bits) and each value is mapped to the four LSBs of the cover image. This mapping provides sufficient security to the
payload because without knowing the mapping rules no one can extract the secret data (payload).

PROPOSED MODEL
In this section, the definitions of performance analysis and proposed model are discussed.
Definitions Related to Performance Analysis

Capacity: The capacity is represented as size of secret information proportional to the size of stego image. It is
defined as bits per pixel (bpp) and in terms of percentage using Equation 1.

Mean Square Error (MSE): It is defined as the square of error between cover image and the stego image. The
distortion in the image can be measured using equation 1.
(1)
Where CI (i, j) is the cover image pixel
SI (i, j) is the stego image pixels
N N is the image size.

Peak Signal to Noise Ratio: PSNR is the measurement of the quality between the cover image and stego-image

www.tjprc.org

editor@tjprc.org

26

Kanzariya Nitin K, Nimavat Ashish V & Jadeja Vijaysinh K

and can be measured in db in using equation 2.

(2)
Proposed Model
In this proposed strategy, the secret image is implanted into cover image utilizing LSB substitution method, XBox mapping and Huffman Encoding to produce stego image that gives more security, high embedding capacity and high
PSNR.
Least Significant Bit Replacement
LSB is spatial domain system and it is a straightforward strategy to installing data in cover image. In LSB
methods, Msbs of the Secret data specifically displaced with Lsbs of pixel of the cover image [1][2][4].
8-bit gray scale image that means every pixel speak to with 8bits stream. For instance 9 pixels of gray scale image
demonstrate below:
(11010010 10101101 01100011)
(00101101 00011100 11011100)
(10100110 11000100 00001100)
Assume our Secret data is decimal number 201 so its binary representation is 11001001. This Secret data is
embedded into the LSBs of pixel of an image; the resulting bit pattern should be as follows:
(11010011 10101101 01100010)
(00101100 00011101 11011100)
(10100110 11000101 00001100)
As over, the 8 bits were embedded in initial 8 pixels from trace of an image; just the 3 highlighted (underlined)
bits need to be adjusted. On a normal, only half of the pixel values in a cover image need to be adjusted while inserting the
Secret data. Such minor progressions can't be noticeable by the human eye, so the message is effectively hidden into the
cover image.
Huffman Encoding
Huffman code is a prefix and least-length code as in no other encoding has a shorter normal length. Huffman
encoding reduces the amount of bits for every pixel. Table 1 show that if secret image of size 60 x 60 is inserted inside
cover image without compression, 28800 bits must be installed, while utilizing Huffman encoding just 13195 bits must be
installed. Therefore Huffman encoding increases the embedding capacity [1]. Secret image is encoded utilizing Huffman
coding and afterward Resultant Huffman codes are inserted into cover image [1], [17]. Huffman codes are optimal codes
that guide one symbol of cover image to one code word [1]. Figure 1 show the block diagram of Huffman encoding in
which secret image is converted to a 1-D bits stream. Huffman table (HT) represents binary codes of each symbol of cover
image [17].

Impact Factor (JCC): 6.8785

Index Copernicus Value (ICV): 3.0

Highly Secure Images Steganography Techniques Based on LSB, X-Box Mapping and Huffman Encoding

27

The Huffman table utilized at encoder and decoder side must be same. In this way the Huffman table is needed for
deciphering process alongside stego image [1]. Huffman encoding is mostly utilized for the accompanying three attributes:
Lossless Compression
It ensures the preservation of actual data while compressing it [1].
Increase the Security
Huffman encoded bit stream does not discloses anything because to extract the exact meaning, the Huffman table
is required [1].
Authentication
It provides authentication, as if any single bit changes in the Huffman coded bit stream, Huffman table will not be
able to decode the data [1].

Figure 1: The Block Diagram of Huffman Encoding

Table 1: Number of Bits With and Without Huffman Encoding
Secret
Image Size
60 X 60
100 X 100
140 X 140

Without Huffman
Encoding
No. of
Bits /
Bits
Pixels
28800
8
80000
8
156800
8

With Huffman
Encoding
No. of
Bits /
Bits
Pixels
13195
3.665
35429
3.542
65924
3.363

Generation of Four Different X(X-OR)-Boxes

Here 4 X-Boxes are produced whose measurement is a 2x2 lattice; in which 16 different (0 to 15) values are set as
given below [4].

Figure 2: X-Mapping Boxes

www.tjprc.org

editor@tjprc.org

28

Kanzariya Nitin K, Nimavat Ashish V & Jadeja Vijaysinh K

Each X box hold 4 separate values and we utilized X-OR property to place values in X-boxes. Sample of X-OR
property given below [4].
0 XOR 0 = 0,
1 XOR 1 = 0,
0 XOR 1 = 1,
1 XOR 0 = 1
For example 15 is stored in one of the 4 X-Boxes as follow:
15= 1011, 10 XOR 11=01
Thus the position of 15 is 0st row and 1st column [4].
Bit Division
Huffman coding allots a binary code to each intensity value of the image and a secret image is changed to a 1-D
bits stream like 11001010011010010001111
So we first gap the above bit stream into 8bit little stream, such as
11001010
11001001
11000011
.
.
.
Then again, we have to separation this 8bit stream values into 4parts that each one part hold 2bits. Case in point
this 8 bits stream 11001001 is separated into little 4 parts given below.
B1=11, B2=00, B3=10, B4=01
X-Box Mapping
Now the next step is X-box mapping in which we map the values of B1, B2, B3, and B4 with respective X-1, X-2,
X-3 and X-4 map box. For example first we map B1 with X-1 box and whose value is 11. First digits represent the row of
X-1 box and second digit represent column of same X box. So we map the value of 1st row and 1th column of the X-I box;
after mapping we get the resultant value 9 that is 1101. Same this process is repeated for the b2, b3, b4 and we get 15,14,1
sequentially.
Embedding of Secret Bits into the Cover Image
Above we get the new mapping values that are embedded into the cover image. In hiding procedure, 4LSB
replacement technique is used in which new mapping values can be replaced with the 4 LSBs of cover image. For
embedding, we select the pixels one by one in zig zag form from the cover image. 4 LSBs of cover image are replaced by
Impact Factor (JCC): 6.8785

Index Copernicus Value (ICV): 3.0

29

Highly Secure Images Steganography Techniques Based on LSB, X-Box Mapping and Huffman Encoding

mapped values like as 9, 15,14,1 sequentially.

For example, figure 3: considered as a cover image whose size is 128 *128 pixels. First mapped value is 9 that is
embedded into first pixel of cover image whose value is 51 using 4LSBs replacement technique and generate new stego
pixel value which is 56. Same this process can be repeated for all remaining values and generate new stego image as shown
in below figure 4.

Figure 3: (128x128) Cover Image

Figure 4: (128x128) Stego Image

RECOVERY PROCEDURE
Retrieve the 4LSB Bits from Stego Image
We retrieve the 4 LSB bits from each pixels of the stego image that is 1dimensional string such as
110110111110000; then we divide 1-D bit stream into 4bits pattern like 1101, 1011, 1110, 000 [1][4]. Then we perform
the XOR operation on the 4 bits stream. First we take the 2 bits, and we do the XOR operation with the other 2 bits.
Lsb1= 1101= 11XOR 01= 10
Lsb2= 1011 =10XOR 11 = 01
Lsb3= 1110 =11XOR 10 = 01
Lsb4= 0001 =00 XOR 01= 01
Concatenation of the Result of the XOR Operation
Now we combine the 2bits results of the XOR operation and generate new 1-D new bits string [1][4].
Decoded a 1-D Bit Stream is Using Huffman Table
Here we get coded secret information so to get original information; we need to decode this coded information
using Huffman table 1-D bit stream is decoded using Huffman table and get original secret bits [1][17]. Once the Decoded
bits are obtained, they need to be arranged in a sequence so that the secret image can be generated. The Decoded bits are
arranged based on the size of the secret image. Perfect arrangement of extracted bits generates an estimate of the secret
www.tjprc.org

editor@tjprc.org

30

Kanzariya Nitin K, Nimavat Ashish V & Jadeja Vijaysinh K

image.

ALGORITHMS
Problem Definition
The information is communicated securely using Steganography. The spatial domain, X-box mapping and
Huffman encoding techniques are used to generate stego image from cover image.
The Objectives Are

Improve PSNR

Increase capacity

Improve security

Assumptions

An ideal channel is used to transfer the stego image.

Huffman Table is also transmitted over an ideal channel.

Embedding Algorithm
Input: A Cover image and secret Image.
Output: A stego-image.

Impact Factor (JCC): 6.8785

Index Copernicus Value (ICV): 3.0

Highly Secure Images Steganography Techniques Based on LSB, X-Box Mapping and Huffman Encoding

31

Extraction Algorithm
Input: A M x N stego-image.
Output: A P x Q secret image.

RESULTS ANALYSIS
For performance analysis, several cover images such as Barbara, peppers, jet, Lena and boats shown in Figure 5
are used. Secret images of different file formats are selected for performance analysis. The secret image is embedded in the
cover image Lena using proposed algorithm to generate stego image Lena as shown in Figure 6. It is observed that
perceptibility of stego image and cover image are same. Figure 7 shows the recovery process in which secret image is
successfully recovered from stego image without any loss. By appearance the quality of stego image and secret image is
same as cover image and original secret image. Also statistical characteristic of stego image is almost same as cover image.

(a) Peppers.jpg

www.tjprc.org

(b) Jet.jpg

editor@tjprc.org

32

Kanzariya Nitin K, Nimavat Ashish V & Jadeja Vijaysinh K

(c) Barbara.jpg

(d) Leena.jpg

Figure 5: Cover Images

Figure 6: Cover Images and Secret Image

Figure 7: Stego Image and Recover Secret Image

Table 2: PSNR Values for Different Secret Image Formats
Cover Image
(128 X 128)
Baboon.jpg
Lena.jpg
Cameraman.jpg
Plane.jpg

Payload
(64 X 64)
Cameraman.gif
Hamilton.bmp
India.gif
Cameraman.png

PSNR
(CI & SI)
+36.125
+37.953
+38.954
+39.005

The PSNR values between cover image & stego image for different cover images with constant capacity are given
in Table 2. It shows that different cover image has less effect on the quality of stego image as PSNR values are almost
constant.
Table 3 demonstrates the compression between existing method [4] and the proposed strategy. The estimations of
PSNR of proposed technique are higher than exiting method [4]. It is watched that the estimation of PSNR is better on
account of proposed algorithm contrasted with existing method [4]. So the security to the payload in the proposed
algorithm is superior to existing method.
Table 3: Comparison of PSNR for Existing and Proposed Technique

Impact Factor (JCC): 6.8785

Index Copernicus Value (ICV): 3.0

33

Highly Secure Images Steganography Techniques Based on LSB, X-Box Mapping and Huffman Encoding

CONCLUSIONS
The Steganography is covert communication to protect confidential information. In this paper, a novel image
Steganography method based LSB, X-box mapping and Huffman Encoding. LSB replacement is simple technique. X-box
mapping increase the security of secret information. Huffman encoding is lossless compression technique that increases
embedding capacity. The algorithm enhances the security and the quality of the stego image and is better in comparison
with other existing algorithm. According to the results, the stego images of our proposed algorithm are almost identical to
the cover images and it is very difficult to differentiate between them. Here we used Huffman encoding technique that
provides lossless compression to increase the embedding capacity. Due to lossless compression, we achieved 100%
recovery of the secret image from stego image that means original and extracted secret images are identical. Security is
very high due to randomization and Huffman encoding technique. Decoding the secret image from stego image requires
Huffman table and x-box mapping rules. Without these elements, no one can decode the secret image from stego image.
It is also observed that the due to Huffman Encoding value of PSNR is better in the case of proposed algorithm as
compared with existing algorithm [4].

REFERENCES
1.

Nitin Kanzariya, Ashish Nimavat and Hardik Patel, Security of Digital Images using Steganography Techniques
based on LSB, DCT and Huffman Encoding Elsevier,( Page No.349-359),(ISBN:987-93-5107-184-6).

2.

Nitin Kanzariya and Ashish Nimavat, Comparison of Various Images Steganography Techniques International
Journal of Computer Science and Management Research (Page No. 1213 1217)(ISSN 2278-733X).

3.

Nitin Kanzariya and Ashish Nimavat, A Novel Technique for Image Steganography Techniques Based on LSB
and DCT Coefficients - International Journal for Scientific Research & Development| Vol. 1, Issue 11, 2014 |
ISSN: 2321-0613(Page No. 2405-2408).

4.

Amitava Nag and Saswati Ghosh An Image Steganography Technique using X-Box Mapping
IEEE-International Conference On Advances In Engineering, Science And Management (ICAESM -2012),
pp. 978-981,March 30, 31, 2012.

5.

Abbas Cheddad, Joan Condell, Kevin Curran, and Paul Mc Kevitt: Digital Image Steganography: Survey and
Analysis of Current Methods. ELSEVIER Journal on Signal Processing 90 (2010) 727-752.

6.

Ajit Danti and Preethi Acharya, Randomized Embedding Scheme Based on DCT Coefficients for Image
Steganography, IJCA Special Issue on Recent Trends in Image Processing and Pattern Recognition, pp. 97-103,
2010.

www.tjprc.org

editor@tjprc.org

34

Kanzariya Nitin K, Nimavat Ashish V & Jadeja Vijaysinh K

7.

H S Manjunatha Reddy and K B Raja, Hybrid Domain based Steganography using BPS, LSB and IWT.
Ad International Journal of Computer Applications (0975 8887) Volume 54 No.3, pp.9-16, September 2012.

8.

El-Sayed M. El-Alfy, Azzat A. and Al-Sadi A Comparative Study of PVD-Based Schemes for Data Hiding in
Digital Images IEEE International Conference on Computer Systems and Applications, pp. 144-149, 2011.

9.

J K Mandai and madhumita Sengupta, "Authentication/ Secret Message Transformation through Wavelet
Transform based Subband Image Coding (WTSIC)" International Symposium on Electronic System Design,
pp. 225 - 229, 2010.

10. Tanmay Bhattacharya, Nilanjan Dey and S. R. Bhadra Chaudhuri A Session based Multiple Image Hiding
Technique using DWT and DCT, International Journal of Computer Applications (0975 8887) Volume
38 No. 5, pp.18-21, January 2012.
11. V. Nagaraj, Dr. V. Vijayalakshmia and Dr. G. Zayaraz Modulo based Image Steganography Technique against
Statistical and Histogram Analysis, IJCA Special Issue on Network Security and Cryptography NSC, 2011.
12. Er. Mahender Singh, Er. Rohini Sharma and Er. Dinesh Garg, A New Purposed Issue for Secure Image
Steganography Technique Based On 2-D Block DCT and DCT,ijarcsse, Volume 2, Issue 7,pp.19-23, July 2012.
13. H S Manjunatha Reddy, N Sathisha, Annu Kumad, K B Raja, Secure Steganography using Hybrid Domain
Technique IEEE ICCCNT'12 26th _28th July 2012, Coimbatore.
14. El-Sayed M. El-Alfy, Azzat A. Al-Sadi Improved Pixel Value Differencing Steganography Using Logistic
Chaotic Maps, IEEE International Conference on Innovations in Information Technology (IIT),pp.129-133,
2012.
15. Hardik J. Patel and Prof. Preeti K. Dave. "Least Significant Bits based Steganographic Technique." International
Journal of Electronics Communications and Computer Engineering, January 2012, Vol. 3, Issue 1, pp 4450.
16. Jagvinder Kaur and Sanjeev Kumar, Study and Analysis of Various Image Steganography Techniques IJCST
Vol. 2, Issue 3,pp.535-539, September 2011.
17. A. Nag, S. Biswas, D. Sarkar and P. P. Sarkar, "A Novel Technique for Image Steganography Based on
Block-DCT and Huffman Encoding". International Journal of Computer Science and Information Technology,
Volume 2, pp.561-570, June 2010.
18. Amitava Nag, Sushanta Biswas, Debasree Sarkar and Partha Pratim Sarkar, A Novel Technique for Image
Steganography Based on DWT and Huffman Encoding, International Journal of Computer Science and Security,
(IJCSS), Volume (4): Issue (6), pp. 561-570.
19. C. Kurak and J. McHugh, A cautionary note on image downgrading, Proceedings of the IEEE 8th Annual
Computer Security Applications Conference, November- December 1992, pp 153159.

Impact Factor (JCC): 6.8785

Index Copernicus Value (ICV): 3.0