New High Capacity Secure Steganography Technique

Transcription

1 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - New High Capacity Secure Steganography Technique Nawar S. Al-Seelawi Tarik Z. Ismaeel YamaanE. Majeed M.Sc. Student Dept. Computer Engineering University Baghdad Faculty member Dept. Electrical Engineering University Baghdad Lecturar Dept. Electrical Engineering University Baghdad nawar.alseelawi@gmail.com tarikismaeel@yahoo.com yamaan.majeed@coeng.uobaghdad.edu.iq Abstract With the spread digital data around the world through the internet the security the data has raised a concern to the people. Many methods are coming up to protect the data from going into the hands the unauthorized person. Steganography and cryptography are two different techniques for data security. The main purpose in cryptography is to make message concept unintelligible while steganography aims to hide secret message. Digital images are excellent carriers hidden information. In this paper a new steganography approach proposed based on LSB technique by using Alpha channel on JPG cover images and Bit-slicing decomposition and AES ( Advanced Encryption Standard) on the secrete image. for this method first the secrete image decomposed to bit streams and the data encrypted using AES algorithm. On the cover side an alpha channel is attached to the cover image and the data embedded into LSBs RGBA channels. The method was implemented and tested by using MATLAB (Ra). Keywords LSB AES Alpha channel RGBA Bitslicing I. INTRODUCTION Steganography is an ancient art that has been reborn in recent years. The word Steganography comes from Greek roots which literally means "covered writing" and is usually interpreted to mean hiding information in between other information []. A steganography system is expected to meet three key requirements namely transparency capacity and robustness. Transparency evaluates the image distortion due to signal modifications like message embedding or attacking. Capacity: It is the maximum amount information that a data hiding scheme can successfully embed without introducing any perceptual distortion in the marked media. Robustness measures the ability embedded data or watermark to withstand against intentional and unintentional attacks.[] Steganographic methods can be broadly classified based on the embedding domain digital steganography techniques are classified into (i) spatial domain (ii) frequency domain. In Spatial domain image steganography cover image is first decomposed in to its bits planes and then LSB s ( Least Significant Bits ) the bits planes are replaced with the secret data bits. As LSB s are redundant bits and contributes very less to overall appearance the pixel replacing it has no perceptible effect on the cover-image. Advantages are high embedding capacity ease implementation and imperceptibility hidden data. The major drawback is its vulnerability to various simple statistical analysis methods. The most direct way to represent pixel's color is by giving an ordered triple numbers: red (R) green (G) and blue (B) that comprises that particular color. The other way is to use a table known as palette to store the triples and use a reference into the table for each pixel. For transparent images extra channel called the Alpha value is stored along with the RGB channels. RGBA image stands for Red Green Blue Alpha. It extends the RGB color model with the alpha value representing the transparency pixels. The A value varies from to in which means completely transparent while means opaque. PNG images follow the RGBA color model []. Bit-plane Slicing decomposition highlighting the contribution made to the total image appearance by specific bits. Assuming that each pixel is represented by -bits the image is composed eight -bit planes. Plane () contains the least significant bit and plane () contains the most significant bit. Only the higher order bits (top four) contain the majority visually significant data. The other bit planes contribute the more subtle details []. After holding the contest for three years NIST chose an algorithm created by two Belgian computer scientists Vincent Rijmen and Joan Daemen. They named their algorithm Rijndael after themselves []. On November the Federal Information Processing Standards Publication ( FIPS PUB ) announced a standardized form the Rijndael algorithm as the new standard for Page

2 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - encryption. This standard was called Advanced Encryption Standard and is currently still the standard for encryption []. To date there are no attacks better than brute-force known against AES [ ]. Like DES AES is a symmetric block cipher. This means that it uses the same key for both encryption and decryption. However AES is quite different from DES in a number ways. The algorithm Rijndael allows for a variety block and key sizes and not just the and bits DES block and key size. The block and key can in fact be chosen independently from bits and need not be the same. However the AES standard states that the algorithm can only accept a block size bits and a choice three keys ( ) bits. A number AES parameters depend on the key length. For example if the key size used is then the number rounds is whereas it is and for and bits respectively. At present the most common key size likely to be used is the bit key []. The overall structure AES can be seen in Fig.. The input is a single bit block both for decryption and encryption and is known as the in matrix. This block is copied into a state array which is modified at each stage the algorithm and then copied to an output matrix. Both the plaintext and key are depicted as a bit square matrix bytes. This key is then expanded into an array key schedule words (the w matrix). It must be noted that the ordering bytes within the in matrix is by column. The same applies to the w matrix The algorithm begins with an ( Add round key ) stage followed by rounds four stages and a tenth round three stages. This applies for both encryption and decryption with the exception that each stage a round the decryption algorithm is the inverse it s counterpart in the encryption algorithm. The four stages are as follows [] :. Substitute bytes. Shift rows. Mix Columns. Add Round Key The tenth round simply leaves out the ( Mix Columns ) stage. The first nine rounds the decryption algorithm consist the following :. Inverse Shift rows. Inverse Substitute bytes. Inverse Add Round Key. Inverse Mix Columns Again the tenth round simply leaves out the ( Inverse Mix Columns ) stage. There are many researches in each the steganography techniques and a brief description some these research are presented: For the researches which are presented the high capacity steganography methods are []. In this work an alpha channel is attached to an image with RGB ( Red Green Blue ) color system ( bits depth ) the resulting image is a PNG (Portable Network Graphics ) image with RGBA color system ( bits depth ) on the other hand using Bit-plane Slicing decomposition on the secrete image to compress it and transform the gray-level secrete image to a binary bit stream the AES encryption occurs in the middile decomposition process. The data then embedded in the cover image four color planes. II. THE PROPOSED TECHNIQUE While most steganography techniques work on cover image or secrete image our proposed technique relies on processing both cover and secrete image to reach to the optimum results. For the secrete image side the total data size is decreased i.e. compressing the image to decrease the amount the payload. Bit-plane slicing technique used to compress the secrete image and also to convert it from D image to D bit stream. On the other side working on the cover image to increase its ability to handle the payload. A fourth channel added to the JPG cover image to increase the bit depth from to and to be four channels carrying the four candidate bit-planes. The proposed system is presented in the Fig. for the sender side and Fig. for the receiver side. Fig. The overall structure AES algorithm Page

3 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - The Embedding Sequence (Case ) (Case ) Fig. The resulted image if the value Alpha is (Case ) all ones (Case ) all zeros Fig. The Main block diagram the sender side In this work the Alpha chose to be all ones and this mean alpha channel will be a white plane acts as a transparent background the image. B. Preparation and Decomposition the Secret Image ) The Embedding Sequence Fig. The Main block diagram the reciever side A. Preparation the Cover Image Given cover image is a color image. Let A be an original color image having size represented as : < < < A ( ) () ( ) {.. } Value K varies from to This image has the extension JPG. It has color channels ( Red Blue Green ). To add the fourth channel it must be defined : < < ℎ ( ) () ( ) The size alpha channel is exactly same to that cover color image. There are two cases selecting the value Alpha channel : Case : the image is a color image having size * and alpha channel comprises all ones then output will be a full color image. Case : the image is a color image having size * and alpha channel comprises all zeros then output will be a gray image. ) Fig. depicts the two cases choosing the value Alpha channel. Grayscale Secret Image: Let is a secrete grayscale image having size represented as < < ( ) () ( ) {.. } This D secrete grayscale image is first passed through bit-plane slicing algorithm. For grayscale image having bit-planes this can be represented as follows : < < ( ) () ( ) {} Where : k The th bit-plane contain more information than other plane then for embedding process could only choose th th th and th bit-plane. To convert all the selected bit-planes into D array as shown below : ( ) and represented as follows for each upper bit-planes : <( ) ( )...() ( ) {} to combine all four strings into D binary secrete array : [ ] () Secrete D array then divided into parts. Then by finding the length the and the length each divided string. Suppose this length is : / () + () + + () The content the st secrete string is : (: ) () sec sec The content the nd secrete string is : ( + : ) () sec sec The content the rd secrete string is : ( + : ) () sec sec And the content the th secrete string is : ( + : ) sec sec () Page

4 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - ) Colour Secret Image: Let is a secrete color image having size represented as : < < < ( ) () ( ) {.. } The secrete image is an RGB image with bitdepth which means it has bit-planes ( bit-planes for each color channel r g and b ). This representation can be treated as grayscale images by extracting Red channel Green channel and Blue channel separately and apply BitPlane slicing on each the three channels where : ( ) ( ) ( ) () where is the bit byte with the value {} that is ( ) (). The prime ( ) indicates that the variable is to be updated by the value on the right. The AES standard depicts this transformation in matrix form as follows: channel ( : : ); channel ( : : ); () channel ( : : ); ( Using the formula () to apply bit-plane slicing : Bitplane( Bitplane( ]; () Bitplane( [ )[ )[ The S-box is constructed in the following fashion :. Initialize the S-box with the byte values in ascending order row by row. Thus thevalue the byte at row x column y is {xy}.. Map each byte in the S-box to its multiplicative inverse in the finite field GF( ) the value {} is mapped to itself.. Consider that each byte in the S-box consists bits labeled ( ). Apply the affine transformation to each bit each byte in the S-box: ) ]; ]; Choosing the upper planes from each channel ( for embedding process ) then converting the selected D bit-planes into D array as in formula () and combine all arrays into D binary secrete array : [ The D array is divided into parts preparing to embed in the cover channels. C. Encryption the Secret Image AES encryption used to give maximum security to the system where AES is the most secure encryption algorithm nowadays. In this method encryption is applied on the secrete image after taking the upper bit-planes that selected from bit-plane slicing process. The D is encrypted before it converted to D array. The key used is bits and number rounds. Fig. illustrate the encryption process. ) + () Each element in the product matrix is the bitwise ] XOR elements one row and one column. Further the final addition shown in equation () is a bitwise XOR the inverse S-box is obtained by taking the inverse equation (.) affine transformation followed by taking the multiplicative inverse in GF( ). As an example consider the input value {}. The multiplicative inverse in GF( ) is {} {a} which is in binary. Using equation (.) the result is {A} Fig. shows the secrete image and the encrypted secrete image : Fig. the secrete image and the encrypted secrete image Fig. The encryption secrete image using AES algorithm D. Proposed Embedding Algorithm In this algorithm a secret image will be hidden in a cover image using LSB. Embedding operation is variable. The number LSB bits used to embed could be vary from Page

5 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - bit to bits. These numbers is used to add more security to the system because the receiver cannot extract the secrete image without knowing the number bits each channel used for embedding. This embedding sequence is chosen by the sender. An example the embedding sequence is in the Fig.. The Design flow hiding process is shown in Fig. Fig. Example embedding sequence The Embedding procedure : - The secrete image is decomposed using Bit-plane slicing to bit-planes if it is a gray scale image and bit-planes if it is a color image. - Encrypt the selected bit-planes directly using AES algorithm. - Convert the encrypted bit-planes into D arrays. - Extract Red Green and Blue planes form cover image and define the Alpha channel. - embed into Alpha channel. suppose the number LSB bits can be used for embedding is N If ( N ) (Embed N bits into each and every pixel Alpha channel till message is not finished) End - embed into Blue channel. If ( N ) (Embed N bits into each and every pixel Blue plane till message is not finished) End - embed into Green channel. If ( N ) (Embed N bits into each and every pixel Green plane till message is not finished) End - embed into Red channel. If ( N ) (Embed N bits into each and every pixel Red plane till message is not finished) Fig. Design flow hiding process for AES model The Extraction Procedure :. Extract Alpha channel and Red Green and Blue plane from RGBA stego image.. Use the embedding sequence to extract the bit strings from each plane the image. The original cover image is produced in this stage.. Combine all these bit-planes into one image to find the recovered encrypted secrete image using equation (.).. Decrypt the encrypted secrete image ( ) using AES decryption process to recover the secrete image. The Design flow the extraction process is shown in Fig. End Page

6 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - [ ( ) ( )] The Normalized-Cross Correlation is given by : [ ( ) ( )] [ ( ) ( )] Average Difference an image is given by : [( ) ( )] Large value indicates that image has poor quality. For test the proposed method a personal selfie image used as an original cover image with size * bit depth and color system RGB. Different images with different color spaces ( grayscale and color images ) and different sizes are chosen as secrete images and different amount bits are embedded in each channel RGBA cover image. ) Grayscale Secret Image: In the test given a code name AES-G An image ( S rocket system ) chosen as a secrete image with gray- level * size and bit depth. The results are shown in the Table I : Fig. The Design flow the extraction process using AES model III. RESULTS AND DISCUSSION A. Results A series experiments have been conducted to show the effectiveness the proposed technique. The efficiency the proposed technique is measured by Five metrics which are: (Peak Signal-to-Noise Ratio) ( Mean Square Error ) Normalized Cross Correlation ( Average Difference ) Histogram Analysis is usually measured in db and given by : ( ) [ ( ) ( )] Where: N: height the two images (because the two images must be the same size) M: width the two images i and j : row and column numbers L: is the number the gray scale levels in the two images C(ij): is the original image. St(ij): is the stego image. Typical values range between and db []. Where shows the mean square error between cover image C and stego image S. Table IThe results embedding ( * ) gray-level secrete image into (*) RGB cover image AES model No. bit s Bits per channel R G B A Fig. shows the original image secrete image and stegoimage : Page

7 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - %. Table III shows the results using a secrete image with the same size the cover image. Table III The results embedding ( * ) gray-level secrete image into (*) RGB cover image AES model Fig. Original image secrete image stego-image AESG The second test AES-G- is a secret image with gray- level * size and bit depth. The results are shown in the Table II : R G B A.. Bits per channel Fig. shows the original image secrete image the stegoimage obtained from embedding process : Fig. Original image secrete image stego-image AESG- In the test AES-G an image a military site gray-scale image with the same size cover image ( * ) is chosen to test the system at the capacity Table II The results embedding ( * ) gray-level secrete image into (*) RGB cover image AES model No. bit s No. bit s Bits per channel R G B A Fig. shows the original image secrete image stegoimage Fig. Original image secrete image stego-image AESG ) Colour Secret Image: In the this test which given a code name AESR an image a secret weapon is used as a secret image with size * bit-depth with color space RGB. Table IV shows the results obtained from embedding the secret image into the cover image. Page

8 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - Table IV The results embedding ( * ) color secrete image into (*) RGB cover image AES model No. bit s Bits per channel R G B A Fig. shows the original image secrete image and the stego-image Fig. shows the original image secrete image and stegoimage. Fig. Original image secrete image stego-image AESR- For testing the system in the capacity % in the test AES-R a color secret image is used * size ( the same size the cover image ). Table VI shows the result obtained from this test Table VI The results embedding ( * ) color secrete image into (*) RGB cover image AES model No. bit s Fig. Original image secrete image stego-image AESR For testing the system in the capacity % in the test AES-R- a color secret image is used * size. Table V shows the result obtained from this test Table V The results embedding ( * ) color secrete image into (*) RGB cover image AES model No. bit s R G B A Bits per channel R G B A Fig. shows the original image secrete image and the stego-image obtained from the embedding process.... Bits per channel Page

9 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - Table VII comparison capacity between the proposed method and other methods Fig. Original image secrete image stego-image AESR B. Discussion According to readings above the test results can be discussed from the viewpoint steganography three key requirement ( Capacity Invisibility and security ) ) The capacity the proposed system: The capacity is the size the data in a cover image that can be modified without deteriorating the integrity the cover image. The steganographic embedding operation needs to preserve the statistical and perceptual quality the cover image. Capacity is represented by the maximum number bits can be embedded in the cover image without degrade the stego image quality and the Maximum Hiding Capacity in terms percentage. The size the hidden information relative to the size the cover image is known as embedding rate or capacity []. Capacity the proposed system is calculated by the following formula : ( ) ( ) % () Metho d [] [] [] [] No. bits The embedding capacity in term percentage is compared with the capacity reached by ref. [ ]. [] used Chaotic map and Contourlet transform for embedding. The comparison is listed in table VIII : Table VIII comparison capacity percentage between the proposed method and other methods Method [] [] [] [] propos ed method Capacity %. Where: ( ) : the number pixels secret image ( ) : the number pixels cover image : No. bits in each secret image pixels N : No. cover bits used for embedding To test the performance the proposed system a comparison between the proposed method and ref. [ ] in term maximum embedded bits has been shown in Table VII and from the results below it can be seen that the capacity the proposed system has majorly improved. For this purpose a cover image is selected to be the (lena) image ( ) see Fig.. [] used LSB substitution and RGBA color space for cover imageref. [] is a steganography based on spatial domain and LSB substitution on PNG cover image using Shamir method for secret sharing to embed in alpha channelref. [] used the Reflected Binary Gray Code RBGC in the wavelet domain and ref. [] used BPCS in frequency domain. Propos ed Method Fig. The cover image ( Lena ) The capacity % is easy to define by ( the ability system to embed a secret data the same size the cover image ). According to this definition a test designed to represent a real % capacity this system using the same image for cover and secret image. While the two images ( cover and secret ) are identical the successful embedding and extracting with low distortion means the system has % capacity. Table IX shows the results objective test and Fig. shows the results subjective tests Table IX The results embedding ( * ) color secrete image the same image No. bits used.... Page

10 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - The results shows that in the capacity % the distortion the stego image become noticeable to the human eyes by increasing the number bits the cover image used to embed. Fig. illustrate the stego image by using different amount cover bits to embed. Fig. The stego images obtained from embedding process with changing the number bits used to embed Fig. The cover image the secret image the stego image and the recovered secret image The comparison above is done using only bits from cover image to embed while using more cover image bits increases the capacity where : ( ) () Where : ( ) : the number pixels cover image N : the number bits the cover image used to embed. ) The Invisibility the proposed system: In this method message is hidden in the least significant bits image pixels. Changing the LSB the pixels does not introduce much difference in the image. The secret data hidden in the LSBs the cover image channels (Red Green Blue Alpha). The use Alpha channel gives the advantages increasing the cover capacity as well as acts as a transparent mask that can handle a part secret data with very high efficiency and very low distortion the stego image. The tests above done using the channels the cover image with various channels LSB changing to use the maximum capacity that the cover can handle although the increasing LSBs used increases the distortion the stego image as a trade-f. The subjective test the stego image is very important test to show the strength the algorithm while visual attacks making use the ability human eyes to clearly discern between noise and visual patterns. Hence the presence the secret information must be invisible to the human eyes first. Methods or techniques that can be used to evaluate the undetectability or imperceptibility steganographic systems are different from one system to another depending on the type cover file used for information hiding. For example image quality represents an indication for the undetectability image based steganography while file size may reveal the presence hidden data within a text file and therefore lead to its detection. Two types perceptibility can be distinguished and evaluated in signal processing systems namely fidelity and quality. Fidelity means the perceptual similarity between signals before and after processing. However quality is an absolute measure the goodness a signal. For example a grayscale can be used distorted and low resolution image (considered to be low quality) for data hiding. The stego image looks identical to the cover image but it is also has low quality. However because it is indistinguishable from the cover image it has high fidelity. For image based steganography the fidelity is defined as the perceptual similarity between the original cover image and the stego image. Therefore the fidelity evaluation requires both versions the image before and after embedding. However attackers and most likely recipients do not have access to the unmodified original cover image. Additionally steganographic systems must avoid attracting the attention anyone not involved in the secret communication process and therefore stego images must have very good quality. Therefore quality is the major perceptual concern for most steganography techniques in order to avoid any suspension and therefore detection. Even though the and the mean square error () are by definition fidelity metrics they are pervasively known as quality measures Page

11 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - since they also represent perceptual distance metrics used to measure the distortion amount added to an image. Accordingly a high quality image entails a large value and therefore both cover image and stego image are very similar and quite undistinguishable. Significantly Fidelity is defined as the perceptual quality stego files and therefore and describe how imperceptible the secret message is []. Accordingly the higher the quality stego images the larger the imperceptibility the steganographic system. Therefore evaluating the quality stego images is a significant measure to be used for evaluating the performance image steganography techniques []. To evaluate the performance the proposed system in term invisibility a comparison between the proposed method (with the two encryption methods ) and ref. [ ] has been shown in Table X. For this purpose a cover image is selected to be the ( lena ) image ( ) and hiding capacity is %. [] used BPCS to palette-based image ref. [] is a steganography based on integer wavelet domain ref. [] used the Intermediate Significant Bit Planes. TableX The comparison between the proposed system and other methods Met hod [] [] [] [] [] [] PSN R Pro pose d. ) The Security the proposed system: The embedding algorithm is said to be secure if the embedded information cannot be removed beyond reliable detection by targeted attacks based on a full knowledge the embedding algorithm and to detector and the knowledge at least one carrier with hidden message. According to that the security the proposed system is the security the encryption algorithm used. In fact there are no possible attack on AES better than brute-force attack. Assuming a computer that try keys at the rate one billion keys per second. Under this assumption the attacker will need about ( billions trillions ) years to try all possible keys for the version AES- []. The way that the AES encrypt the secret image also gives a degree invisibility and robustness beside the high degree security where the extracted encrypted image ( in the case detection and successful extraction ) is look like a random image ( or insignificant data ) while there is no sign that the extracted image is actually a secret encrypted image. The attacker cannot distinguish between the encrypted image and any randomized pixel's values image. Figure shows an encrypted secret image which is it not a thing only a random pixels and cannot say it is a secret image or just a random pieces from the attacked image ( stego image). Fig. Encrypted secret image IV. CONCLUSIONS. In this work a new data hiding technique presented that allows hiding a color image (secret object) in another color image (cover object) where both images might be same size or bigger therefore achieving up to % embedding capacity.. The use an encryption algorithm and the capability to control in the number cover image bits used in embed increases the security the system and adds the robustness factor as a third advantage the system beside the high capacity and the undetectability. Using AES The use AES algorithm as an encryption method level up the security the system to high degree while this algorithm is the current standard algorithm for encryption there is no possible attack on AES that is it is impossible to decrypt the embedded secret image ( in the case detection ).. The stego image is very close to cover image in both objective and subjective tests. Statistical results shows that the system has high invisibility.. Using Bit-Slicing technique compresses the secrete image and this results in decreasing the total amount data embedded.. Attaching the alpha channel to the RGB image increases the bit depth the image and this results in increasing the embedding range.. As the results shows Alpha channel can handle more bits than the other channels while maintain a good considering that the Alpha channel is the lowest byte the RGBA pixel. ACKNOWLEDGMENT We wish to acknowledge H.O.D and staff Electronics and Communication engineering department and H.O.D and staff Computer Engineering department our college for their kind support for this project. We also thank our project guide and co-guide for highlighting our path and their gracious guidance. In last we like to thank all the friends who had given some valuable contribution for this work. REFERENCES [] S. A.Laskar and K. Hemachandran" Secure Data Transmission Using Steganography and Encryption Page

12 International Journal Research in Computer and ISSN (Online) - Communication Technology Vol Issue January- ISSN (Print) - [] [] [] [] [] [] [] [] [] [] [] [] [] Technique " International Journal on Cryptography and Information Security (IJCIS) Vol. No. September. V. K. Mann and H. S. Dhaliwal " Colour Image Steganography " International Journal Engineering Trends and Technology (IJETT) Volume Issue August. S. Sharma and U. Kumari " A High Capacity DataHiding Technique Using Steganography " International Journal Emerging Trends &Technology in Computer Science (IJETTCS) Volume Issue May June. R. C. Gonzalez and R. E. Woods " Digital Image Processing " rd edition Prentice Hall Upper Saddle River. Announcing the Advanced Encryption Standard (AES) Federal Information Processing Standards Publication November. C. Paar and J. Pelzl " Understanding Cryptography : A textbook For Students and Practitioners " Springer. W. Stallings " Cryptography and Network Security Principles and Practice " th edition Pearson Education Inc.. K. Kamalam and S. Saranya " An Effective Method in Steganography to Improve Protection Using Advanced Encryption Standard Algorithm " International Journal Engineering Trends and Technology (IJETT) Volume Number Dec. H. Sajedi and M. Jamzad" Secure steganography based on embedding capacity" Springer-Verlag International Journal Information Security volume Issue. R. English " Comparison High Capacity Steganography Techniques " IEEE International Conference St Computing and Pattern Recognition S. K. Muttoo and Sushil Kumar" A Multilayered Secure Robust and High Capacity Image Steganographic Algorithm " IEEE rd International Conference on Communication Systems Stware and Middleware and Workshops COMSWARE. C. W. Lee and W. H. Tsai " A New Steganographic Method Based on Information Sharing via PNG Images " IEEE. A. Cheddad " Steganlage : A New Image Steganography Algorithm " PhD Thesis School Computing & Intelligent Systems Faculty Computing & Engineering University Ulster September. S. Khaire et. al. " Review : Steganography Bit Plane Complexity Segmentation (BPCS) Technique " International Journal Engineering Science and Technology Volume Issue. [] A. J. Sadiq " Comparison Steganography in spatial domain Image " Journal Baghdad College Economic Sciences No. Baghdad. [] S. Venkatraman A. Abraham and M. Paprzycki " Significance Steganography on Data Security " The International Conference on Information Technology : Coding and Computing ITCC. [] Z. N. Abdulhameed and M. K. Mahmood " High Capacity Steganography based on Chaos and Contourlet Transform for Hiding Multimedia Data " International Journal Electronics and Communication Engineering & Technology (IJECET) Volume Issue January. [] I. J. Cox et.al. " Digital Watermarking and Steganography " Second Edition Burlington MA USA Elsevier Inc.. [] N. I. Wu and M. S. Hwang " Data Hiding : Current Status and Key Issues " International Journal Network Security Vol. No. January. [] M. Niimi H. Noda E. Kawaguchi and R. O. Eason " High Capacity and Secure Digital Steganography to Palette-Based Images " IEEE. [] S. A. Parah and J. A. Sheikh " Data Hiding in Intermediate Significant Bit Planes A High Capacity Blind Steganographic Technique " IEEE International Conference on Emerging Trends in Science Engineering and Technology ----/. [] K. B. Raja S. Sindhu and T. D. Mahalakshm " Robust Image Adaptive Steganography using Integer Wavelets " IEEE Image Processing. [] A.W. Naji et al. " Novel Framework for Hidden Data in the Image Page within Executable File Using Computation Between Advance Encryption Standard and Distortion Techniques " International Journal Computer Science and Information Security (IJCSIS) Volume No. Aug Page