A STEGANOGRAPHY ALGORITHM FOR HIDING DATA IN AN … · Least Signi cant Bit (LSB)is the most popular Steganography tech-nique. It hides the secret message in the RGB image based on

A STEGANOGRAPHY ALGORITHMFOR HIDING DATA IN AN IMAGE

BY ENHANCED LSBSUBSTITUTION METHOD

M.N.Jayaram1, M.Abhishek2,Madhu Bharadwaj3,Prabhava4,

Venkatesh.M5,1Associate Professor,2,3,4,5Final year B.E students

1,2,3,4,5Dept. of E&C Engineering, S.J.C.E ,Mysore-570006.

July 30, 2018

Abstract

The art of information hiding has been around nearly aslong as the need for covert communication. Steganographyis art and science of invisible communication which aroseearly on as an extremely useful method for covert infor-mation transmission. One of the most important factors ofinformation technology and communication has been the se-curity of the information. For security purpose the conceptof Steganography is being used. Our paper deals with im-age steganography. This paper explains about how a secretmessage can be hidden into an image using the improvedleast significant bit substitution method that makes use ofrandomization for hiding data and minimizes the chances ofdetection.Several problems arise among the existing LSB-based image steganographic schemes due to distortion in astego image and limited payload capacity. Thus, a schemehas been proposed and developed with the aim to improve

1

International Journal of Pure and Applied MathematicsVolume 120 No. 6 2018, 11645-11656ISSN: 1314-3395 (on-line version)url: http://www.acadpubl.eu/hub/Special Issue http://www.acadpubl.eu/hub/

11645

the robustness of the embedded information, increase pay-load of the secret data, by ensuring data integrity and alsoretaining the quality of the stego-image produced within anacceptance threshold.

Key Words:Security, Steganography, Least SignificantBit, Decryption, Encryption , Stego-key.

1 Introduction

Steganography is the art of passing information through originalfiles in a manner that the existence of the message is unknown.The term steganography is arrived from Greek word means, Cov-ered Writing. The innocent files can be referred to as cover text,cover image, or cover audio as appropriate. After embedding thesecret message it is referred to as stego-medium. A stegokey isused to control the hiding process so as to restrict detection and/orrecovery of the embedded data. While cryptography is about pro-tecting the content of messages (their meaning), steganography isabout hiding the message so that intermediate persons cannot seethe message. Steganography refers to information or a file that hasbeen concealed inside a digital Picture, Video or Audio file. Sincethe rise of the internet, themost important factor of informationtechnology and communication has been the security of informationThe motivation behind developing image Steganography methodsaccording to its use in various organizations to communicate be-tween its members, as well as, it can be used for communicationbetween members of the military or intelligence operatives or agentsof companies to hide secret messages or in the field of espionage. Inthe Steganography system scenario as depicted in Figure 1.1, beforethe hiding process, the sender must select the appropriate messagecarrier (i.e. image, video, audio, text) and select the effective se-cret messages as well as the robust password (which suppose to beknown by the receiver). The effective and appropriate Steganogra-phy algorithm must be selected that is able to encodethe messagein more secure technique. Then the sender may send the Stego fileby email or chatting, or by other modern techniques. The Stegofile is the Image embedded with the secret information. After re-ceiving the message by the receiver, he/she can decode it using the

2

International Journal of Pure and Applied Mathematics Special Issue

11646

extracting algorithm and the same password used by the sender[7][11]. Message is the data that the sender wishes to remain con-fidential. It can be plain text,cipher text, other image, or anythingthat can be embedded in a bit stream such as a copyright mark, acovert communication, or a se-

Figure 1.1 The Steganography model

rial number. Password is known as stego-key, which ensures thatonly recipient who knows the corresponding decoding key will beable to extract the message from a cover-object. The cover-objectwith the secretly embedded message is then called the Stego-object.Recovering the message from a stegoobject requires the cover-objectand a corresponding decoding key if a stego-key was used during theencoding process then the same key must be used in the decodingprocess. The original image may or may not be required in mostapplications to extract the message. Many carrier messages can beused such as Image, text video and many others. The image fileis the most popular used for this purpose because it easy to sendduring the communication between the sender and receiver. Theimages are divided into three types: binary (Black- White), Grayscale and Red-Green-Blue (RGB) images. The binary image hasone bit value per pixel represent by 0 for black and 1 for white pix-els. While the gray scale image has 8 bits value perpixel representfrom 00000000 for black and 11111111 for white pixels. The RGBimage has 24 bits values per pixel represent by (00000000, 00000000and 00000000) for black and (11111111, 1111111 and 11111111) forwhite pixels. The RGB image is the most suitable because it con-tains a lot of information that help in hiding the secret informationwith a bit change in the image resolution which does not affect theimage quality and make the message more secure. In this research

3

International Journal of Pure and Applied Mathematics Special Issue

11647

paper the RGB images are used as a carrier message to hide thesecret messages by the Least Significant Bit hiding method (LSB)as well as the proposed method. [6][5]

2 LeastSignificantBitHidingTechnique

Least Significant Bit (LSB)is the most popular Steganography tech-nique. It hides the secret message in the RGB image based on it itsbinary coding. LSB algorithm is used to hide the secret messagesby using algorithm1. LSB makes the changes in the image resolution quite clear aswell as it is easy to attack. [8][9] Algorithm (1)Least SignificantBit Hiding Algorithm. Inputs: RGB image, secret message and thepassword. Output: Stego image.1. Begin2. Scan the image row by row and encode it in binary.3. Encode the secret message in binary.4. Check the size of the image and the size of the secret message.5. start sub-iteration 1:6. choose one pixel of the image to start encoding process7. Hide a bit of the secret message in each part of the pixel in leastsignificant bit (8th bit).8. Set the image with the new values. 9. End sub-iteration 1.10. Set the image with the new values and save it.11. EndAlthough LSB hides the message in such way that the humans donot perceive it, it is still possible for the opponent to retrieve themessage due to the simplicity of the technique. Therefore, maliciouspeople can easily try to extract the message from the beginning ofthe image if they are suspicious that there exists secret informa-tion that was embedded in the image. Therefore, a system namedSecure Information Hiding System is proposed to improve the LSBscheme. It overcomes the sequence-mapping problem by embed-ding the message into a set of random pixels, which are scatteredon the cover image

4

International Journal of Pure and Applied Mathematics Special Issue

11648

3 TheProposedMethod

With respect to the Figure 3.1,the algorithm for the proposed LSBhiding tech

Figure 3.1 The proposed LSB hiding technique

nique can be formed.Algorithm(2) indicates the proposed LSB hid-ing technique. Algorithm (2): The Proposed Hiding Algorithm.Inputs: RGB image, secret message and the password. Output:Stego image. Begin1. Scan the image row by row and encode it in binary.2. Encrypt the secret message using RSA algorithm.3. Encode the encrypted secret message in binary.4. Check the size of the image and the size of the secret message.5. start sub-iteration 1:6. Choose one pixel of the image randomly7. Divide the image into three parts (Red, Green and Blue parts)8. Hide two by two bits of the secret message in each part of thepixel by searching about the identical.9. If the identical is satisfied then set the image with the new val-ues. Otherwise hide in the two least significant bits and set theimage with the new values.10. Save the location of the hiding bits in the next 4 pixels corre-sponding to each character hidden in the first 4 pixels.11. Same procedure is repeated for each character.12. End sub-iteration 1.13. Set the image with the new values and save it.14. Calculate the CRC for the encrypted message bits and store itin 3 different locations. End

5

International Journal of Pure and Applied Mathematics Special Issue

11649

4 Enhancingsecurity

We can make our secret message further secure by implement-ing RSA algorithm (for encryption) and CRC (Cyclic redundancycheck). CRC (Cyclic Redundancy Check) Consider a case whereyou transmit a password to one of your friend using image steganog-raphy. Because of some noise in the transmission channel, thatpassword may get changed by a letter. Even though message wastransmitted in a secure way, the receiver was not able to retrieve it.This could be overcome with the help of Cyclic Redundancy Check(CRC).A CRC-enabled device calculates a short, fixedlength bi-nary sequence, known as the check value or improperly the CRC,for each blockof data to be sent or stored and appends it to thedata, forming a codeword. When a codeword is received or read,the device either compares its check value with one freshly calcu-lated from the data block, or equivalently, performs a CRC on thewhole codeword and compares the resulting check value with anexpected residue constant. If the check values do not match, thenthe block contains a data error. The device may take correctiveaction, such as rereading the block or requesting that it be sentagain. Otherwise, the data is assumed to be error-free (though,with some small probability, it may contain undetected errors; thisis the fundamental nature of error-checking).

5 SteganographyProcess

Steganography process mainly consists of two phases, encryptionphase and decryption phase. During the encryption phase, mes-sage is hidden in the image file resulting in a stego image which istransmitted. In the decryption phase, message is extracted from thestego image. These two phases are well described in the followingsections.

5.1 EncryptionP hase

Figure 5.1 depicts the process that takes place during the encryp-tion phase. The Encryption Phase uses two types of files for encryp-tion purpose. One is the secret message which is to be transmittedsecurely, and the other is a carrier file such as image. In the en-

6

International Journal of Pure and Applied Mathematics Special Issue

11650

Figure 5.1 Encryption phase

Figure 5.2 Flowchart for Encryption process

cryption phase data is embedded into the image using proposedalgorithm by which least significant bits of the secret message arearranged with the bits of carrier file such as image, Such that themessage bits will merge with the bits of carrier file. In this pro-cedure LSB algorithm helps for securing the originality of image.Initially the message and the location of the image where messagehas to be hidden are entered. This location along with some spe-cial characters just like 23Q@l34 can be used as password wherethe starting location will be [23, 34].

5.1.1 Flowchart for Encryption process

Figure 5.2 indicates the Encryption process of the proposed method.Forthe CRC calculation, we must enter a generator polynomial at thebeginning of the encryption process. Message is encrypted usingRSA algorithm which produces different 8 bit sequence for eachcharacter. This encrypted message is hidden .Each character ishidden in four pixels by dividing 8 bit encrypted message into fourgroups of 2 bits each. These 2 bits are stored in each pixel .usingthe proposed algorithm. These 2 bits (message bits) are compared

7

International Journal of Pure and Applied Mathematics Special Issue

11651

with 2 successive bits of each pixel (each pixel contains 8 bits ina plane).If a match is found then the first bit position where databits are hidden is noted which may be 1, 2, 3, 4, 5, 6. If no match isfound between the data bits and image bits in a pixel then data bitsare stored in 7th and 8thbits and the bit position is noted as 7.Eachcharacter is stored in 4 pixels and the corresponding bit positionsare stored in last 3 bits of next 4 pixels. Therefore, we require 8pixels for embedding each character. This process is continued forall the characters. The length of the message will be stored in theimage which is of length 10 bits i.e., supporting maximum lengthof 1023. For the given message CRC is calculated using genera-tor polynomial and will be stored in three different locations sothat any error during transmission is found and can be asked forre-transmission.

Figure 5.3 Decryption process

Figure 5.4 Flowchart for Decryption phase.

5.2 Decryption Phase

5.2.1 Flowchart for the Decryption phase

Figure 5.3 depicts the pictorial view of the decryption phase andFigure 5.4 indicates the flowchart for decryption phase.From thepassword which gives the starting location of data bits , we need to

8

International Journal of Pure and Applied Mathematics Special Issue

11652

access last 3 bits of 5th to 8th pixel (from the starting location) toobtain the bit positions of first 4 pixels where data has been embed-ded. From these bit positions we can recover the data bits easily.For example if last 3 bits of 5th pixel (from starting location) gives101 i.e. 5 then 5th and 6th bit of 1st pixel (from the starting loca-tion) is extracted which gives the first 2 bits of encrypted character.Same process

Figure 6.1 The proposed and the stegoimage obtained.

is repeated to extract the encrypted data. CRC will be calculatedfor these data bits and will be compared with the CRC calculatedin the encryption process. Since we have stored same CRC in 3 dif-ferent locations, CRC obtained during decryption process will becompared with all those 3. If 2 or more out of 3 matches with theCRC calculated during encryption then we may assume that therewill be no errors and extracted bits are correct. Otherwise we mustask for retransmission. Using RSA algorithm, original message isobtained from this extracted data bits. Thus after giving the cor-rect password, the decryption section uses the Least Significant Bitalgorithm by which the encoded bits in the image is decoded andturns to its original state and gives the output as a text message.

9

International Journal of Pure and Applied Mathematics Special Issue

11653

6 Experimental Results

6.1 Comparision between LSB

Table 6.1 Comparision of image parameter values obtained byLSB and the proposed technique.

andproposedtechnique(image)parameters With reference to Figure6.1,we can calculate the image parameters for the LSB and pro-posed hiding techniques.Table 6.1 indicates the corresponding val-ues of the parameters for the LSB and proposed methods.The valuesindicated in the table are experimentally calculated and indicatesthat the proposed method has a far greater advantage than the LSBapproach in terms of the image ,noise presence and errors present.

7 Conclusion

In this a data hiding method by improved LSB substitution pro-cess is proposed. The image quality of the stego-image can begreatly improved with low extra computational complexity. Ex-perimental results show the effectiveness of the proposed method.The results obtained also show significant improvement in PSNRthan the method proposed in Ref. [2] with respect to image qualityand computational efficiency. A good balance between the securityand the image quality is achieved. Our future work will focus onimproving the efficiency of the proposed algorithm. The proposedscheme contributes a multi embedding feature that enable sendersto encode secret data into several cover create a stealth camouflage

10

International Journal of Pure and Applied Mathematics Special Issue

11654

to avoid intruders unwanted attention. Another advantage is theimplementation of bit substitution using sequential colour cycle al-gorithm to ensure the capacity of stego images remain unchangeddespite having multiple layers of encoding and decoding embedded.

References

[1] Amirthanjan,R.Akila, Deepika chowdavarapu, P., 2010. AComparative Analysis of Image Steganography, InternationalJournal of Computer Application, 2(3), pp.2-10.

[2] Bishop, M., 2005. Introduction to computer security. 1st ed.Pearson publications.

[3] Chan, C.K. Cheng, L.M., 2004. Hiding data in images by sim-ple lsb substitution: patternrecognition.vol 37.Pergamon.

[4] El-Emam,N.N., 2008. Embedding a large amount of informa-tion using high secure neural based steganography algorithm.International Journal of Signal Processing, 4(2), pp.5-4.

[5] Johnson, N.F. Jajodia, S., 1998. Exploring Steganography:seeing the unseen Computing practices, IEEE journal, Vol 1.Johnson, N.F. Jajodia, S.Duric, Z., 2001. Information hiding:steganography attacks and countermeasures. Kluwer academicpublishers.

[6] Krenn, J.R., 2004. Steganography and Steganalysis. IEEEcommunication magazine.

[7] Morkel et al., 2005. An overview of image steganography. Infor-mation security south Africa conference (ISSA) research group,vol .1, no .1.

[8] Provos, N. Honeyman,P.,2003. Hide and seek: an introductionto steganography. IEEE computer society.

[9] ZaidoonKh, A. Zaidan,A.A. Zaidan,B.BAlanazi.H.O., 2010.Overview: main

[10] fundamentals for steganography. Journal of Computing, 2(3),pp.40-4

11

International Journal of Pure and Applied Mathematics Special Issue

11655

11656