HIGHLY SECURED AND RANDOMIZED IMAGE STEGANOGRAPHIC ALGORITHM

Dr. R.Sridevi

HIGHLY SECURED AND RANDOMIZED IMAGE STEGANOGRAPHIC ALGORITHM

Dr. R.Sridevi
Associate Professor Department of Computer Science and Engineering JNTUH College of Engineering, JNTUH, Hyderabad

Corresponding Author: Dr. R.Sridevi, E-mail: sridevirangu@yahoo.com

Related article at Pubmed, Scholar Google

Visit for more related articles at Journal of Global Research in Computer Sciences

Abstract

In today’s internet scenario, secured data transfer is very difficult if not impossible due to the technology and computing power availability to the attackers. Hence more robust methods are required to provide a secured data transfer. Though steganographic algorithms are existing, no algorithm is fool proof for long time, as hackers gain more knowledge over time [1]. In this proposed work, a new technique to improve the security of steganographic algorithm by using the high level of randomization is proposed and implemented. It has the high embedding capacity and more robustness in the stego key. In proposed algorithm, message to be transmitted is encrypted. The encrypted message is embedded on image in randomized pixels. The randomness of the position of pixels on which the encrypted message to be embedded will be decided by the stego key. The stego key itself is encrypted and transmitted to other party in a secured form. Hence it is more robust and secured algorithm. The algorithms used for steganography process is Pixel Value Differencing with Modulus (PVDM) [6] and Least Significant Bit(LSB) algorithms[5] with randomization.

Keywords

LSB, AES, RSA

INTRODUCTION

Here we are using the following two algorithms:

LSB Algorithm:

A simple and well known approach is to directly hide secret data into the least-significant bit (LSB) of each pixel in an image. Many modifications have been proposed to this LSB algorithm [11].

The proposed method uses a stego key to determine the pixels to embed the data bits (contrary to LSB of each pixel), thus provides more random spread of data.

Pixel Value Differencing with Modulus Function Algorithm:

In 2008 C.-M. Wang et al. proposed a refined version of Wu and Tsai scheme, Pixel Value Differencing with Modulus function [6]. In this method the modulus of two consecutive pixels is modified to embed the secret data instead of the difference of the pixel values.

The given image is scanned in a zigzag manner to obtain the pixels. Blocks of two consecutive pixels are obtained. Given a sub-block Fi composed of two continuous pixels P(i,x) and P(i,y) from the cover image, obtain the difference value di, the sub-range Ri such that Ri belongs to [li,ui], the width wi = ui – li + 1, the hiding capacity ti bits, and the decimal value v of ti for each Fi. Where li is lower limit and ui is upper limit of the Range Ri.

The remainder values Prem(i,x), Prem(i,y) and Frem(i) of P(i,x), P(i,y) of sub-block Fi are computed respectively by using the following equations:

Prem(i,x) = P(i,x) mod wi

Extraction Scheme:

In recovery process, the secret data can be extracted without using the original image. Nevertheless, it is essential to use the original range table R designed in the embedding phase inorder to figure out the embedding capacity for each sub-block Fi. Given a sub-block Fi with two consecutive pixels from the stego-image with their pixel values being P(i,x) and P(i,y) respectively, the difference value di of P(i,x) and P(i,y) is computed. Each Fi can be related to its optimal sub-range Ri from the original table R according to the difference value di. Hence, the width of the sub-range can be calculated as wi= ui - li, and the number of bits ti of the secret data can be extracted from Fi by equation

ti =floor (log2|wi|)

Eventually, compute the remainder value of Fi by using Equation

Frem(i) = (P(i,x) + P(i,y)) mod wi

And transform the remainder value Frem(i) into a binary string with the length ti. Frem(i) is nothing but decimal value to be extracted.

PROPOSED SYSTEM

Though many steganographic algorithms are available, none of them is completely randomized. Since all steganographic algorithms are public, once if an attacker suspects message bits on the stego image, the attacker can try all the possible methods of deciphering existing algorithms (brute force) to extract message bits.

Sender’s side Process:

In proposed method, the pixels of the carrier file, in which original message bits are stuffed is completely random and is decided by the stego key. The stego key inturn is transmitted to the other party after encrypting with the public key algorithm like RSA.

Figure 2 presents the block diagram of the proposed system on receiver’s side. Encrypted Stego key and the stego object are received by the receiver. Stego key must be decrypted and retrieved with the receiver’s private key. De-stuff the encrypted form of message bits using the parameter values in the stego key.

These bits can be decrypted using the secret key (which is one of the parameter in stego key) inorder to get the original message bits at the receiver’s end.

Below are the steps for the proposed algorithm to improve the security by randomization given by stego key:

a. Initially cover image having a resolution of 512 x 512 is taken and array of pixels with dimension 500 x 500 is considered for computational simplicity to divide an image as equal size of parts. The total number of parts an image is divided into is the first field of the stego key.

b. The part of the cover image at which the encrypted bits are to be stuffed is decided by second field of stego key.

c. Pixel position, at which stuffing of secure data bits starts in that corresponding image part, is given in third field of stego key.

d. The inter pixel distance between the pixels is fourth field of stego key.

Considering all these random variables, the position of the bits to be stuffed in the cover image is decided. Exchange of key can be achieved through various methods not just limited to network.

Once after deciding, instead of using the simple steganographic techniques, the proposed method uses the combination of methods, either the PVDM or 3-bit LSB method depending on the inter pixel value difference.

The proposed algorithm provides improvement of stuffing capacity. In PVDM method, the pixel pairs are classified as smooth area pixels, where the pixel value difference is small and edge area pixels, where the pixel value difference is large. It can be realized that the number of pixel pairs in smooth areas is considerable in amount.

Inorder to improve the capacity of the technique, these smooth area pixels can be used for embedding secret data using LSB replacement method, which accommodates more number of bits. PVDM method is used for the pixel pairs in edge areas.

PVDM method:

Consider a pixel pair in smooth area with values 32, 34.

The new difference is |38-35|=3< threshold value.

So the new values are 38 & 35.

The original Pixel values of cover image are considered are (246,100).The stego pixel values of the cover image after stuffing 7bits of secret bits are (247,101). Hence without much difference in pixel values, 7 secret data bits can be embedded which shows an improvement in data stuffing capacity.

RESULTS ANDANALYSIS

From Table 3, it can be observed that the hiding capacity of data bits is increased in the range of 56.57% to 99.70% for various images, by comparing the capacity of Pixel Value Differencing with Modulus function method.

The increase in data stuffing capacity for LSB, PVDM and proposed methods are shown in Graph 1. It is evident that the proposed method offers the accepted image qualities with the images having a PSNR value more than 36.

CONCLUSION

The proposed method is a randomized method using robust key for embedding encrypted data bits with higher capacity maintaining acceptable image quality. The stego key chosen by the user gives randomization property which can withstand steganalysis process. The stego key has good level of robustness, because it is encrypted using RSA algorithm while transmitting on the channel.

Finally this method gives a good quality because the PSNR values are greater than 36 and high embedding capacity (increase in hiding capacity ranges from 56.57% to the 99.70% for various images considered). The average increase in hiding capacity of the proposed method is 84.05%.

This proposed method is highly secured and has high embedding capacity with randomization properties to provide better data security while transmitting data bits on the channel compared to other existing algorithms.

References

Adem Orsdemir, H. Oktay Altun, Gaurav Sharma and Mark F. Bocko, “steganalysis aware steganography: statistical indistinguishability despite high distortion”, SPIE-IS&T, Vol.6819, 2008, pp.1 - 9.
Ahmad T. Al-Taani and Abdullah M. AL-Issa, “A Novel Steganographic Method for Gray-Level Images”, International Journal of Computer, Information, Systems Science and Engineering 3:1 2009.
Alvaro Martín, Guillermo Sapiro and Gadiel Seroussi, “Is Image Steganography Natural?”, IEEE Transactions on Image Processing, Vol.14, 2005, pp.2040-2050.
C.-C. Chang and H.-W. Tseng, “A steganographic method for digital images using side match”, Pattern Recognition Letters, Vol.25, 2004, pp.1431–1437.
C.-K. Chan and L.M. Cheng, “Hiding data in images by simple LSB substitution”, Pattern Recognition, Vol.37, 2004, pp.469 – 474.
Chung-Ming Wang, Nan-I Wu, Chwei-Shyong Tsai and Min-Shiang Hwang, “A high quality steganographic method with pixel-value differencing and modulus function”, The Journal of Systems and Software, Vol.81, 2008, pp.150–158.
Da-Chun Wu and Wen-Hsiang Tsai, “A steganographic method for images by pixel-value differencing”, Pattern Recognition Letters, Vol.24, 2003, pp.613–1626.
H.B.Kekre, Archana Athawale and Pallavi N.Halarnkarg, “Increased Capacity of Information Hiding in LSB’s Method for Text and Image”, International Journal of Electrical, Computer and Systems Engineering, 2008, pp.246-249.
Hong–juan zhang and Hong-jun tang,” A Novel Image Steganography Algorithm against Statistical Analysis”, in proceedings of ICMLC, 2007, pp.3884-3888.
Huaiqing Wang and Shuozhong Wang,”Cyber Warfare: Steganography vs. Steganalysis”, Communications of the ACM, Vol.47, No.10, 2004, pp.76-82.
Johnson N. and Jajodia.S, “Exploring steganography: Seeing the unseen”, IEEE Computer, Vol.31, 1998, pp.26–34.
Sorina Dumitrescu, Xiaolin Wu and Zhe Wang, “Detection of LSB Steganography via Sample Pair Analysis”, IEEE Transactions on Signal Processing, Vol.51, 2003, pp.1995-2007.
Tse-Hua Lan and Ahmed H. Tewfik, “A Novel High-Capacity Data-Embedding System” IEEE Transactions on Image Processing, Vol.15, 2006, pp.2431-2440.
Tseng Y.C, ChenY.Y and Pan H.K, “A secure data hiding scheme for binary images”, IEEE Transactions on Communications, Vol.50, 2002, pp.1227–1231.
Tseng Y.C and Pan H.K,” Data hiding in 2-color images”, IEEE Transactions on Computers, Vol.51, 2002, pp.873–878.
Behrouz A.Forouzan, “Cryptography & Network Security “, Special Indian Edition, 2007.
William Stallings, “Cryptography & Network Security Principles &Practices”, 3rd Edition, 2003.