Sanjive Tyagi et al.

/ (IJAEST) INTERNATIONAL JOURNAL OF ADVANCED ENGINEERING SCIENCES AND TECHNOLOGIES

Vol No. 3, Issue No. 1, 029 - 033

MULTI LAYERS SECURITY SCHEME FOR

EMBEDDING SECRETS IN STEGO IMAGE
(MSSESSI)

Sanjive Tyagi * Ajay Agarwal

* Prof. & Head of the Department, Deptt. of M.C.A.
Deptt. of M.C.A., Radha Govind Engineering College,
Meerut, U.P. (India), K.I.E.T, Gzb., U.P.(India),
(Research Scholar Singhania University, jhunjhunu, E-mail: [email protected]
Rajsthan)
E-mail: [email protected]

Abstract—To improve the capacity of the hidden secret data and detected by non-authorized people. [3] Steganographic

T
to provide an imperceptible stego-image quality, a new technique is a means of covert communication and needs to
steganography method based on four layered security has been satisfy two basic requirements--good security and high hiding
constructed. We kept in mind the security of transmission and capacity
capacity of channel, triple EHDES provides more complexity and
enhanced security for efficiency of communication channel. The The purpose of steganography is to hide the very presence
Image segmentation schemes improve steganographic security of communication by embedding messages into innocuous-
and are less vulnerable to steganalytic attacks. looking cover objects, such as digital images. To accommodate

Keywords- EHDES,
Segmentation, and Image File.
Cryptography,
ES Stegnography,
a secret message, the original cover image is slightly modified
by the embedding algorithm to obtain the stego image. The
embedding process usually incorporates a secret stego-key that
governs the embedding process and it is also needed for the
extraction of the hidden message [4].
1. INTRODUCTION
There are three basic views behind hiding information. The
Steganography is a technology and science about information first is capacity, which is the amount of information that can be
hiding, which can ensures any unauthorized receivers can not embedded within the cover file. An information-hiding
discover the secret message except the authorized receiver [1]. algorithm has to be able to compactly store a message within a
file. Next is security, which refers to how a third-party can
A more erudite method of steganography is by merging the
detect hidden information within a file. Intuitively, if a message
two techniques to produce more security to secure data
is to be hidden, an ideal algorithm would store information in a
A
transmission such that if intruders detect the presence of data way that was very hard to notice. High security layers have
even then message cannot be decode without the knowledge of been proposed through four layers to make it difficult to break
key. through the encryption of the input data and confuse
The most common stegno method is the LSB approach, or steganalysis too. Various encryption techniques like
Least Significant Bit. As we know digital pixels are cryptography, digital watermarking, steganography etc have
represented by three colors: red, green and blue. These colors already been introduced in attempt to address these growing
together form digital pictures or video. Each color of every concerns [5].
IJ

pixel requires 1 byte or 8 bits of information. Since the first bit

is the “least significant” or carries the least amount of Steganography have four application areas:
importance in the byte, this steganographic technique chooses • Copyright Protection. It has security, invisibility and
to overwrite the first bit of successive bytes until the entire robustness requirements. Watermark techniques fit in this area.
secret message is embedded into the original source file, or the
cover data. Since we have only modified the least significant • Authentication. It has security and invisibility
bits of a portion of the source file, the human eye should not be requirements. Digital signature fits in this area.
able to detect the degradation in the picture or video [2].
• Secret and Invisible Communication. It has requirements
2. PRELIMINARIES for security, invisibility and insertion of high volumes of secret
data. [6].
2.1 Stegnography
2.2 Image Segentation
Steganography is a technique used to embed secret information
The main objective of image segmentation is to
into non-secret information, preventing the message from being
extract various features of image which can merged or split in

ISSN: 2230-7818 @ 2011 http://www.ijaest.iserp.org. All rights Reserved. Page 29

 Sanjive Tyagi et al. / (IJAEST) INTERNATIONAL JOURNAL OF ADVANCED ENGINEERING SCIENCES AND TECHNOLOGIES
Vol No. 3, Issue No. 1, 029 - 033

order to build objects of interest on which analysis and 1. M is a finite set of possible plain texts.
interpretation can be performed. [10, 11]
2. C is a finite set of possible ciphertexts.

2.2.1Region Approach to Image Segmentation

3. K, the keyspace, is a finite set of possible keys.
Regions in an image are a group of connected pixels
4. For each K ε k, there is an encryption rule eK ε E. and a
with similar properties. In the region approach, each pixel is
assigned to a particular object or region. In the boundary corresponding decryption rule dK ε D. Each eK : M →C and
dK : C → M are functions such that dK(eK(x)) = x for every
approach, the attempt is to locate directly the boundaries that
exist between the regions. In the edge approach, the edges are plaintext x ε M.
identified first, and then they are linked together to form The main property is property 4. It says that if a plaintext x
required boundaries. is encrypted using eK, and the resulting ciphertext is
subsequently decrypted using dK, then the original plaintext x
results.
2.2.2 Image Segmentation Based On Region Growing
Region growing is an approach to image segmentation
in which neighboring pixels are examined and adds to a region 2.3.1 Triple EHDES
class if no edges are detected. This process is iterated for each Triple EHDES uses the cascading or chain of Enhanced
boundary pixel in the region. If adjacent regions are found, a Data Encryption Standard (EHDES) [7, 8, 9].

T
region-merging algorithm is used in which weak edges are
dissolved and strong edges are left intact. Let EK(P.T.) and DK(P.T.) represent the EHDES
encryption and decryption of P.T. using EHDES key K
Region Growing requires a seed to begin with. respectively. Each EHDES encryption/decryption operation is
Ideally, the seed would be region, but it could be a single pixel. a compound operation of EHDES encryption and decryption
A new segment is grown from the seed by assimilating as operations. The following operations are used:
many neighboring pixels as possible that meet the homogeneity
ES
criterion. The resultant segment is then removed from the
process. A new seed is chosen from the remaining pixels. This
continues until all pixels have been allocated to segment. As
pixels are aggregated, the parameters for each for each segment
1) EHDES encryption operation:
the transformation of a 64-bit block P.T. into a 64-bit block
C.T. that is defined as follows:
have to be updated. The resulting segmentation could heavily C.T = EK3(DK2(EK1(P.T.))).
on the initially seed chosen and the order in which neighboring 2) EHDES decryption operation:
pixels are examined.
the transformation of a 64-bit block P.T into a 64-bit block
C.T. that is defined as follows:
2.2.3 Image Segmentation Based On Thresholding C.T. = DK1(EK2(DK3(P.T.))).
Thresholding techniques produce segments having The standard specifies the following keying options for
A
pixels with similar intensities. Thresholding is a use ful bundle (K1, K2, K3)
technique for establishing boundaries in image that contain
solid objects resting on contrasting background. There exist a 1) Keying Option 1: K1, K2 and K3 are independent keys.
large number of gray-level based segmentation methods using 2) Keying Option 2: K1 and K2 are independent keys and
either global or local image information. The thresholding K3 = K1.
technique requires that an object has homogenous intensity
and a background with a different intensity level. Such an 3) Keying Option 3: K1 = K2 = K3.
IJ

image segmented into two regions by simple thresholding [13].

3. OUR APPROACH
2.3 Cryptography In our new concept, we encrypt the original text message
letter by letter applying a function, which involves certain
Cryptography is a branch of applied mathematics that aims mathematical operation using corresponding letters and also
to add security in the ciphers of any kind of messages. numbers from the original image, then we use highly secure
Cryptography algorithms use encryption keys, which are the encryption using Double - EHDES algorithm to encrypt the
elements that turn a general encryption algorithm into a message. For encryption we need to use secret key for plain
specific method of encryption. The data integrity aims to verify text M and Triple - EHDES encryption function.
the validity of data contained in a given document. [7]
Cipher Text: C = EK(EHDES)(Message).
DEFINITION:
The image segmentation method has been introduced to
A cryptosystem is a five -tuple (M, C, K, E, D), where the decomposed cover image into two regions, i.e. foreground and
following conditions are satisfied: background region.

ISSN: 2230-7818 @ 2011 http://www.ijaest.iserp.org. All rights Reserved. Page 30

 Sanjive Tyagi et al. / (IJAEST) INTERNATIONAL JOURNAL OF ADVANCED ENGINEERING SCIENCES AND TECHNOLOGIES
Vol No. 3, Issue No. 1, 029 - 033

Then Hide encrypted text into background region of cover Calculate, Knew i = F(R and K)
image using Stegnography algorithm i.e List Significant Bit
(LSB) coding is the way to embed information in cover image Where R is a random number.
file. In this LSB technique is applied on encryption process on Step 6:
message. It is really appreciable four layer method to provide
high security to the high confidential image. Encryption using secret key and Tripe- EHDES

The proposed method is enhanced or characterized by Plain Text M and Cipher Text: C =
robustness, larger amount of secrete data, less time complexity EK(EHDES)(X).
and especially high security. 3.2 Segmentation Approach
Proposed work deals with the security of text message by We have proposed a steganographic method to embed
applying symmetric key cryptography algorithm Triple - secrete data into foreground/background region of still image.
EHDES in which we use generated secret key which are Foreground segmentation of image is used to divide the image
calculated using Triple - EHDES key generation process. into background and foreground.
Secret key is used at both sender side and receiver side. Secret We take an image file where cover image composed of
key are always different using Triple - EHDES algorithm with
two semantic objects; a foreground object and a plain
modification that a mathematical function F. This function
using a value depends on the decimal value of the R array of background object. Here, the goal is to separate two semantic
each pixel of cover image. The first letter corresponding to the objects for which we are using image segmentation method is
first pixel and next to the second pixel and so on . to extract various features of the image which can be merged

T
or split in order to build objects of interest on which the
A mathematical function F is using R array of each pixel of steganography can be performed. It has the advantages of
cover image and initial key K for generating the Secret key imperceptibility and robustness.
Knew i. The encrypted code is taken digit by digit. This The cover-image is to be decomposed into
approach constitutes the phase one security in our work. Background-cover-region and Foreground-cover-region. Then
In next phase, we have introduced the hiding of encrypted the embedding method determines the regions of foreground
and compressed text file into any cover image. and background in which the data can be embedded. The data
ES
In our work secret key are always different because we are
generating randomly number based on the confidential
message text and original cover image. This method is a unique
is embedded in to the appropriate region of cover image using
proper LSB algorithm. Then inverse steps are done to create
the stego data.
to generate random number such that no one can guess the At the destination the stego image received is
random number to crack the secret key. subjected to segmentation. Then employing suitable decoding
algorithm the embedded data is extracted. The same key used
3.1 Algorithm for encrypting the confidential message by the sender is used by the recipient.
Step 1: 3.2.1 Forground Segmentation of Cover Image
Convert the text to number system, which are ASCII
number of character.
A
Step 2:
A mathematical function f is used which gives the number
of random numbers below given number say R.
Step 3:
Here, the value of R depends on the decimal value of Original Image
IJ

character of cipher text.

Step 4:
Forground Segmentation
The F function is then applied on the random number R by
checking the parity of decimal value of character of cipher text.
X =ASCII Converted Character numbers of confidential
Message.
R =Decimal value of the R array of the pixels the original
cover image.
Knew i = F(R) = Result value after applying the F function
Step 5: Background Region (BR) Foreground mask (FR)
Now using random number R, generate secret key

ISSN: 2230-7818 @ 2011 http://www.ijaest.iserp.org. All rights Reserved. Page 31

 Sanjive Tyagi et al. / (IJAEST) INTERNATIONAL JOURNAL OF ADVANCED ENGINEERING SCIENCES AND TECHNOLOGIES
Vol No. 3, Issue No. 1, 029 - 033

3.2.2. The Stego-composite-Image, Pseudocode for 3.2.3 Following figures illustrates the encoding process:
embedding secrete data into cover image.
Let us denote secrete image by S and cover image by I and SG
is segmentation function, which takes cover image as an input
and decompose it into background-region (BR) and
Original Image foreground-region (FR).

SG (I) =BR and FR

Image simplification

Segmentation

Cover-image (I) BR
SG FR
Foreground Segmentation

T
Extraction of Foreground region -FR 3.2.4 The Encoder for embedding secrete data into cover
image:
E is an encoder function; it takes input as secrete-data (S) and
Foreground-region (FR). It generates a new image which is
called secrete-foreground-region (SFR).
Extraction of Background region-BR
ES E(S, FR) =SFR

Find size of Foreground region f-size

Encoder

Foreground-region secrete-foreground-region
E

Find size of secrete image s-size (FR) (SFR)

A
Secrete-data (S)

Secrete Image (S)

3.2.5 The Merger for embedding secrete data into cover

If s-size < f-size image:
then apply steganography algorithm to embed secrete-data in
IJ

FR, get secrete-foreground-region-SFR

M is Merger function, it takes a secrete-foreground-region
else (SFR) and Background-region (BR). It generate Stego-
choose another secrete file composite-Image î
endif M (SFR, BR) = î

Merger

Produce Stego-composite-Image by merging secerte-

foreground-region-SFR and Background region BR Secrete-foreground-region M Stego-composite-Image ( î )
(SFR)

Background-region (BR)

ISSN: 2230-7818 @ 2011 http://www.ijaest.iserp.org. All rights Reserved. Page 32

 Sanjive Tyagi et al. / (IJAEST) INTERNATIONAL JOURNAL OF ADVANCED ENGINEERING SCIENCES AND TECHNOLOGIES
Vol No. 3, Issue No. 1, 029 - 033

3.3 Algorithm to embed confidential message into cover A steganographic algorithm based on segmentation is
image file. proposed, which is quite other than the conventional
Algorithm to embed confidential message into cover image steganographic method. In this paper, we propose an
file named inFile generate new file with embedded message appropriate scheme by using the LSB matching method to
file named outFile. embed secure data into the segmented-cover-image and Triple -
EHDES is to provide high level security to secrete message.
Encoded-Message (msg,inFile on input-mode, outFile on Triple – EHDES is the three times cascading of EHDES with
output-mode) different mode. Compared with conventional steganographic
Step 1: method, this steganographic method has its strong and weak
Read offset bytes from input inFile and writes to output File points. This method has a value of a wide range of
outFile applications.
Step 2:
Calculate message length and write it into output file by REFERENCES
embedding using XOR function it in last two bits for every
byte. Suppose, Message length being 16 bits, will be stored in [1] Nameer N. EL-Emam, Hiding a Large Amount of Data with High
8 pairs of 2 bits. Security Using Steganography Algorithm Applied Computer Science
Step 3: Department, Faculty of Information Technology, Philadelphia University,
Jordan
Embed each byte of message in 4 pairs of 2 bits each is
embedded in 4 byte of input file and written into output file [2] Alain C. Brainos, A Study Of Steganography And The Art Of Hiding
Information, East Carolina University,

T
named outFile.
http://www.infosecwriters.com/text_resources/pdf/steganographyDTEC6
Step 4: 823.pdf
Write the remaining bytes of the input file into output file.
[3] Niels Provos, Peter Honeyman, Hide and Seek: Introduction to
3.4 Algorithm for generate of message from Image Steganography, IEEE Security and Privacy, Volume 1 , Issue 3 (May
2003), Pages: 32 - 44
The picture is received at receive side. This function decode
[4] Jessica Fridrich and Miroslav Goljan, Digital image steganography using
message from a file named outFile open on output mode.
Decode Message (outFile on Input-mode)
Step 1:
ES
Read offset bytes from the input file and apply again XOR
[5]
stochastic modulation, Department of Electrical and Computer
Engineering, SUNY Binghamton, Binghamton, NY, 13902-6000, USA.
Swarnendu Mukherjee, Swarnendu Bhattacharya, Amlan Chaudhury
Triple Layer Data Security ACM Ubiquity, Volume 9, Issue 17, April
29-May 5 ,2008
function, Generate message bit.
[6] Zhao, J. In business today and tomorrow, ACM Communications of the
Step 2: ACM, p. 7, 1998.
[7] Diego F. de Carvalho, Rafael Chies, Andre P. Freire, Luciana A. F.
Read last 2 bits of consecutive 8 bytes and concatenate
Martimiano and Rudinei Goularte, Video Steganography for Confidential
them to get the message length. Documents: Integrity, Privacy and Version Control , University of Sao
Step 3: Paulo – ICMC, Sao Carlos, SP, Brazil, State University of Maringa,
Computing Department, Maringa, PR, Brazil.
Read last 2 bits from input file in pairs of 4 and concatenate
A
[8] Ramveer Singh , Awakash Mishra and D.B.Ojha “An Instinctive
them to get message of 1 byte. Approach for Secure Communication – Enhanced Data Encryption
Standard (EHDES)” International Journal of Computer Science and
Step 4:
Information Technologies, Vol. 1 (4) , 2010, 264-267.
Repeat step 3 until the message is extracted of calculated [9] D.B. Ojha, Ramveer Singh, Ajay Sharma, Awakash Mishra and Swati
length. Garg “An Innovative Approach to Enhance the Security of Data
Encryption Scheme” International Journal of Computer Theory and
Step 5:
IJ

Engineering, Vol. 2,No. 3, June, 2010,1793-8201

Decrypt the message. [10] Digital Image Processing - S Jayaraman, S Esakkirjan, T Veerakumar,
Publisher-Mc Graw Hill
[11] Joel Molin “Foreground Segmentation of Moving Objects” Instutitionen
4. CONCLUSIONS for Systemteknik, Department of Electrical Engineering, Linkoping
University, SE-581 83, Sweden

ISSN: 2230-7818 @ 2011 http://www.ijaest.iserp.org. All rights Reserved. Page 33