A Secret-Key Image Steganography Technique using Random Chain Codes

With the wide range use of digital communication technologies, the Internet has been commonly used as a channel for transmitting various images. Steganography practises have been implemented for achieving such secure transmission. The main focus of steganography is data hiding where, digital images are utilized as the cover image. One of the image steganography techniques is based on LSB method, where the secret message bits are embedded sequentially in LSB of the bytes of the carrier image. This makes the hidden message vulnerable to detection by attackers. Many secret key image steganography techniques have been developed as alternative techniques to achieve a high level of security for the hidden secret message. But, these techniques failed to use the full capacity of the carrier image. In this paper, a secret key image steganography technique has been implemented using chains of a random sequence of indices (codes) of the bytes in the carrier image. These chains have been constructed based on the secret key used. This makes the hidden message more secure and difficult to depict by attackers. Furthermore, the proposed technique uses the full capacity of the carrier image. Visual and numerical tests have been conducted for the performance of the proposed technique, the recorded results proved it can be used effectively in the field of information hiding.

Chains of pixels; Hiding capacity; Image steganography; Information hiding; Information security; Secret key

Secret  image  represents  the  image  to  be  embedded  in  the  carrier  image, and stego image represents the image that holding the hidden message. The main goal of steganography is to make a stego image, which carries the secret messages. A secret stego keyis defined as password, which is an additional secret information required to protect the hidden secret image (Prasad & Pal, 2017).

Recently, images are commonly used as a carrier to hide information. According to type embedding information in the image content, steganography methods can be categorized into: spatial domain techniques (Gul & Kurugollu, 2010) and frequency domain techniques. While, the spatial domain techniques apply direct manipulation over pixels of the image, the transform domain techniques transform the image into the frequency domain and then hide the secret message. It is important to notice that the hiding capacity of secret messages in spatial domain technique is relatively larger than frequency domain techniques.

Steganography techniques which use stego key for hiding information are classified into three classes (Sumathi et al., 2014): pure steganography, secret key steganography, public key steganography.

A large number of images transmitted via the Internet have encouraged researchers to use these images as a cover media in developing the steganography methods for protecting data in the field of information security. Image steganography methods concentrate on the techniques of hiding a secret data in a cover image and generate a stego image which is carrying a hidden secret message (Kaur et al., 2014).  An image steganography model consists of three elements: secret data, cover image, and key. Hence, stego image is the cover image contains the secret data and stego key is the key used to embed the secret data in the cover image (Subhedar & Mankar, 2014).

The success of the image steganography methods depends on exploiting the weak point of detecting minor changes happened in the stego-image pixels. These changes made by the human visual system (HVS). There are some major properties that determine the strength and weaknesses of the image steganography techniques (Purohit & Sridhar, 2014): capacity, robustness, undetectability.

These goals are hard to achieve at the same time. Hiding long message makes it more vulnerable to detect by attackers, (i.e. It becomes less secure) and vice versa. One of the common techniques to do steganography is hiding secret messages in the least significant bit (LSB) of the image plane. LSB techniques in terms of the competing major properties are considered as a practical way to conceal messages in the spatial domain of the image plane. Furthermore, it can hide large quantities of data, for instance, high payload capacity (Hamid et al., 2012).

Least significant bit (LSB) is the traditional method of embedding secret information in a digital image (Yadav et al., 2014), (Uliyan & Al-Husainy, 2017). The traditional method uses the LSB of the pixels in the cover image to hide the bits of the secret message. This usually causes distortion in the stego image and the ratio of distortion depends mainly on the number of changes that occur in the LSB of pixels. This distortion must keep at the minimum to drive away any doubt about the presence of the secret message in the stego image.

In the field of information security, many researchers focused on innovation of strong steganography techniques. The main focus of this paper is on spatial domain techniques based on LSB. Various LSB techniques for image steganography have been covered (Trivedi et al., 2016).

One of the major goals of steganography techniques is to increase the amount of data hidden in the cover image; this certainly needs to use a large number of bits of pixels of the cover image. But, it raises the distortion ratio in the stego image and affects negatively on the quality of the stego image (Tiwari et al., 2014).

(Al-Husainy, 2012) proposed a steganography technique ,which reduces the distortion that occurs in the Stego image pixels through dividing the secret message into a set of blocks of the same length and finding the best similarity between LSBs of pixels and the blocks.

An image steganography technique based on dynamic pattern has been proposed by (Thiyagarajan et al., 2012). Through generating a dynamic pattern in the selection of indicator sequence, this technique aims to strengthen the security of hidden data. To minimize the distortion in the pixels, data should embed in the insignificant color channel of pixels and exclude the significant color channel.

(Rana & Singh, 2010) suggested LSB image steganography technique by using a pre-determined random selection of pixels; dividing the cover image to a set of segments and dividing the secret message into four blocks after encrypting it using data encryption standard (DES) method. A predetermined method is used to select a pixel in each block; each pixel represents the stego key. Three levels of security used in this technique through a combination of odd and even rows and columns respectively.

Novel steganography technique for the RGB format images has been presented by (Nilizadeh & Nilchi, 2013). It embeds a secret text in the blue layer of certain blocks. At the first, each block chooses a unique t1 × t2 matrix of pixels as a pattern, using the bit difference of neighbourhood pixels, for each keyboard character. Then, a secret message is hidden in the remaining part of the block. Blocks are chosen randomly using a random generator for increasing the security.

(Singh et al., 2014) proposed an image steganography method that hides a secret message using the N-Queen matrix (pattern) as the stego key. The value of N in the N-Queen matrix reflects the level of security in the steganography method. The numbers of solutions increase with the increase of N.

All the above, techniques are based on a randomly selecting of LSB of pixels do not achieve the use of the full capacity of the cover image but they are achieving good protection against attackers. On the other hand, the traditional LSB techniques, which are not based on the random selection of LSB of pixels, are more vulnerable to penetrate by attackers through reading the LSB of pixels in the stego image sequentially but it is achieving the use of the full capacity of the cover image.

The proposed technique works to use all the LSB of the pixels in the cover image in a random sequence. This will help to achieve the use of the full capacity of the cover image and provide high protection of the hidden data against the attackers. It proposed a secret key steganography technique, where it uses a secret key of size (256×8 bits = 2048 bits).

The rest of this paper is organized as follows: Proposed method will be introduced in Section 2. Section 3 presents Experimental results and performance analysis. In Section 4, Conclusions are drawn.

This paper presents a secret key image steganography technique to hide the secret message bits in the cover image. This technique achieves the use of all the pixels of the cover image and the selection of the pixels in the cover image randomly. The extraction of chains of elements from the secret key and the use of different key blocks for each block of pixels in the cover image is the new approach adopted in this technique. The evaluation tests of the proposed technique showed that the technique succeeded to achieve the three goals of the strong steganography techniques which are capacity, robustness, and undetectability.

The technique uses all the cover image pixels to achieve the full capacity. The technique uses relatively large Stego secret key (16×16) with the extraction of a chain of elements from it to achieve maximum randomization in selecting pixels in the hiding stage which lead to making this technique more robust.

The authors are grateful to the Middle East University, Amman, Jordan for the financial support granted to cover the publication fee of this research article.

Al-Husainy, M.A.F., Uliyan, D.M., 2017. Image Encryption Technique based on the Entropy Value of a Random Block. International Journal of Advanced Computer Science and Applications, Volume 8(7), pp. 260–266

Al-Husainy, M.A.F., 2012. Message Segmentation to Enhance the Security of LSB Image Steganography. International Journal of Advanced Computer Science and Applications, Volume 3(3), pp. 57–62

Nilizadeh, A.F., Nilchi, A.R.N., 2013. Steganography on RGB Images based on a "Matrix Pattern" using Random Blocks. International Journal of Modern Education and Computer Science, Volume 5(4), pp. 8–18

Gul, G., Kurugollu, F., 2010. SVD-based Universal Spatial Domain Image Steganalysis. IEEE Transactions on Information Forensics and Security, Volume 5(2), pp. 349–353

Hamid, N., Yahya., A., Ahmad, R.B., Al-Qershi, O. M., 2012. Image Steganography Techniques: An Overview. International Journal of Computer Science and Security (IJCSS), Volume 6(3), pp. 168–187

Kaur, S., Kaur, A.,  Singh , K., 2014. A Survey of Image Steganography. International Journal of Research in Electronics and Computer Engineering (IJRECE), Volume 2(3), pp. 102–105

Pal, A.K., Pramanik, T., 2013. Design of an Edge Detection Based Image Steganography with High Embedding Capacity. International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness, Springer, Volume 115, pp. 794–800

Prasad, S., Pal, A.K., 2017. An RGB Colour Image Steganography Scheme using Overlapping Block-based Pixel-value Differencing. Royal Society open science, Volume 4(4), pp. 161066

Purohit, A., Sridhar, P.S.V.S., 2014. Image Steganography: A Review. International Journal of Computer Science and Information Technologies, Volume 5(4), pp. 4891–4893

Rana, R., Singh, D., 2010. Steganography-concealing Messages in Images using Lsb Replacement Technique with Pre-Determined Random Pixel and Segmentation of Image. International Journal of Computer Science and Communication, Volume 1(2), pp. 113–116

Singh, A., Dhanda, S.K., Kaur, R., 2014. Secure Image Steganography using N-Queen Puzzle and its Comparison with LSB Technique. International Journal of Innovations in Engineering and Technology (IJIET), Volume 3(4), pp. 4–8

Subhedar, M.S., Mankar, V.H., 2014. Current Status and Key Issues in Image Steganography: A Survey. Computer science review, Volume 13-14, pp. 95–113

Sumathi, C.P., Santanam, T., Umamaheswari, G., 2014. A Study of Various Steganographic Techniques Used for Information Hiding. International Journal of Computer Science & Engineering Survey (IJCSES), Volume 4(6), pp. 9–25

Thiyagarajan, P., Aghila, G., Venkatesan, V.P., 2012. Dynamic Pattern Based Image Steganography. Journal Of Computing, Volume 2(8), pp. 1–9

Tiwari, N., Sandilya, M., Chawla, M., 2014. Spatial Domain Image Steganography based on Security and Randomization. International Journal of Advanced Computer Science and Applications, Volume 5(1), pp. 156–159

Trivedi, M.C., Sharma, S., Yadav, V.K., 2016. Analysis of Several Image Steganography Techniques in Spatial Domain: A Survey. In: Proceedings of the Second International Conference on Information and Communication Technology for Competitive Strategies, ACM, Article Number 84

Uliyan, D.M., Al-Husainy, M.A.F., 2017. Detection of Scaled Region Duplication Image Forgery using Color Based Segmentation with LSB Signature. International Journal Of Advanced Computer Science And Applications, Volume 8(5), pp. 126–132

Yadav, R.M., Tomar, D.S., Baghel, R.K., 2014. A Study on Image Steganography Approaches in Digital Images. Engineering Universe for Scientific Research and Management, Volume 6(5), pp. 1–6