effect of noise on hidden data -...

Effect of Noise on hidden data

Garima Tomar

Faculty of Electrical and Electronics,MITS Ujjain

[email protected]

Abstract -This paper simulates an effective data hiding technique

i.e. steganography based on LSB insertion and RSA encryption in

order to provide seven million times better security than the

previous work. The Main idea of proposed scheme is to encrypt

secret data by RSA 1024 algorithm, convert it in to binary sequence

bit and then embedded into each cover pixels by modifying the

least significant bits (LSBs) of cover pixels. The result image is also

known as steganography image. The PSNR value of this

steganography image is 54.34 db. In this paper Baboon image is

used for experimental purpous.This steganography image is

transmitted through AWGN channel, and performance is

simulated. The image and hidden data are reconstructed with the

SNR level ≥9 dB.

Key words: Steganography, LSB, RSA algorithm, AWGN channel

I INTRODUCTION

On internet the amount of digital images has increased

rapidly but security of images becomes increasingly

important for many applications, like, confidential

transmission, military and medical applications. In

steganography, the secret message is embedded into an

image (or any media) called cover image, and then sent to the

receiver who extracts the secret message from the cover

image. After embedding the secret message, the cover image

is called a stego-image. This image should not be

distinguishable from the cover image, so that the attacker

cannot discover any embedded message. The security of the

transformation of hidden data can be obtained by two ways:

encryption and steganography. A combination of the two

techniques can be used to increase the data security [1]. In

encryption, the message is changed in such a way so that no

data can be disclosed if it is received by an attacker [2].

Whereas in steganography [3], the secret message is

embedded into an image often called cover image, and then

sent to the receiver who extracts the secret message from the

cover message [4]. When the secret message is embedded

into cover image the it is called a stego-image. The visibility

of this image should not be distinguishable from the cover

image, so that it almost becomes impossible for the attacker

to discover any embedded message. There are many

techniques for encrypting data [5], which vary in their

security, robustness, performance and so on. Also, there are

many ways for embedding a message into another one.

The most popular one is embedding a message into a gray

image using LSB [6]. In this method the data is being hidden

in the least significant bit of each pixel in the cover image.

After hiding data in image the form of image is called stego

image. In this paper LSB insertion method is used for

embedding data in cover image ,resulting stego image is

transmitted through AWGN channel and performance is

simulated.

Fig 1 steganography Mechanism

II DATA HIDING ALGORITHM

In the application the user first specifies the data that they would

like to hide, which can be in any file format. The application

then encrypts this data using the recipient‟s RSA public key.

Once the encrypted data is obtained, the procedure described in

the following paragraph is repeated for each image in a user‟s

image library. Each bit of the encrypted data is compared to the

least significant bit of the pixel bytes in an image. The

comparisons are made starting from the first byte in the image

until the last byte that permits all the data to be hidden in that

image. The application cycles through the pixels of the image

Garima Tomar et al, International Journal of Computer Science & Communication Networks,Vol 2(1), 12-15

12

ISSN:2249-5789

looking for the block of bytes those results in the least number

of LSB changes.

A. least significant bit insertion method

Least significant bit insertion is a common, simple approach to

embed information in a cover file [7]. Usually, three bits from

each pixel can be stored to hide an image in the LSBs of each

byte of a 24-bit image. The resulting stego-image will be

displayed indistinguishable to the cover image in human visual

system [18].

Fig. 2: Least Significant Bit

The last bit of the byte is selected as least significant bit in one

bit LSB as illustrated in Figure 2 because of the impact of the bit

to the minimum degradation of images [32]. The last bit or also

known as right-most bit is selected as least significant bit, due to

the convention in positional notation of writing less significant

digit further to the right [20]. In bit addition, the least significant

bit has the useful property of changing rapidly if the number

changes slightly. For example, if 1 (binary 00000001) is added

to 3 (binary00000011), the result will be 4 (binary 00000100)

and three of the least significant bits will change (011 to 100).

Fig. 3: Example of bit addition

There are numbers of steganograpic tools which employ LSB

insertion methods available on the web. For example, S-Tools

take a different approach by closely approximating the cover

image which may mean radical palette changes. Instead, S-Tools

reduce the number of colours while maintaining the image

quality, so that the LSB changes do not drastically change

colour values. Another tool which is using LSB manipulation is

EzStego. It arranges the palette to reduce the occurrence of

adjacent index colours that contrast too much before it inserts

the message. This approach works quite well in gray-scale

images and may work well in images with related colours [18].

On the other hand, StegCure keeps the advantage of S-Tools and

EzStego that is maintaining the image quality, but it can prevent

the attack from hackers by restricting user to have only one

attempt to perform destego method. If the user has used the

wrong destego method for the first time, there is no second

attempt to recover the hidden data in the image even though the

user has chosen the correct destego method.

B Encryption Algorithm

As mentioned previously, the data to be hidden is first encrypted

using the RSA public key algorithm. Encrypting the data before

hiding it provides defense in depth, and makes the job of the

attacker more difficult if their goal is to recover the secret data.

The application uses the RSA algorithm for two reasons. First,

by using a public key algorithm the need for a private shared

key between the sender and recipient of the data is eliminated.

Shared keys are impractical because they require a secure way

of distributing the key to every person who you may want to

communicate with. A public key for a person can be distributed

fairly easily by publishing it on a website, or by emailing it to

people you expect would need to send you secret information.

Second, the RSA algorithm is also widely known and

demonstrably secure if large enough prime numbers are used to

generate the keys. Using an algorithm such as RSA which is

public knowledge is in keeping with the principle of open design

of secure software systems. Adhering to this principle was also

the reason we chose not to use our own encryption algorithm.

III SIMULATED RESULTS

Tool used: MATLAB Simulink version 7.0 .The name of image

is baboon, size of image is 256x256, and format of image is

BMP which we have taken for simulation .Here 256x256

meaning that number of rows and column of pixels, and BMP

meaning that the format of image i.e. bit map format. These four

images are gray scale images i.e. 8 bits per pixel. The proposed

method hides secret data bits in LSB of each pixel so we can

hide 65536 secret data bits in an image by using row ×column

× n relationship. Where n is no of LSBs used.The results are

tabulated in Table 1. In Table 1, the column labeled „Capacity‟

is the number of bits can be embedded into the host-image and

the column labeled „PSNR‟ is the peak-signal-to-noise-ratio of


13

ISSN:2249-5789

the stego-image. The results are the average value of embedding

65536 random bit-streams into the host images.

A simulation results when secret data bits are hide in cover

image

Here PSNR is calculated by following relation

MSE=1

𝐻×𝑊 .𝐻

𝐼=1 𝑃𝑉𝑂𝐹𝐶𝑂𝑉𝐸𝑅 𝐼𝑀𝐴𝐺𝐸 𝐼, 𝐽 −𝑊𝐽=1

𝑃𝑉 𝑂𝐹 𝑆𝑇𝐸𝐺𝑂 𝐼𝑀𝐴𝐺𝐸 𝐼, 𝐽 )2 (1)

PSNR =1𝑂 log 2552

𝑀𝑆𝐸 (2)

Cover image of Baboon Stego image of baboon PSNR =53.94

Fig 4 image used in this simulation

When this image is transmitted through AWGN channel, noises

are introduced in it, thereby this stego image may be corrupted

by noise and also hidden secret data bits are affected by the

noise. The experimental results shown in table 2 that how much

stego bits and data bits are corrupted by noise.The solution of

this problem is that detects these errors and corrects these errors

.In proposd method use channel coding to correct corrupted

secret hidden data bits by noise. For this purpouse use

convolutional coding and viterbi algorithm to detect and correct

errors .The experimental results shown in table 2 that how much

secret data bits and stego bits are corrected by channel coding.

The size of image is 256x256 i.e.65536 total no of pixels or

524288 data streams are transmitted through noisy channel.

Additive white Gaussian noises are used in this experiment. The

characteristic of this communication system with bit error rate

(BER) versus signal noise ratio (SNR, 𝐸𝑏 /𝑁0, dB) , where 𝐸0 is

energy per bit and 𝑁0 is noise spectral density. Such a controlled

noise was added in every channel and that stego image is

transmitted over the channel bit by bit. The number of error bits

was measured at every controlled noise level to obtain BERs for

test image during the stego image transmission. The received

stego bits are used to reconstruct the stego image, and extract

secret data bits by using decryption algorithm.. The error

correction results of the proposed method are given in the table

2.

B. simulation results when data stream is transmitted trough

awgn for the Baboon picture

IV CONCLUSION

The conclusion of these experimental results is that- The

proposed data hiding scheme is an efficient data hiding scheme

based on the LSB insertion and RSA encryption method by its

PSNR value of image like baboon is 54.00 dB, which is not

noticeable by human eyes. Enhance security of hidden data by

7x106 times than the RSA-512 in terms of its time complexity,

and 2650 times in space complexity. The steganography image

is transmitted through AWGN channel, and performance

TABLE 1 SIMULATION RESULS WHEN SECRET

DATA BITS ARE HIDE IN COVER IMAGE

Name of

the images

Embedding

secret data

capacity in bits

PSNR in db

Baboon 65536 54.34

Table 2 simulation results when stego image is transmit

through AWGN channel

SNR

IN db

No of bits corrupted

through

AWGN/524288

No of hidden data bits

corrupted through

AWGN /65536

0 41086 5080

1 29086 3688

2 11958 1466

3 6491 815

4 3231 411

6 1217 146

8 86 14

9 24 4

10 2 1


14

ISSN:2249-5789

simulated. The image and hidden data are reconstructed with the

SNR level ≥9 dB.

REFERENCES

1. Receiver Compatible Data Hiding in Color Image Debnath

Bhattacharyya1, Arpita Roy2, Pranab Roy2, and Tai-hoon

Kim3International Journal of Advanced Science and

TechnologyVolume 6, May, 2009.

2. W. Bender, D. Gruhl, N. Morimoto, A. Lu, “Techniques for data

hiding”, IBM Systems Journal, Vol. 35,Issue 3-4, 1996, pp. 313-336.

3. F. Petitcolas, R. Anderson and M. Kuhn, “Information Hiding-A

Survey”, Proceedings of the IEEE, Special Issue on Protection of

Multimedia Content, Vol. 87, Issue 7, July 1999, pp. 1062-1078.

4. N.F. Johnson and S. Jajodia, “Exploring steganography: seeing the

unseen”, Computer, Vol. 31, Isuue 2,February 1998, pp. 26

5. K. Stefan and A. Fabien, “Information hiding techniques for

steganography and digital watermarking”, Artech House Books,

December 1999.

6. Alfred J. Menezes, Paul C. van Oorschot and Scott A. Vanstone,

“Handbook of Applied Cryptography”,CRC Press, 1996. 7. R. Chandramouli and Nasir Memon, "Analysis of LSB Based Image

Steganography Techniques", 2001International Conference on Image

Processing, October 7-10, 2001, Thessaloniki, Greece, Vol. 3, pp.

1019-1022.

8. (FIPS), F. I. P. S. “Data encryption standard (des)” , Federal

Information Processing Standards Publication,October 25,1999.

9. M. Wu and J. Lee, "A Novel Data Embedding Method for Two-

Color Facsimile Images", International Symposium on Multimedia

Information Processing, Taiwan, December 1998.

10. M. Wu, E. Tang, B. Liu, "Data Hiding in Digital Binary Image",

2000 IEEE International Conference on Multimedia and Expo, Vol.

1, July 30-August 2, 2000, New York City, NY, USA, pp. 393-396.

11. Hsiang-Kuang Pan, Yu-Yuan Chen, Yu-Chee Tseng, "A Secure Data

Hiding Scheme for Two-Color Images", Fifth IEEE Symposium on

Computers and Communications, 3-6 July, 2000, Antibes-Juan

LesPins, France, pp.750.

12. M. U. Celik, G. Sharma, A. M. Tekalp, and E. Saber, “Reversible

data hiding”, IEEE ICIP-2002, Rochester, NY, Vol. 2, September 22-

25, 2002, pp. 157-160.

13. A.D. McDonald and M.G. Kuhn, “StegFS: A Steganographic File

System for Linux”, Third International.

14. Ismail Avcibas, Nasir Memon and B. Sankur, "Steganalysis Using

Image Quality Metrics", IEEE.

15. Transactions on Image Processing, Vol. 12, Issue 2, February

2003, pp. 221-229.Workshop on Information Hiding, LNCS, Vol.

1768, 1999, pp. 462-477.

16. Petticolas, Fabien, "Information Hiding Techniques for

Steganography and Digital Watermarking", Artech House Publishers,

December 1999.

17. Implementing Cisco IOS Network Security (IINS) Catherine Piquet

Copyright © 2009 Cisco Systems, Inc. Published by: Cisco Press 800

East 96th Street Indianapolis, IN 46240 USA.

18. StegCure: An Amalgamation of Different Steganographic Methods

inGIF Image L.Y. Por1, W.K. Lai2, Z. Alireza3, B. Delina4 Faculty

of Computer Science and Information Technology University of

Malaya 50603, Kuala Lumpur MALAYSIA .

19. S. Katzenbeisser, Fabien A.P. Petitcolas, Information Techniques for

Steganography and Digital Watermarking, Artech House,

Boston,London, 2000.

20. John Kadvany, “Positional Value and Linguistic Recursion”,

Springer Netherlands,Vol. 35, December 2007, pp. 487-520.

21. N. F. Johnson, S. Jajodia, “Exploring Steganography: Seeing the

Unseen,” IEEEComputer, February 1998, pp.26-34.Image, Vol. 3,

pp. 1019-1022. L.Y. Por1, W.K. Lai2, Z. Alireza3, B. Delina4,2008.

22. Mohammed Al-Mualla and Hussain Al- Ahmad, “Information

Hiding: steganography and Watermarking”. [Online].

Available:http://www.emirates.org/ieee/information_hiding.pdf,[Acc

essed: March 12, 2008].

23. Andrew D. Ker,“Steganalysis of Embedding inTwo Least-Significant

Bits”,IEEETransactions On Information Forensics And Security,Vol.

2, No 1, March 2007, pp. 46-54.

24. NeetaDeshpande,SnehalKamalapurandJacobsDaisy,“Implementation

of LSBsteganography and Its Evaluation for VariousBits”,1st

International Conference on DigitalInformation Management, 6 Dec.

2006 pp.173-178.

25. Muthiyalu Jothir,Navaneetha Krishnan,“Statistical models for Secure

steganographySystems”, Digital Rights Management Seminar, 15th

May, 2006.

26. S. Reinsberg, et. al., Phys. Med. Biol. 50, 2651-2661, (2005)T.

Morkel, J.H.P. Eloff, M.S. Olivier, “An Overview of Image

Steganography”,Proceedings of the Fifth Annual InformationSecurity

South Africa Conference (ISSA2005).

27. Mehmet Utku Celik, Gaurav Sharma, Ahmet Murat Tekalp and Eli

Saber.“LosslessGeneralized-LSBDataEmbedding”, IEEETransaction

on Image Processing, Vol. 14, No.2, Feb 2005, pp. 253-266.Sandton,

South Africa, 2005.

28. Alain Brainos, “A Study of steganography and the Art of Hiding

Information”, SecurityWriter,July 27, 2004.

29. Der-Chyuan Lou and Jiang-Lung Liu,“Steganographic Method for

SecureCommunications”, Elsevier Science Ltd, Vol21, No 5, 2002,

pp 449-460.

30. Simon Haykin, 'Communication Systems', 4th edition, John Wiley &

sons, Inc. 2001.

31. John G. Proakis, 'Digital Communications', Mc.Graw Hill, 4th

edition, 2001.

32. Petitcolas, Fabien A.P., “Information Hiding: Techniques for

Steganography and Digital Watermarking.”, 2000


15

ISSN:2249-5789

effect of noise on hidden data -...

Documents