155

CHAPTER 8

Conclusions and Future Work

The growth of modern communication needs a special means of security especially on

computer network. As there appears a risk that the sensitive information transmitted

might be intercepted or distorted by unintended observers for the openness of the

internet. So it has resulted in an explosive growth in secure communication and

information hiding. Moreover, the information hiding technique can be used

extensively in applications like business, military, commercials, anti-criminal, digital

forensic and so on. Steganography is the technique of secret communication which

has received much attention. In this thesis image based steganography methods have

been proposed to increase the performance of the data hiding techniques. This thesis

focuses on the analysis and development of image steganography techniques that can

hide data with a low detection rate and high payload.

8.1 Summary of contributions

The main contribution of this thesis is providing an enhanced image based

steganographic technique for achieving the goal of data hiding using steganography.

To achieve this goal the various existing image based steganographic techniques i.e.

spatial domain based and frequency domain based with an application to data hiding

have been investigated. Some of the methods related to such domain available in

literature are discussed in chapter 2. All digital file formats can be used for

steganography, but the formats those are with a high degree of redundancy are more

suitable. The redundant bits of an object are those bits that can be altered without the

alteration being detected easily. Chapter 3 presents a study of the different file

formats that can be used in steganography. The most popular cover objects used for

steganography are digital images. Digital images often have a large amount of

redundant data, and this is what steganography uses to hide the message. In Chapter 4,

the techniques related to image based steganography on both spatial domain and

Conclusions and Future Work

156

frequency domain are investigated to understand the details of the basic working

process. Along with it, the attacks and techniques related to image steganalysis are

discussed. The basic evaluation measures of the image based steganography are also

discussed that are used to examine the performance of a steganographic technique.

Related to all of the above facts, the existing steganographic algorithms are expanded

by combining them with the cryptographic process. In the first case the cryptography

and DCT based steganography is combined to form a process that holds the features

of steganographic and cryptography technique to increase the security of the secret

data as stated in Chapter 5. The secret message is first encrypted using substitution

cipher method. Then the cover-image which is to hold the secret data is preprocessed

to reduce noise present in the cover-image and increase the dependence between

neighboring pixels, so that the embedding process may better utilize the bits. The

encrypted message is then embedded in the DCT coefficients of high frequencies of

the cover-image. In the process of embedding a modified standard quantization table

is used by putting ones (1s) in the coefficients located in the high frequency part. The

process of extracting the secret message from the resulting stego-image is also stated.

In the experimental analysis, a comparative study of the proposed method with Jpeg–

Jsteg (Hsu and Wu, 1999) and Chang et al., 2002 based steganography is also

conducted. The hiding capacity of the proposed method is more than Jpeg–Jsteg and

Chang et al., 2002 and also shows better PSNR values than both the methods. Along

with this, results of the various other evaluation parameters are also discussed.

In Chapter 6, the existing LSB based steganography is also combined with encryption

technique to enhance the embedding capacity of image steganography. The secret

message undergoes double encryption firstly using transposition cipher method and

then with substitution method before embedding into the cover-image file. Then

encrypted message is embedded into an image by using least-significant-bit (LSB)

technique that enables high capacity of data embedding. Once all the message

characters are embedded into the cover-image, the target character is inserted in the

pixel of the cover-image immediately next to the one containing the last input

character of the message. The process of extraction of secret message is also stated.

The main security lies in the encryption method where the secret message that is to be

embedded goes though double encryption process and the encryption process is

Conclusions and Future Work

157

controlled by two different keys. In examining the performance of the proposed

steganographic technique, an evaluation scheme for steganographic system is

conducted using various performance parameters on various images. A comparative

study is also conducted with OPAP (Chan and Cheng, 2004) based on embedding bits

of secret message in the cover images. From both the experiments it is found that the

proposed method shows better PSNR than OPAP. From other experiments it is found

that the proposed method shows better PSNR. Various evaluation measures are also

performed to test the proposed method.

Chapter 7 introduces an approach of least significant bit (LSB) based steganography

in digital images that can override some statistical and structural measures of

detection by spreading message bits randomly in which the secret messages are

embedded only in the red plane of the cover-image’s pixel determined by a

pseudorandom number generator (PRNG) initiated by a stego-key. In the process of

LSB embedding process, the random number generator selects the hiding points in the

pixel’s red plane of the cover image by using a random interval method. The random

interval produces a random sequence of locations of the secret data. In this method the

red plane is selected for data embedding while the blue and green planes are left

unmodified. So after performing the embedding operation the unmodified green and

blue planes are added to the modified red plane to form the final stego-image. In

extraction of the secret message, only the pixel’s red plane is selected using the same

stego-key that produces the same sequence of random locations of the secret bits. In

the experimental analysis, secret message of fixed and variable sizes using different

stego-keys are used to embed into the cover-images for examining the effect of

increasing key values and message size. A comparison of the proposed method is also

done with Amirtharajan et al. 2013 scheme using PSNR and MSE as performance

measure. It is seen that the PSNR value dynamically changes with the change in key

value in Amirtharajan et al. scheme whereas in the proposed method the PSNR value

are mostly static as the PSNR value are all same for different stego-keys. The method

is also evaluated using histogram analysis, visualizing LSB bit-plane and various

other performance measures to examine its ability to withstand from attacks.

In this thesis, the focus is not only on the embedding strategy, but also on the pre-

processing stages, such as secret message encryption and embedding area selection to

Conclusions and Future Work

158

improve the security. A comparative study of the three proposed method is presented

in table 8.1 based on certain criteria which are used in the process of their

development.

Table 8.1(a): Comparative study of the proposed image steganography methods

Criteria DCT with

encryption

LSB with

encryption

Random LSB

Basic operation Encrypts the secret

message and pre-

process the cover-

image before

embedding data in

the high frequency

coefficients of DCT

Encrypts the secret

message using two

encryption technique

and then embeds

data in the LSB of

the cover-image

Embedding operates

only in the red plane

of the cover-image’s

pixel by modifying its

LSB determined by a

pseudorandom

number generator

Operates in the

Image domain

Frequency domain Spatial domain Spatial domain

Image compression

used

Lossy Lossless Lossless

Pre-processing of

cover-image

Yes No No

Data embedding

process

Sequential Sequential Scattering(random)

Key used Key is used only in

the encryption

process

Key is used only in

the encryption

process

Stego-key is used for

random pixel

selection in the cover-

image

Embeds secret data

in

High frequency

coefficients of the

cover-image

Red, green and blue

plane of the cover-

image

Only in red plane of

the pre-selected pixel

of the cover-image

Encryption of secret

data

Yes Yes No

Conclusions and Future Work

159

Table 8.1(b): Comparative study of the proposed image steganography methods

Criteria DCT with

encryption

LSB with

encryption

Random LSB

Robustness against

attacks

Good Good Excellent

Invisibility of Secret

data

Highly invisible Highly invisible Highly invisible

Capacity Average in

comparison to other

two methods

High High

Stego-image size Lower than cover-

image

Same as cover-image Same as cover-image

Number of secret key

used

One key in

encryption

Two keys, one for

transposition and

substitution

encryption

One stego-key is

used for random

selection of pixels

Extraction process

depends on

Embedding and

encryption process

Embedding and

encryption process

Embedding process

and the stego-key

PSNR value Average High High

Overall Security Secure Secure Highly secure

Distribution of secret

data on the cover-

image

It embeds secret

data in order.

It embeds secret data

in order

Spreads throughout

the cover-image

Average PSNR value

for capacity of 28 K.B

42 dB 57 dB 62 dB

Time Computation

(For embedding and

extraction)

More Less Less

Key used in extraction Key is used in

decryption process

Two Keys are used in

decryption process

Key is used in

extracting the pixels

consisting of secret

data

Conclusions and Future Work

160

8.2 Scope for future Work

The world of digital media is in a continuous state of evolution. Steganography is

regarded as technology that has major competitive applications. While a significant

progress in the image steganography techniques has been achieved, still there is scope

for the improvement as there is yet to be evolved a standard method and the proposed

algorithms can be further enhanced. In this thesis, the study and analysis related to the

image based steganography relating to LSB and DCT has been done. The

enhancements can be done by using soft computing techniques such as Neural based

steganography, Fuzzy and Genetic algorithms based approaches. The future work can

also take into considerations of the Quantum computation approaches which can

extend the classical steganography for performance enhancement of the existing

techniques.

The existing transform and spatial domain based approaches can be enhanced with

certain variations. The DCT and DWT techniques can also be enhanced by using

randomization approach where the secret bits can be embedded randomly selected

blocks. Additionally, improving the embedding capacity of these methods that can

withstand severe compression can be considered. In the spatial domain enhancement

can also be done. The LSB based random embedding where the secret data is

embedded only in red plane can be enhanced using two planes for embedding (Red,

Green or blue plane) that will increase the embedding capacity and will also preserve

the security. The embedding capacity can also be enhanced by using more LSBs and

maintaining the statistical properties of the images.

References

180

Swathi, A. and Jilani, D. S. (2012). Video Steganography by LSB Substitution Using

Different Polynomial Equations. International Journal of Computational

Engineering Research, ISSN 2250-3005, Vol. 2, Issue 5, pp. 1620-1623.

Tan, K. T.; Ghanbari, M. and Pearson, D. E. (1998). An objective measurement tool

for MPEG video quality. Signal Processing, Elsevier, Vol. 70, No. 3, pp. 279-

294.

Tomar, G. (2012). Effect of Noise on image steganography based on LSB insertion

and RSA encryption. IOSR Journal of Engineering, Vol. 2, No. 3, pp. 473-477.

Tseng, H. W. and Chang, C. C. (2004). High capacity data hiding in JPEG-

compressed images. Institute of Mathematics and Informatics, Vilnius,

INFORMATICA, Vol. 15, No. 1, pp. 127–142.

Tyagi, V.; Kumar, A.; Patel, R.; Tyagi, S. and Gangwar, S. S. (2012). Image

Steganography using Least Significant Bit with Cryptography. Journal of global

research in computer science, Vol. 3, No. 3, pp. 53-55.

Upham, D. (1997). Jpeg–Jsteg, http://www.tiac.net/users/korejwa/jsteg.htm

Vassaux, B.; Nguyen, P.; Baudry, S.; Bas, P. and Chassery, J. M. (2002). Scrambling

technique for video object watermarking resisting to MPEG-4. In Video/Image

Processing and Multimedia Communications 4th EURASIP-IEEE Region 8

International Symposium on VIPromCom (pp. 239-244). IEEE.

Venkatraman, S.; Abraham, A. and Paprzycki, M. (2004). Significance of

Steganography on Data Security. In Information Technology: Coding and

Computing, 2004. Proceedings. ITCC 2004. International Conference on (Vol. 2,

pp. 347-351). IEEE.

Wang, H. and Wang, S. (2004). Cyber warfare: Steganography vs. Steganalysis.

Communications of the ACM, Vol. 47, No.10, pp. 76-82.

Wang, X.; Yao, Z. and Li, C. T. (2005). A palette-based image steganographic

method using colour quantisation. In Image Processing, 2005. ICIP 2005. IEEE

International Conference on (Vol. 2, pp. 1090-1093). IEEE.

References

181

Wang, Z.; Bovik, A. C. and Lu, L. (2002a). Why is image quality assessment so

difficult?. In Acoustics, Speech, and Signal Processing (ICASSP), 2002 IEEE

International Conference on (Vol. 4, pp. 3313-3316). IEEE.

Wang, Z.; Sheikh, H. R. and Bovik, A. C. (2002b) No-reference perceptual quality

assessment of JPEG compressed images. In Image Processing. 2002.

Proceedings. 2002 International Conference on (Vol. 1). IEEE.

Watson, A. B. (1994a). Image compression using the discrete cosine transform.

Mathematica Journal, Vol. 4, No. 1, pp. 81-88.

Watson, A. B. (1994b). Perceptual optimization of DCT color quantization matrices.

In Image Processing, 1994. Proceedings. ICIP-94., IEEE International

Conference (Vol. 1, pp. 100-104). IEEE.

Wayner, P. (2009). Disappearing cryptography: information hiding: steganography

and watermarking. Morgan Kaufmann.

Weng, S.; Zhao, Y.; Ni, R. and Pan, J. S. (2009). Lossless data hiding based on

prediction-error adjustment. Science in China Series F: Information Sciences,

Springer Vol. 52, No. 2, pp. 269-275.

Westfeld, A. and Pfitzmann, A. (2000). Attacks on steganographic systems.

In Information Hiding (pp. 61-76). Springer Berlin Heidelberg.

Wu, D. C. and Tsai, W. H. (2003). A Steganographic method for images by pixel-

value differencing. Pattern Recognition Letters, Elsevier, Vol. 24, No. 9, pp.

1613-1626.

Wu, H. R. and Rao, K. R. (Eds.). (2005). Digital video image quality and perceptual

coding. CRC press.

Wu, M. Y.; Ho, Y. K. and Lee, J. H. (2004). An iterative method of palette-based

image Steganography. Pattern Recognition Letters, Elsevier, Vol. 25, No. 3, pp.

301-309.

References

182

Wu, N. I. and Hwang, M. S. (2007). Data Hiding: Current Status and Key Issues.

International Journal of Network Security, Vol. 4, No. 1, pp.1-9.

Yu, Y. H.; Chang, C. C. and Lin, I. C. (2007). A new steganographic method for color

and grayscale image hiding. Computer Vision and Image Understanding, Vol.

107, No. 3, pp. 183-194.

Zhang, T.; Zhang, Y.; Ping, X. and Song, M. (2006). Detection of LSB

Steganography based on Image Smoothness. In Multimedia and Expo, 2006

IEEE International Conference on (pp. 1377-1380). IEEE.

Zhang, W.; Zhang, X. and Wang, S. (2007). A double layered ―plus-minus one‖ data

embedding scheme. Signal Processing Letters, IEEE, Vol. 14, No. 11, pp. 848-

851.

183

Appendix A

Definition of Terms

Artifact: Artifacts are the irregularities that may be present in an image after

processing. They are not related to the details of the image and sometimes

accompany transmitted signals.

Bit-plane: The bit-plane of an image is a set of bits corresponding to a given bit

position in each of the binary numbers representing the image.

Carrier: Carriers are the digital media that are used as a medium to carry the secret

data. It is the medium into which the secret data is hidden.

Ciphertext: It is the translated text which is formed by encrypting the secret data. It

is the unordered or substituted text created by changing its readability and

meaning.

Coefficient: Coefficients are formed when a signal is transformed a from an image

representation into a frequency representation by using discrete transformation.

Cover-image: The image which is taken as a medium of covering the secret data. It is

the original image into which the secret data is inserted into its redundant bits.

Cryptanalysis: It is the study of attacks on the cryptography to find weaknesses in

them, without necessarily knowing the key or the algorithm.

Decibel: The decibel (dB) is a logarithmic unit used to express the ratio between two

values of a physical quantity. It is used to measure signal level after processing

and is widely used in electronics, signals and communication.

Decryption: Decryption is the process of taking encoded or encrypted text and

converting it back into original plaintext which are understandable.

Eavesdropper: Eavesdroppers are the attackers (unauthorized person) who try to

break a signal through communication channel to check if the signal contains any

secret data.

Appendix A

184

Encryption: The process of transformation from plaintext (secret message) to

ciphertext (unreadable format) to create a data that is not understandable.

Entropy coding: Entropy coding encodes an image by rounding the coefficient

values to integer to reduce the size. Entropy coding is lossless compression.

Fidelity: Fidelity is the perceptual similarity between images before and after

processing.

Keystream generator: It generates random sequence of numbers by partitioning the

stego-key into random sequences. These sequences allocate positions of the

secret bits.

MSE: Mean squared error (MSE) shows variation between the cover- image and

resultant (stego) image. A high quality image should have less MSE value.

Payload: In steganography, payload relates to the amount of secret data that can be

hidden into digital media.

Pixel indices: Pixel indices are the random locations of the pixels which are formed

using random number generator.

Plaintext: It is the original secret message that is transformed into unordered state

using encryption process.

PRNG: The pseudorandom number generator (PRNG) generates sequence of random

numbers using a stego-key which is used to select random pixels of the cover-

image for hiding the secret data. The key is used as seed for the random number

generation.

PSNR: The Peak-Signal-to-Noise Ratio (PSNR) is the performance measurement

criteria that show the relationship between the bit- or detection-error of two

similar signals. A high quality image should have higher PSNR value.

Randomization: It is the process of scattering the secret bits in different pixel

position of the cover-image that makes the secret data to be in random order.

Seed: The seed defines the starting point of a random number generator. It initiates

the process of random number generation.

Appendix A

185

Steganalysis: Steganalysis is the technique to detect whether a given digital media

contains hidden data. The steganalysis plays a role in the selection of features or

properties of the digital media to test for suspicious data.

Steganalyst: Steganalyst is the individual (attacker) who performs the steganalysis

with the purpose of detecting suspicious or secret data into a medium when

transmitted.

Stego-image: The resultant image formed as a result of the steganography algorithm

which contains secret data embedded into it.

Stego-key: The secret key used in the steganographic method to choose the random

pixel position in the image. The security of a stego-key

Target character: It is the last character that is embedded into the cover-image. The

target character terminates the embedding and extraction process.

Zigzag: Zigzag order is performed to group similar frequencies together by sorting

the coefficients in zigzag ordering.

186

Appendix B

List of Publications

Journals

1. Laskar, S.A. and Hemachandran, K. (2012). An Analysis of Steganography

and Steganalysis Techniques, Assam University Journal of Science and

Technology, ISSN: 0975-2773, Vol. 9, No. 2, pp. 88-103.

2. Laskar, S.A. and Hemachandran, K. (2012). High Capacity data hiding using

LSB Steganography and Encryption. International Journal of Database

Management Systems (IJDMS), ISSN: 0975-5705, Vol. 4, No. 6, pp. 57-68.

3. Laskar, S.A. and Hemachandran, K. (2013). Steganography Based on Random

Pixel Selection for Efficient Data Hiding. International journal of Computer

Engineering and Technology (IJCET), ISSN: 0976-6367, Vol. 4, Issue 2, pp.

31-44.

4. Laskar, S.A. and Hemachandran, K. (2014). A Review on Image Steganalysis

techniques for attacking Steganography”, International Journal of

Engineering Research & Technology (IJERT), ISSN: 2278-0181, Vol. 3, Issue

1, pp. 3400-3410.

Book Chapters

1. Laskar, S.A. and Hemachandran, K. Combining JPEG Steganography and

Substitution Encryption for Secure Data Communication. In David C. Wyld,

et al. (Eds). Computer Science & Information Technology (CS & IT): ISSN:

2231-5403, CCSEA, SEA, CLOUD, DKMP, CS & IT 05, pp. 149–160,

(2012).

Appendix B

187

Conference and Symposium

1. Laskar, S.A. and Hemachandran, K. “Combining JPEG Steganography and

Substitution Encryption for Secure Data Communication” has been presented

in Second International Conference on Computer Science, Engineering and

Applications (CCSEA)-2012, CCSEA, SEA, CLOUD, DKMP, CS & IT 5,

Organized by AIRCC, May, 26-27, 2012, Delhi, India.

Citation Count -

(a) Laskar, S.A. and Hemachandran, K. (2012). An Analysis of Steganography

and Steganalysis Techniques, Assam University Journal of Science and

Technology, ISSN: 0975-2773, Vol.9, No-2, pp. 88-103. Cited by 1

Cited by

1. Maan, V. K. and Dhaliwal, H. S. (2013). Vector Quantization in Image

Steganography. International Journal of Engineering Research & Technology

(IJERT) Vol. 2, No. 4 (2013), pp. 421-424.

(b) Laskar, S.A. and Hemachandran, K. (2012). High Capacity data hiding using

LSB Steganography and Encryption." International Journal of Database

Management Systems (IJDMS), ISSN: 0975-5705, Vol.4, No. 6, pp. 57-68.,

DOI: 10.5121/ijdms.2012.4605. Cited by 5

Cited by

1. Bansal, T. and Lamba, R. (2013). Steganography on Colour Images using

32x32 Quantization Table. International Journal for Advance Research in

Engineering and Technology (IJARET), Volume 1, Issue V, 2013, pp. 68-72,

ISSN: 2320 6802.

2. Vasudev, P. and Saurabh, K. (2013). Video Steganography Based on Improved

DCT 32*32 Vector Quantization Method. International Journal of Software

& Hardware Research in Engineering (IJSHRE), ISSN: 2347-4890, Vol. 1, No.

4, 2013, pp. 46-51.

Appendix B

188

3. Garg, T. and Vatta, S. (2014). A Review on Data Compression using

Steganography. International Journal of Computer Science and Mobile

Computing (IJCSMC), Vol.3 Issue.6, June- 2014, pg. 275-278, ISSN 2320–

088X.

(c) Laskar, S.A. and Hemachandran, K. (2013). Steganography Based On

Random Pixel Selection for Efficient Data Hiding, International journal of

Computer Engineering and Technology (IJCET), ISSN 0976-6367, Vol. 4,

Issue 2, pp. 31-44. Cited by 4

Cited by

1. Qasim, M. A. and Pawar, D. (2013). Encryption & Steganography in IPv6 source

address” International Journal of Computer Engineering & Technology (IJCET),

ISSN: 0976 6375, Vol.4, No. 2 (2013) pp. 315-324.

2. Sumathi, C. P.; Santanam, T. and Umamaheswari, G. (2013). A Study of Various

Steganographic Techniques Used for Information Hiding." International Journal

of Computer Science & Engineering Survey (IJCSES), ISSN: 0976-2760, Vol.4,

no. 6 (2013), pp. 9-25.

3. Abdulhameed, Z. N. and Mahmood, M. K. (2014). High Capacity Steganography

Based on Chaos and Contourlet Transform for Hiding Multimedia Data.

International Journal of Electronics and Communication Engineering &

Technology (IJECET), ISSN: 0976-6472, Volume 5, Issue 1, (2014), pp. 26-42.

4. Mala, R. and Manimozi, I. A Novel Approach for Reversible Data Hiding in

Encrypted Images Using Key Based Pixel Selection. International Journal of

Computer Science & Engineering Technology (IJCSET), ISSN: 2229-3345, Vol. 5

No. 06 Jun 2014, pp. 715-719.

189

Appendix C

Participation in Conferences and

Workshops

1. “National conference on Current Trends in computer Science (CTCS 2010)”,

Department of Computer Science, Assam University, Silchar 22-24, February

2010, AUS.

2. “ISI-AU Workshop On Intelligent Data Analysis: Theory and Application”, 1-

5, March 2011, Computer Vision and Pattern Recognition Unit, Indian

Statistucal Institute, Kolkata and Department of Information Technology,

Assam University, Silchar, India.

3. “ISI-NEHU Winter School on Soft Computing, Pattern Recognition and

Image Processing”, October 20 – 24, 2011, Department of Information

Technology, North- Eastern Hill University, Shillong, Meghalaya.

4. “Workshop on Application of Mathematics in Computer Science and

Engineering”, August 2-4, 2011, Department of Mathematics, NIT, Silchar,

India.

5. “International Conference on Computer Science, Engineering and

Applications”, Organized by AIRCC, May, 26-27, 2012, New Delhi, India.

6. “13th

Workshop on Computational Information Processing”, December 3-7,

2012, Electronics and Communication Sciences Unit, ISI Kolkata and

Department of Information Technology, Assam University, Silchar, India.

7. “Workshop on Role of IPR in Electronics, Communication, Computing and

Devices”, November 27-28, 2013, Tezpur University IPR Cell and Institute of

Engineers, Silchar, India.