International Journal of Computer Science Engineering and Information Technology Research (IJCSEITR) ISSN 2249-6831 Vol. 3, Issue 3, Aug 2013, 185-192 © TJPRC Pvt. Ltd.

SECURED STEGANOGRAPHY APPROACH USING AES

© TJPRC Pvt. Ltd.,

MANOJ RAMAIYA1, NAVEEN HEMRAJANI2 & ANIL KISHORE SAXENA3 © TJPRC Pvt. Ltd., 1

Research Scholar, Department of Computer Engineering, Suresh Gyanvihar University Jaipur, Rajasthan, India 2

Professor, Department of Computer Engineering, Suresh Gyanvihar University Jaipur, Rajasthan, India

3

Professor and Director, SRCEM Banmore, Rajiv Gandhi Technical University Bhopal, Rajasthan, India

ABSTRACT Today’s cyber world is growing rapidly over Internet technologies & its applications needing high level of safeguard for expensive data during transmission over the open communication channel. Image steganography is a technique for hiding information into a cover image. Least Significant-Bit (LSB) replacement based method is most common steganographic technique in spatial domain owing to its simplicity and hiding capacity. Most of existing approaches focus on the embedding strategy with less consideration to the pre-processing such as encryption of secrete image. The conventional steganography algorithm does not provide the preprocessing required in image for better security and privacy. The proposed work presents an exclusive technique for Image steganography based on the Advance Encryption Standard (AES) using 128 bit block size of plaintext & 128 bits of Secrete key. The preprocessing provide high level of security as extraction of image is not possible without the knowledge of mapping rules of AES and secrete key .

KEYWORDS: Steganography, AES, S-Box, LSB Technique, Cryptography, Preprocessing of Secrete Image INTRODUCTION The emerging scenarios of communications system need the exceptional means of security especially in computer network and communication. The security standard assumes a clear separation between users and attackers and network security is gaining significance as the data being exchanged on the internet increases. Therefore, the security and privacy are required to safeguard against unauthorized access. This has resulted in an explosive growth of the field of information hiding, which covers applications such as copyright protection for digital media, Cryptography, Steganography, Digital Watermarking and fingerprinting. All these applications of information hiding are quite diverse. Cryptography and Steganography are widely used in the field of data hiding and has received significant attention from both industry and academia in the recent past. Former conceals the original data while latter conceals the very fact that data is hidden. Steganography provides high level of secrecy and security by combining with cryptography. Until quite recent times, the science of Cryptography was used regularly and exclusively by government and defense sector in order to protect the privacy of classified communication.

RELATED WORKS There are large numbers of steganography embedding techniques proposed in the literature. These procedures modify the cover image with different methods. But the entire embedding techniques share the considerable goal of maximizing the capacity of the stego channel [15]. In other words, the endeavor is to embed at highest possible rate while remaining imperceptible to steganalysis attack. Special domain embedding technique operates on the principle of correcting the parameter of the cover image like payload or disruption, so that the divergence between cover and stego image is insignificant and undetectable to the human eyes.

186

Manoj Ramaiya, Naveen Hemrajani & Anil Kishore Saxena

Steganography generally exploit human perception because human senses are not skilled to look for file that has hidden information inside them. Therefore steganography disguises information from people trying to hack them. Payload is the amount of data that can be hidden in the cover object. The most widely known image steganography algorithm is based on modifying the least significant bit of pixel value, hence known as LSB technique [1,2,6,8,12,13]. They are based on two techniques i.e. LSB replacement and LSB matching. Steganography with cryptography provide powerful tools for image security over communication. There are various cryptographic technique combined with steganography for additional security [1,2] like DES , S- box mapping etc. However they are very complicated and involves large computation. A software implementation of DES is not fast enough to process the large amount of data generated by multimedia applications and hardware implementation adds extra cost to both senders and receiver. In order to tackle these problems, system based on Advanced Encryption Standard (AES)[3,4] being very fast in some application such as smart card , cellular phones and image – video encryption is gaining importance. However, a central consideration for any cryptographic system is its susceptibility to possible attacks against the encryption algorithms such as statistical attack, differential and brute attacks. There are two levels of security for image encryption, Low level encryption, where the encrypted image has degraded visual quality compared to original image while in case of high level security, the content is completely jumbled and the image looks like random noise and not comprehensible at all to the viewers. The proposed new scheme in this paper provide both low level security using steganography & high level security using AES. Peak Signal to Noise Ratio (Psnr ) The measurement of the quality between the cover image f and stego-image g of sizes N x N (for 8 bit gray level) is defined by PSNR as: PSNR = 10 × log (2552 / MSE) N 1

Where MSE =

N 1

( f ( x, y) g ( x, y))

N 0

2

/ N2

N 0

Where f(x,y) and g(x,y) represent the pixel value at the position (x, y) in the cover-image and the stego-image respectively. The PSNR is expressed in dB. PSNR is representative of the quality of image i.e. the higher the PSNR, lower is the difference between cover image and stego image and vice – versa. Steganographic Triangle Several important issues need to be considered when studying steganographic systems. They are steganographic robustness, capacity, and security [6, 7]. The affiliation between them can be expressed by the steganography triangle shown in Figure 1. It represents balance of the desired characteristics associated with steganographic method. They are interdependent on each other and in order to improve one element, one or both of other elements needs to be sacrificed.

. Figure 1: The Steganography Triangle

Secured Steganography Approach Using AES

187

PROPOSED IMAGE STEGANOGRAPHY MODEL Proposed steganography model is based on DES function comprising of permutation, substitution, S–Box mapping and secrete key.

Figure 2: Proposed Model for Steganography First the secrete image is selected (e.g. of 64×64). Now the intensity value of each pixel of secrete image is converted from decimal to binary.

Figure 3: Conversion of Decimal Pixel Value to Binary Now taking sixteen consecutive pixel values from secrete image one block of 128 Bits is formed. Input this block to AES encoding function [3] described below. Encodind Function using AES Rijndael is a block cipher developed by Joan Daemen and Vincent Rijmen. The algorithm is supporting any combination of data and key size of 128/192/256 bits.The proposed method used its 128 bits version allow 128 bits blocksize and 128 bits keysize. AES is divided into four basic operational blocks. These operates on array of bytes and organised as a 4×4 matrix called state. For complete encryption , the data is passed through 10 rounds. These rounds are governes the following transformstion .

Bytesub Transformation: It is a nonlinear byte substitution using a subsitution table (s-box) , which is constructed by multilicative inverse and affine transformation. Byte Substitution controls each byte of state individually, with the input byte used to index a row and column in the lookup table to receive the substituted value.

188

Manoj Ramaiya, Naveen Hemrajani & Anil Kishore Saxena

Shiftrows Transnsformation: It is a simple byte transformation . The bytes in the last three rows of the state are cyclically shifed : the offset of the leftshift varies from one to three bytes.

Figure 4: Encoding Function (AES) Detail

Mixcolumns Transformations: Its operates on every column independently. Each byte of a column is mapped into a new value that is a function of all four bytes in that column. The substitution makes use of arithmetic over GF(28). It is designed as a matrix multiplication where each byte is treated as a polynomial in GF(2 8). The inverse used for decryption involves a different set of constants. This gives good mixing of the bytes within each column. Combined with the “shift rows” step provides good avalanche, that reflect within a few rounds, all output bits depend on all input bits.

Addroundkey Transformation: It is a simple XOR between the working stste and the round key. This transformation is its own inverse.

AES Key Expansion: The AES key expansion algorithm takes as input a 4-word (16-byte) key and gives a linear array of words, providing a 4-word round key for the initial AddRoundKey stage and each of the 10 rounds of the cipher. It involves copying the key first in to the group of 4 words, and then constructing subsequent groups of 4 based on the values of the previous 4th words. The first word in each group of 4 gets “special treatment” with rotate + S-box + XOR constant on the previous word before XOR’ing the one from 4 back. One complete execution of AES gives sixteen pixels value of secrete image into repective sixteen pixels value of

encrypted secrte image. This encrypted image gives high level of security during communication .

Figure 5: Output of Encoding Function

189

Secured Steganography Approach Using AES

Embedding Module Taking the encrypted secrete image, the values are converted from decimal to binary. Next divide this 8 bit values into 4 part taking 2 bits in each. Suppose The binary value of (173)10 = (10101101)2 Next divide this 8 bit value into 4 part taking 2 bits in each

Figure 7: Bit Division After bit division we get value of b1 = 10, b2 = 10, b3 = 11, b4 = 01.

Insertion of Bit into the Cover Image : After receiving values of b1, b2, b3, b4 , these values are inserted into the cover image. These values are placed into the 2 bit LSB of the four consecutive pixels in cover image. Taking the pixels one by one from the cover image, the 2 LSB bits are replaced by 10,10,11,01 respectively.

Figure 6: Cover Image (128 × 128)

Figure 7: Insertion of Bit into Cover Image

Formation of Stego Image: After receiving the new pixel value the stego image is formed by replaceing these values at their original position. Likewise the pixels value on by one from encrypted secrete image and insertion into the cover image and replaced them. Result becomes the stego image.

Encoding Algorithm Input: A gray level Secrete Image (m × n), A gray Level Cover of size (2m × 2n); Output: Stego Image of size (2m × 2n); Steps:

Input sixteen pixel value of the secrete image form block of 128 bits to the image encoding Function (AES), which produces the encrypted secrete image.

Divide the each pixel value of encrypted secrete image into 4 parts containing 2 bits each.

Insert these pixel values into the LSB position of first four pixels in the cover image one by one.

End.

190

Manoj Ramaiya, Naveen Hemrajani & Anil Kishore Saxena

Image Retrival At the receiving end decoding of stego image perform the following process:

Generate the 2 LSB Bits from the Stego Image The pixels are processed one by one from the stego image. Convert it into binary values and take 2 LSB bits from

four consecutive pixel values: Similarly taking next three pixels. i.e. 242, 35, 97; (242)10 = (1 1 1 1 0 0 1 0) 2 (35)10 = (0 0 1 0 0 0 1 1) 2 (97)10 = (0 1 1 0 0 0 0 1) 2 receiving, b1 = 1 0; b2 = 1 0; b3 = 1 1; b4 = 0 1;

Figure 8: LSB (2 Bits) Extension of Stego Image

Concatenation of Results: Now concatenating theinput, the 8 bits of first pixel value of encrypted secrete image is obtained as

Figure 9: Formation of Stego Image

Formation of Encrypted Secrete Image: Now the generated value are placed into first position. Likewise taking the next four pixel value from stego image, the process is repeated and the whole encrypted secrete image is retrived.

Formation of Secrete Image: Now the eight consequitive pixel value from encrypted secrete image are again inputted to AES decoding function with same parameter and keys one by one (but used in reverse order) to obtain respective eight pixels value of the original secrete image.

Decoding Algorithm Input: Stego Image of size (2m × 2n); Output: A grey level Secrete Image (m × n); Steps:

Input each pixel and take 2 bit LSB from 4 consecutive pixel value of the stego image.

191

Secured Steganography Approach Using AES

Concatenate four 2bit LSB to obtain 8 bits of each pixel value of encrypted secrete image.

Now sixteen consecutive pixel value forming block of 128 bits are inputtted to decoding Function (AES) using same parameter but keys value used in reverse order, to get first eight pixel value of secrete image.

End.

RESULTS AND ANALYSIS Proposed model is stronger Steganography technique as without knowing the secrete keys and the S-box mapping function, the extraction of secrete image from the stego image is impossible. Moreover quality of cover image is also not degrade due to variation in two LSB of each pixel which reflects only difference of 0 – 3 (insignificant) pixel value. Additionally, the proposed scheme is capable of not just scrambling data but also change the intensity of the pixels which adds to the safety of the encryption. Tabel 1: Capacity & Psnr Name of Image Baboon.jpg Cameraman.jpg Lena.jpg

Size (Pixel ) 64× 64 64× 64 64× 64

Capacity 25 % 25 % 25 %

PSNR In dB 54.58 55.01 59.28

CONCLUSIONS In the proposed AES based steganography model, the strength of S-box mapping and secrete key for encrypting secrete image, improves security and image quality compared to existing algorithms. All of the cryptographic algorithms presented so far have some problems. The earlier ciphers can be broken with ease on modern computation systems. The DES algorithm was broken in 1998 and Triple DES turned out to be too slow for efficiency as the DES algorithm was developed for mid-1970’s hardware and does not produce efficient software code. Triple DES has three rounds as DES and is correspondingly slower. 64 bit block size of triple DES and DES is not very efficient and is questionable when it comes to security. DES, AES is a symmetric block cipher. However AES is quite different from DES in a number of ways. The AES allows for a variety of block and key sizes not just the 64 and 56 bits as of DES’ block and key size. AES standard states that the algorithm can only accept a block size of 128 bits and a choice of three keys – 128/ 192/ 256 bits in the creation of a new standard. Also AES is not a feistel structure. In the feistel structure half of the data block is used to modify the other half of the data block and then the halves are swapped. In the AES entire data block is processed in parallel during each round using substitutions and permutations. Rijndael was designed to have resistance against all known attacks like brute force etc. Speed and code compactness on a wide range of platforms and design Simplicity. Steganography, especially combined with the cryptography is a powerful tool which enables to communicate secretly with the rapid development of digital technology and internet. Steganography has advanced a lot over past years. The proposed method characterizes gives high level of security using cryptography which is not visible to unauthorized access and low level of security using steganography.

ACKNOWLEDGMENTS The principal author acknowledgment is due to Sh.R.S.Sharma, Chairman, ShriRam Group of Colleges (SRGOC) for the encouragement and blessing to carry the research.

192

Manoj Ramaiya, Naveen Hemrajani & Anil Kishore Saxena

REFERENCES 1.

Manoj Kumar Ramaiya, Naveen Hemrajani, Anil Kishor Saxena, “Security Improvisation in image Steganography using DES”,3rd IEEE Trans. International Conference IACC -2013, Page(s): 1094 – 1099.

2.

Manoj Kumar Ramaiya, Naveen Hemrajani, Anil Kishor Saxena, Monika Sharma “Image Stenography: Self Extraction Mechnanism”, UACEE International Journal of Advances in Computer Science and its ApplicationsIJCSIA Vol -3 Issue -2 ,ISSN 2250-3765 Pg-145-148, 2013.

3.

Federal Information Processing Standared Publication (FIPS 197), “Advance Encryption Standard (AES) “, 26 Nov. 2001.

4.

M. Zeghid, M. Machhout, L. Khriji, A.Baganne, and R. Tourki, World Academy of Science and Technology March 2007 , 526-531.

5.

Yambin Jina Chanu, Themrichon Tuithung, Kh Manglem singh,“ A Short Survey on Image Steganography and Steganalysis Technique “, IEEE Trans. ,2012 .

6.

Amitava Nag, Saswati Ghosh, Sushanta Biswas, Debasree Sarkar, Parta Pratim Sarkar " An Image Steganography Technioque using X-Box Mapping", IEEE Trans. International Conference Advances in Engineering, science and Mamagement (ICAESM- 2012) 709 -713.

7.

Ge Huayong, Huang Mingsheng, Wang Qian, "Steganography and Steganalysis Based on Digital Image", IEEE Trans. International Congress on Image and Signal Processing,(2011) 252-255.

8.

W. Luo, F. Huang, J. Huang, Edge adaptive image steganography based on LSB matching revisited, IEEE Trans. Inf. Forens. Security 5 (2) (2010) 201-214.

9.

Guilliang Zhu, Weiping Wang, “Digital Image Encryption algorithm based on pixel”, ICIS – 2010 IEEE International Conference 29-31 Oct 2010, pp – 769 – 772.

10. Jasmin Cosic, Miroslav Bacai, “ Steganography and Steganalysis Does Local web Site contain “Stego” Contain“, 52 th IEEE Trans. International Symposium ELMAR-2010, Zadar, Croatia 2009 , pp 85 – 88. 11. Zhang Yun-peng , Liu Wei “ Digital Image Encryption Algorithm Based on chaos and improved DES “, System, man and Cybernatics ,SMC 2009 , IEEE International Conference 11-14 Oct 2009, pp 474-479. 12. J. Mielikainen, LSB Matching Revisited, IEEE Signal Process. Lett. 13 (5) (2006) 285-287. 13. Saeed R. Khosravirad, Taraneh Eghlidos and Sharokh Ghaemmaghami, “Higher Order Statistical of Random LSB Steganography”, IEEE Trans. 2009, pp 629 - 632. 14. N Provos and P. Honeyman, "Hide and seek: An Introduction to Steganography", IEEE Security and Privacy, 2003, pp32-44. 15. Donovan Artz " Digital Steganography: Hiding Data within Data ", Los Alamos National Laboratory, IEEE Trans. 2001, pp 75-80. 16. Schaefer " A Simplified Data Encryption Standard Algorithm ", Cryptologia, January 1996