International Journal of Computer Science Engineering and Information Technology Research (IJCSEITR) ISSN 2249-6831 Vol.

3, Issue 1, Mar 2013, 67-74 TJPRC Pvt. Ltd.

IMAGE ENCRYPTION TECHNIQUES: A CRITICAL COMPARISON

RAVI SHANKER YADAV1, MHD. RIZWAN BEG2 & MANISH MADHAVA TRIPATHI3
1

M.Tech Final Year (CSE), Integral University, Lucknow,Uttar Pradesh, India

2

HOD ,CSE & IT, Integral University, Lucknow, Uttar Pradesh, India

Associate Professor, Dept.of CSE, Integral University, Lucknow, Uttar Pradesh, India

ABSTRACT
Image, which covers the highest percentage of the multimedia data, its protection is very important. This can be achieved by image encryption. As the use of digital techniques for transmitting and storing image are increasing, it becomes an important issue that how to protect the confidentiality, integrity, and authenticity of image. There are various techniques which are discovered from time to time to encrypt the images to make images more secure. Image encryption techniques scrambled the pixels of the image and decrease the correlation among the pixels, so that we will get lower correlation among the pixel and get the encrypted image. There are so many different technique used to protect confidential image data from unauthorized access. In this paper, we have done the literature review on existing work which was done form different techniques for image encryption and also given the general introduction about images and image encryption with advantages and disadvantages.

KEYWORDS: Cryptography, Image, Encryption, Security and AES INTRODUCTION

Images are generally the collection of pixels. Basically Image Encryption means that convert the image into unreadable format. Many digital services require reliable security in storage and transmission of digital images. Due to the rapid growth of the internet in the digital world today, the security of digital images has become more important and attracted much attention. The prevalence of multimedia technology in our society has promoted digital images to play a more significant role than the traditional texts, which demand serious protection of users privacy for all applications. Encryption techniques of digital images are very important and should be used to frustrate opponent attacks from unauthorized access. Digital images are exchanged over various types of networks. It is often true that a large part of this information is either confidential or private. Encryption is the preferred technique for protecting the transmitting data. There are various encryption systems to encrypt and decrypt image data. However, it can be argued that there is no single encryption algorithm which satisfies the different image types. In general, most of the available encryption algorithms are used for text data. However, due to large data size and real time constants, algorithms that are good for textual data may not be suitable for multimedia data. Even though tripledata encryption standard (T-DES) and international data encryption algorithm (IDEA) can achieve high security, they may not be suitable for multimedia applications Therefore, encryption algorithms such as Data Encryption Standard (DES), Advanced Encryption Standard (AES), and International Data Encryption Standard (IDEA) were built for textual data. Although we can use the traditional encryption algorithm to encrypt images directly, this may not be a good idea for two reasons. First, the image size is often large than text. Consequently, the traditional encryption algorithms need a longer time to directly encrypt the image data. Second, the decrypted text must be equal to the original text but this

68

Ravi Shanker Yadav, MHD. Rizwan Beg & Manish Madhava Tripathi

requirement is not necessary for image data. Due to characteristic of human perception, a decrypted image containing small distortion is usually acceptable. The intelligible information in an image is due to the correlation among the image elements in a given arrangement. This perceptive information can be reduced by decreasing the correlation among image elements using certain transformation techniques. In addition to cryptography, chaotic image transformation techniques are getting significantly more sophisticated and have widely used. The chaotic image transformation techniques are perfect supplement for encryption that allows a user to make some transformation in the image, and then the image is totally diffracted, so nobody could see that what information could be shown through that image. Thus, it is often used in conjunction with cryptography so that the information is doubly protected, that is, first it is transformed by chaotic map transformation techniques, and then it is encrypted so that an adversary has to find the hidden information before the decryption takes place.

ALGORITHM ON IMAGES
Most of the algorithms specifically designed to encrypt digital images were proposed in the mid-1990s. There are two major groups of image encryption algorithms: Non-chaos selective methods Chaos-based selective or non-selective methods However, most of these algorithms are designed for a specific image format, either compressed or uncompressed. These are methods that offer light encryption (degradation), while offer strong form of encryption. Some of the algorithms are scalable and have different modes ranging from degradation to strong encryption. The user is expected to choose a method based on its properties, which will be best for image security. Image encryption has applications in internet communication, multimedia systems, medical and military imaging systems. Each type of multimedia data has its own characteristics such as high correlation among pixels and high redundancy. Thus, different techniques should be used to protect confidential image data from unauthorized access.

SURVEY ON PAPERS
In modified AES based algorithm for image encryption 2007 author analyze the advance encryption standard and in their image encryption technique they add a key stream generator (A5/1, W7) to AES to ensure improving the encryption performance[1]. AES is very fast symmetric block algorithm especially by hardware implementation [2, 3, 4, 5]. This Algorithm gives the better performance of image encryption in comparison to previous one. This algorithm is time taking and risky process. Secondly, in image encryption using block-based transformation algorithm, 2008 is based on the combination of image transformation and a well known encryption and decryption algorithm called blowfish. The original image divided into blocks, which were rearranged in to a transformed image using a transformation algorithm, and then the transformed image was encrypted using the Blowfish algorithm. And finely the result showed the correlation between image elements was significantly decreased .Their result also show that increasing the number of blocks by using smaller block sizes resulted in a lower correlation an higher entropy. In this algorithm there is no key generator .Here we use the Blowfish algorithm which divide image into number of blocks. Due to large data size and real time constrains, algorithm that are good for textual data may not be suitable for

Image Encryption Techniques: A Critical Comparison

69

multimedia data. In this algorithm the correlation between image elements was significantly decreased [6]. In most of natural images, the values of the neighboring pixels are strongly correlated [7, 8]. It has higher entropy. It has the possibility of losing image data in the duration of image blocking. It has lower correlation. Next is image encryption using self-invertible key matrix of Hill Cipher algorithm, 2008 that presents image encryption technique using the Hill Cipher. They are generating self- invertible matrix for hill cipher algorithm. Using this key matrix they encrypted gray scale as well as color image except for the image with background of same gray level or same color [9]. This is matrix based algorithm. It directly encrypts gray scale. It is simple to implement. This algorithm cannot work on image with background of same gray level or same color. Another survey is an image encryption approach using a combination of permutation technique followed by encryption, 2008 approaches a new permutation technique based on the combination of image permutation and a well known encryption algorithm called RaijnDael. The original image was divided into 4 pixels blocks, which were rearranged into a permuted image using a permutation process, and then the generated image was encrypted using the RijnDael algorithm. Their results showed that the correlation between image elements was significantly decreased by using the combination technique and higher entropy was achieved. It gives the higher entropy. In this method correlation between image elements is decreased. Here permutation process on image was very complicated. It is also a time taking process and chances of mistake are greater [10]. Next according to Novel image encryption algorithm based on Hash function, 2010, we use the image encryption based on SHA-512 hash function. The algorithm consists of two main sections; the first does preprocessing operation to shuffle one half of image. The second use the hash function to generate a random number mask. The mask is the XORed with the other part of the image which is going to be encrypted [11]. Here very few chances of mistakes, because encryption done in two sections. It is a lazy process of image encryption and work very slowly. Another survey done was a digital image encryption algorithm based composition of two chaotic logistic maps, 2010. In this encryption algorithm introduces an efficient chaos-based stream cipher, composting two chaotic logistic map and a large enough external secret key for image encryption. In this proposed image encryption scheme, an external secret key of 104 bit and two chaotic logistic maps are employed to confuse the relationship between the cipher image and the plane image. Further, to make the cipher more robust any attack, the secret key is modified after encrypting each pixel of the plain image. The robustness of the proposed system is further reinforced by a feedback mechanism, which makes the encryption of each plain pixel depends on the key, the value of the previous cipher pixel and the output of the logistic map. (Data dependent property) It is more efficient in comparison to previous encryption methods, because in this algorithm, we have used two logistic maps. It uses an external secret key. In this method security is very strong. Because here each plain pixel depends on key and output depends on the logistic map so, there is a lot of confusion created [12]. In new modified version of advance encryption standard based algorithm for image encryption, 2010, the author analysis advance encryption standard (AES) algorithm and present a modification to the advanced encryption standard (MAES) to reflect a high level security and better image encryption. Their result so that after modification image security is high. They also compare their algorithm with original AES encryption algorithm [13].

70

Ravi Shanker Yadav, MHD. Rizwan Beg & Manish Madhava Tripathi

It gives the high security in compression of others algorithms. They also compare their algorithm with original AES encryption algorithm. The algorithm and the secret key, consequently a same data will be ciphered to the same value; which is the main security weakness. In image encryption using affine transform and XOR operation, 2011this image encryption author propose a two phase encryption and decryption algorithm that is based on shuffling the image pixels using affine transform and they encrypting the resulting image using XOR operation. They redistribute the pixel values to different location using affine transform technique with four 8-bit keys. The transformed image then dived into 2 pixels x 2 pixels block and each block is encrypted using XOR operation by four 8-bit keys. The total key size used in algorithm is 64 bit. Their results proved that after the affine transform the correlation between pixel values was significantly decreased. It gives a better solution because in this algorithm we use the XOR operation. The correlation between pixel values was significantly decreased. It is very lengthy and complicated process. There is a lot of chance of mistakes because key distribution is very complicated [14]. The permutation based image encryption technique, 2011 is based on random pixel permutation with the motivation to maintain the quality of the image. The technique involves three different phases in the encryption process. The first phase is the image encryption. The second phase is key generation phase. The third phase is the identification process. This provides confidentiality to color image with less computations permutation process is much quick and effective. The key generation process is unique and is a different process. It is more effective because it is a three steps process. The first phase is the image encryption. The second phase is key generation phase and the third phase is the identification process. In this there is a High chance of error in key generation process [15]. In year 2011 one more algorithm was developed as image security by genetic algorithm, this is based on a hybrid model composed of genetic algorithm and a chaotic function for image encryption .In their technique, first a number of encrypted image are constructed using the original image with the help of the chaotic function. In the next stage, these encrypted images are employed as the initial population for starting the operation of the genetic algorithm. Then the genetic algorithm is used to optimize the encrypted image as much as possible. In the end, the best cipher- image is chosen as the final encryption image [16]. According to image encryption using chaotic maps and DNA addition operation and noise effects on it, 2011, author compared four chaotic maps cross chaotic, logistic, and Henon map and noise effects are observed on image. Firstly, they use the image encryption algorithm to convert original image to encrypted image. Then they apply noise on the encrypted image and then decrypt cipher image with noise back to original image. They have found out that cross chaotic map showed best results than other three chaotic maps. This algorithm applies noise on the encrypted image and then decrypt cipher image with noise to obtain original image. They have found out the cross chaotic map that shows best results than other three chaotic maps. It is a map base process so it is easy to represent. It is not a cost effective process [17]. Image encryption based on the general approach for multiple chaotic system, 2011, in this author proposed new image encryption technique based on new chaotic system by adding two chaotic systems, the Lorenz chaotic system and the Rossler chaotic system. From experimental analysis they demonstrate that the image encryption algorithm has the advantage of large key space and high-level security, high obscure level and high speed.

Image Encryption Techniques: A Critical Comparison

71

It gives the large key space and high-level security, high obscure level and high speed. It is a demonstrate process [18].Another new and effective method for image encryption using differential evolution approach in frequency domain, 2011, which employs magnitude and phase manipulation using differential evolution (DE) approach. They have carried out key space analysis, statistical analysis, and key sensitivity analysis to demonstrate the security of the new image encryption procedure. It employs magnitude and phase manipulation using differential evolution (DE) approach. It uses different evaluation approach to go on the critical situation [19]. In statistical analysis of s-box in image encryption application based on majority logic criterion, 2011, Author propose a criterion to analyze the prevailing S-Boxes and study their strengths and weakness in order to determine their suitability in image encryption applications. The proposed criterion uses the results from correlation analysis, entropy analysis, contrast analysis, homogeneity analysis, energy analysis and mean of absolute deviation analysis. These analysis to applied in advance encryption standard (AES), Affine-power- Affine (APA), gray, Luij, residue prime, S8 AESSKIPJACK and XYIs boxes. In this method we study their strengths and weakness in order to determine their suitability in image encryption applications [20]. We find the result in the form of correlation analysis, entropy analysis, contrast analysis, homogeneity analysis, energy analysis and mean of absolute deviation analysis. It is Complicated and lengthy process because there are lot of analysis done in single technique. Also here time factor will be increases. A new image encryption approach using the integration of a shifting technique and the AES algorithm march 2012, Author proposed a new encryption technique based on the integration of shift image blocks and basic AES, here the shifted algorithm is used to divide the image into blocks .Each block consists of number of pixels and these blocks are shuffled by using a shift method thats move the rows and columns of the original image in such a way to produce a shifted image. This shifted image is then used as an input image to the AES algorithm to encrypt the pixels of the shifted image. The main idea is that an image can be encrypted by shifting the rows and columns of original image and not to change the positions of the blocks but by shifting all the rows a number of times depending on the shift table, and then the same number of times for the columns for an arrangement of blocks [21]. It is an improved and effective method then previous technique, because here shifting algorithm is used which shift the original image into shift image. Some mistakes are possible in the maintaining shifting table. It is a lengthy and difficult process.According to design and analysis of a novel digital image encryption scheme march 2012, a non-chaos based image encryption scheme using an external key of 144-bits is presented. The Proposed encryption scheme uses both pixel substitution as well as pixel permutation process. A feedback mechanism, to avoid differential attack and make the cryptosystem more robust, is also applied. The proposed encryption technique scheme has high encryption rate, requires less computations and sensitive to small changes in the secret key so even with the knowledge of approximate key values, there is no possibility for the attacker to break the cipher. The proposed scheme is simple, fast and secured against any attack. It uses the substitution and permutation process. It is very large and complicated process. Here very difficult performance and security analysis [22]. Another algorithm came in 2012 was secret key encryption algorithm using genetic algorithm April 2012, a new approach of genetic algorithm is proposed in which, the operations of gross over and mutation are exploited to produce this

72

Ravi Shanker Yadav, MHD. Rizwan Beg & Manish Madhava Tripathi

encryption method .this new method was applied to the candidate type of data .example IMAGS. This paper presented in many sections firstly we describe the introduction of genetic algorithm and cryptography under the heads of introduction in section-1.The encryption method satisfies the goals that are required in any encryption method to encrypt image.It is a complicated process and algorithm is too lengthy [23].Lastly, one more paper was implemented of new advance image encryption to enhance security of multimedia concept July-2012, current proposal algorithm that fall into the representative image encryption technique, nave, position permutation, value transformation, substitution transposition and selective techniques will be described, evaluated, and compared, with respect to security level and encryption speed. When we will compare from many commonly used algorithms, the proposed algorithm result will produced best performance, the lowest correlation and the highest entropy. This algorithm completed after three phases and every phase is very complicated. It is very time taking process [24]. Table 1: Comparison Table TECHNIQUE Modified AES Based Algorithm 2007 Block-Based Transformation Algorithm, 2008 Self-Invertible Key Matrix Of Hill Cipher Algorithm, 2008 A Combination Of Permutation Technique Followed By Encryption, 2008 A Novel Image Encryption Algorithm Based On Hash Function, 2010 A Digital Image Encryption Algorithm Based Composition Of Two Chaotic Logistic Maps, 2010 New Modified Version Of Advance Encryption Standard Based Algorithm For Image Encryption, 2010 Image Encryption Using Affine Transform And XOR Operation, 2011 Permutation Based Image Encryption Technique, 2011 Image Encryption Using Chaotic Maps And DNA Addition Operation And Noise Effects On It, 2011 Image Encryption Based On The General Approach For Multiple Chaotic System, 2011 Statistical Analysis Of S-Box In Image Encryption Application Based On Majority Logic Criterion, 2011 The Integration Of A Shifting Technique And The AES Algorithm March 2012 ADVANTAGES Better performance No key generator, correlation between image elements decreased and higher entropy Matrix Based and Encrypt Gray Scale Higher Entropy and Correlation between image elements decreased Because of encryption done in two phases chances of mistakes is low Better than all above because of two logistics maps, Uses external sacred keys and Strong security Higher security DISADVANTAGES Time taking and risky Image loosing and lower correlation Cannot encrypt image with same gray level or color Permutation process is too complex, Time taking and also chances of mistakes are high Encryption done in two phases so will be increases Lot of confusion in process

Better Solution and Correlation between pixels values significantly decreases Three phases process Easy to represent

The algorithm and the secret key, consequently a same data will be ciphered to the same value; which is the main security weakness. Lengthy, complicated and chances of mistakes is high High chances of error in key generation Not a cost effective process

Large key space and high-level security, high obscure level and high speed Correlation analysis, entropy analysis, contrast analysis, homogeneity analysis, energy analysis and mean of absolute deviation analysis Improved and effective method

Demonstrate process

Complicated and lengthy process because there are lot of analysis done in single technique. Also here time factor will be increases. Possibility of mistakes while preparing shifting table, it is lengthy and difficult process

Image Encryption Techniques: A Critical Comparison

73

Design And Analysis Of A Novel Digital Image Encryption Scheme March 2012 Secret Key Encryption Algorithm Using Genetic Algorithm April 2012 New Advance Image Encryption To Enhance Security Of Multimedia Concept July 2012

Table 1: Comparison Table Contd., Simple, fast and secured against Large, complicated and very any attack difficult performance and security analysis Encryption method satisfies the Complicated and algorithm is goal of encrypting the images too lengthy Best performance, the lowest correlation and the highest entropy Three phase process and every image is very complicated

CONCLUSIONS
There are many techniques to make an image secure. In general sense todays digital world where nothing is secure, the security of images over network is very important. In this paper, we have surveyed different image encryption techniques from different research papers. We conclude that all techniques are good for image encryption and have their own advantages and disadvantages and they also gives better security at their level so that no unauthorized access can be done on images, which is in the open network. Each technique has its own suitability and its own limitations. But still lot more has to be done in this context.

REFERENCES
1. M.Zeghid, M.Machhout, L.khriji, A. Baganne, and R.Tourki, A modified AES based algorithm for image encryption2007 2. K. Gaj, p.chodowiec,Fast implementation and fair comparison of the final candidates for advance encryption standard using field programmable gate arrays, in: CT-RSA2001, pp, 8499. 3. A.Hodjat, I .VerbauwhedeA21.54 B bits/s fully pipelined AES processor on FPGA. Proceeding of the 12th annual IEEE symposium on field programmable custom computing machines (FCCM04). 4. K.Janvinen, M.Tominisko, j, Skytta,A fully pipelined memorys 17,8Gpbs AES-128 encryptior, in international symposium of field programmable gate arrays, 2003, pp, 207-215. 5. M.Mclone, J.V.McCanny,Rijindael FPGA implementations utilizing look-up tables, J.VLSI signal process, sits, 34(3)(2003)261-275. 6. S.K.Panigrahy, B.Acharya, D.jena, Image Encryption using self-invertible key matrix of Hill cipher algorithm2008. 7. S.P.Nanavati, P.K. Panigrahi. Wavelets: applications to image compression-I, Joined of the scientific and engineering computing, vol.9, no.3, 2004, pp.4-10. 8. AL.Vitali, A. Borneo, M.Fumagalli and R. RINALDO, VIDEO over IP standard- compatible multiple description coding, journals of Zhejiang university- science A, vol. 7, no.5, 2006, pp 668-676. 9. M.aliBani, younes, and a.jantan, Image Encryption using block based Transformation Algorithm 2008

10. M.aliBani, younes1+, and A.jantan2++, an image encryption Approach using a combination of permutation technique followed by Encryption 2008 11. Seyedzade, S.M, Atani, RE., Mirza Kuchaki, S, Novel image encryption algorithm based on hash function2010.

74

Ravi Shanker Yadav, MHD. Rizwan Beg & Manish Madhava Tripathi

12. I.A.Ismail, M.A, and H.Diab Digital image encryption algorithm based on composition of two chaotic logistic maps2010. 13. K.S.H, Shankeria.R, m.and Rahmani M.New modified version of advance encryption standard based algorithm for image encryption, 2010. 14. A.Nag, J.P.SINGH, S.Khan, S.Ghosh, S.Viswas, Image encryption using affine transformed XOR operation, 2011. 15. S. P. Indrakanti, P. S. Avadhani, permutation based image encryption technique, 2011. 16. R. Enayatifkr and A. H. Abdullah, image Security via genetic algorithm, 2011. 17. K. Singh, K. Kaur, image encryption using chaotic match and DNA addition operation and noise effect on it, 2011. 18. Q.H.Alsafasfeh, and, A.A.Arfoa, Image encryption based on the general approach for multiple chaotic system, 2011. 19. I. S. I. Abuhaiba and M. A. S. Hassan, Image encryption using differential evolution approach in security Domain, 2011. 20. T. Shah, I. H, M.A Gondal and H. Mahmood, Statistical analysis of S box in image encryption application based on majority logic criterion, 2011. 21. A.B.Abugharsa, A.s.b.h.Basari, H.Almangush, A new image encryption Approach using the integration of A Shifting Technique and the AES algorithm 2012 22. N.K.Pareek, Design and Analysis of a novel digital image encryption scheme 2012 23. Ankita Agarwal, secret key encryption algorithm using genetic algorithm2012 24. Varsha Bhatt, Implementation of new advance image encryption algorithm to enhance security of multimedia component2012. 25. www.ieee.org 26. www.google.com 27. http://en.wikipedia.org