### 1. Introduction

### 1.1 STEGANOGRAPHY

From a decennary, the western universe is sing the incorporation of the computing machine into civilization. We are frequently considered as information society. The easy manner for communicating and informations sharing is by agencies of computing machine and web engineering. Data which contain information often has some redundancy, which is known as noise. Noise is enclosed because optimum compaction being is really rare and common compress ratio is the issue of efficiency. Steganography is the art and scientific discipline of implanting information to data ‘s noise. Such description is really general and it is better to believe approximately Steganography as a manner of concealing secrets in guiltless looking messages. These messages are referred to as screen text. After the secret is inserted we call them stegotext, military literature besides uses Steganography and is referred to as broadcast security. Here comes the chief difference between Steganography and Cryptography, Cryptography is about hiding the content of the message ; Steganography is about hiding their being. On the other manus, a good cryptography system should be designed in such a manner that everything is unafraid and does non cognize the being of message and a secret key is provided. Depending on the manner of encoding the Steganography is classified into authoritative or historic. Steganography has many existent life applications. Those applications are divided into two major classs. First class falls into the same category as cryptanalysis and takes topographic point in a unafraid passage of sensitive informations. In many instances this class is known as systems with a inactive antagonist. The purpose of an antagonist is merely to comprehend by seeing if any secret communicating takes topographic point. The implanting algorithm strives hard for high redundant informations security and capacity. Unfortunately, the demand for security and capacity entails that the concealed information is delicate. Second class comes under the portion of digital watermarking and fingerprinting. Watermarking is the procedure of as implanting the tantamount digital cast to some peculiar goods to demo its beginning and so right of first publication. On the other manus fingerprinting engineering produces goods with consecutive Numberss so that they can be tracked. This class can be called as systems with an active antagonist. As it is normally good known that some object is marked, the antagonist ‘s undertaking is to take the grade by actively pull stringsing the object ( e.g. cropping resampling and change overing ) . Here we strive to accomplish high hardiness. Robustness refers to the sum of alteration the stegotext can keep up. Relationship among security, capacity and hardiness are such that they create vertices of equilateral trigon.

### 1.2 DEFINITION OF CRYPTOGRAPHY

Cryptography trades with acquisition of numerical methods associated with supplying security for the information such as confidentiality, informations unity, entity hallmark and informations beginning hallmark. There are many methods of supplying security to the information cryptanalysis is one among those [ 5 ] .

Cryptanalytic ends [ 5 ] :

1. Privacy or confidentiality: This is related to keep the informations in secret to everyone but can be accessed to those who are certified to see the information.

2. Data unity: Data unity is related to do certain that the information is non changed by any uncertified or by any unusual manner.

3. Authentication: Authentication is sing designation. When two groups have communication both groups have to acknowledge each other.

4. Non-repudiation: Non renunciation is based forestalling the entry of informations which has been done earlier.

### 1.3 DIFFERENCES BETWEEN STEGANOGRAPHY AND CRYPTOGRAPHY [ 6 ]

S.no

Cryptography

Cryptanalysis

1

In this the information is hidden within a different information and seems like an ordinary image.

It seems to look like nonmeaningful sentence or word when the encoding is performed.

2

When all the stegno images are collected and placed at one topographic point they do n’t look to be leery.

When all the encrypted information is collected and placed at one topographic point they are likely to be leery.

3

An adept stegnologist may place when there is immediate alteration in the informations format but can non observe the content of the concealed informations.

An adept cryptanalyst will place the communicating that is encrypted cryptographically from the information which is been passed.

4

When recycling the images which are been one time encrypted demands attention to utilize those images.

Careful attending is necessary for recycling the keys.

5.

Steganography has no legal regulation that stops utilizing this.

Cryptanalysis has legal regulations that ban utilizing this.

### 1.4 CLASSIFICATION OF STEGANOGRAPHY

Cryptography is classified based on ages. They are

1. Ancient Cryptography

2. Medieval Cryptography

3. Modern Cryptography

### 1.4.1 ANCIENT STEGANOGRAPHY

From the clip when adult male started pass oning over written messages, the necessary for secretiveness was in high bid. In the earlier period, messages could easy be read by the 3rd party because there were no secret devices. A alteration was taken topographic point during the clip of Greeks which was around 500BC. Demaratus was the first to utilize the technique of Steganography. The word Steganography is taken from two Grecian words steganos, intending “ Covered” and grahein, intending “to write” . Steganography is the art of concealing a message so that it seems like a message does non exists at all.

An old illustration of first cryptography is that Demaratus saved his state, Greece communicated to the fellow members by messaging sing the onslaught of the Persian King by utilizing Wax Tables. He encrypted the message which was to be passed on a tabular array and was covered equally with wax coating, making a false feeling that no message is being transmitted, and this lead to the victory of Greece. This was the first identified instance of the usage of Steganography and from so, the complexness of Steganography has started.

### 1.4.2 MEDIVAL STEGANOGRAPHY

World-War II gave a new facet to Steganography by contriving two new Steganography techniques. They were

1. Micro-Dot engineering

2. Open codification messaging

### MICRO-DOT TECHNOLOGY

The micro-Dot is basically a immensely elaborate image is being shirked to the size of a period or DOT. We are capable of lading a whole page of text or a whole image into small Micro-Dot so successfully, and once more the image can be seen by either blowing up the Dot or by the usage of a microscope.

### OPEN CODE MESSAGING

Open coded message usage regular words and messages to compose the buffer text that hides the message because they seem to be normal and frequently pass security cheques.

### 1.4.3 MODERN STEGANOGRAPHY

Staganography was modernized by the addition in the computing machine universe. The present twenty-four hours universe is utilizing the digital cryptography which includes a broad assortment of Fieldss like medical specialty, armed forces, peer-to-peer communicating, etc. [ 9 ] .

One of the latest application of cryptography is to hedge mishandling of patient ‘s records in infirmary by cryptography the specifics of the patient in a digital image, in such a manner that it is accessible to merely few of import individuals like physician and nurse will be cognizing the watchword, so that the information is non handled by the other individual.

The armed forces during war are utilizing techniques like voice cryptography, graphical cryptography because security of information is of chief of import at that clip.

At present twenty-four hours, with rapid growing of computing machine engineering it is much easier to conceal messages and on the other manner it is hard to happen out that message. As the engineering, is increasing exponentially, the high demand or growing for cryptography seem vast. The undertaking is focused on the applications that are derived from the usage of computing machine.

### 1.5 TYPES OF MODERN STEGANOGRAPHY

Computer engineering has given a new form to Steganography. Steganography is most normally used in three different formats. They are

1. Text format

2. Audio format

3. Graphic format

### 1.5.1 TEXT FORMAT

Textual cryptography follows nearer to the basic significance as it a concealed communicating by agencies of authorship.

### 1.5.2 AUDIO FORMAT

Auditory cryptography uses the conversation or music to transport or convey the born-again messages.

### 1.5.3 GRAPHICAL FORMAT

Graphic cryptography is the procedure of concealing messages in picture or in graphical images most normally be electronic media or by publishing.

### 1.6 GRAPHICAL STEGANOGRAPHY

A Graphical cryptography is which the symbols are drawn. There are computing machine plans in which the plans are written to suggest a digital image and conceal the message in that. With the addition in the cyberspace use, a web browser which is text based is moved to the graphical interface of the web. Now a twenty-four hours the higher cyberspace broadband velocity is available to the common adult male utilizing in the house, the usage of artworks and picture have become most common. All of these effects have built a rich environment in which cryptography can be used to convey proprietary corporate information, secret military programs or even tools for computing machine felons. The file formats used for this type of cryptography are JPG, GIF, BMP, AVI and MPEG [ 4 ] .

### 1.7 BASIC STRUCTURE OF STEGANOGRAPHIC SYSTEM

The basic construction of stegographic system consists of four objects which can be described as [ 1,2 ]

1. Cover object ( bearer image ) -which act as a vas for transporting or reassigning the secret information ( making a false cogent evidence ) .

2. Information to be hidden- secret information, which is to be transmitted.

3. Steganography algorithm to code the secret information in to the screen object.

4. Stego Key- for keeping privateness.

### 1.8 STEGANOGRAPHY IN COMMUNICATION

Hiding the information within an image is most common process used at present. Through the cyberspace secret messages can be spread easy by infixing the secret message in an image. To hide a message inside an image without changing its seeable belongingss, “noisy” countries which have colour alterations can be changed for the screen image.The general technique used to do the alterations engaged is the use of the Least-Significant Bit ( LSB ) , dissembling, filtrating and transmutations on the screen image.

### 1.9 BLOCKADE IN STEGANOGRAPHY

Two types of onslaughts are discussed for the procedure for implanting the informations and they work strongly. Ocular onslaught is the first sort of onslaught which makes the aptitude of worlds equivocally distinguish between noise and ocular forms. Statistical onslaughts are 2nd sort of onslaughts which make it harder to computerise.

### 1.10 STEGANOGRAPHY BEFORE

If a 24 spot colour image is used, each pel can hive away three spots of colour constituents ruddy, green, and blue. When an image of 800 ? 600 pel is considered the image can keep an sum of 14, 40,000 spots of concealed informations. From that whole if merely three spots are used the whole bytes are being wasted [ 1, 2 ] .

### 1.11 PROPOSED STEGANOGRAPHY

The chief purpose of the undertaking is to infix more than one spot at each byte in one pel of the cover-image and obtain the consequences like the LSB. This purpose can be reached by developing a Steganography algorithm to conceal big sum of any type of information through electronic image image by utilizing maximal figure of spots per byte at each pel. Any type of informations can be hidden in a electronic image image which has 24- spots by utilizing the Steganography algorithm. The 24 spots has three bytes of RGB colourss, each byte has four spots called as Nibbles. The highest value is stored in the left nibble and the lowest value is contained in the right nybble in a byte [ 1, 2 ] .

### INPUT DATA

0

1

0

1

0

1

0

1

0

0

0

0

1

1

1

1

0

0

0

0

0

0

0

0

0

0

0

0

1

1

0

1

Red GREEN BLUE

0

0

0

0

1

1

1

0

0

0

0

0

0

0

0

1

0

0

0

0

1

1

0

1

### Red GREEN BLUE

The byte value can alter is there is a alteration in the right nibble. A byte can be efficaciously specified harmonizing to the left nibble value. The nibble value is fixed by the interval, in such a manner that 16 degrees of precedence are concluded [ 1 ] .

### 2. PREVIOUS STEGANOGRAPHY ALGORITHM

Steganography algorithm provides high security by utilizing the two theories but the new algorithm is designed with three beds which works separately and supply unbreakable security wall [ 1 ] .

Layer 1: The first bed of security done is Encryption mechanism bed by utilizing Advance Encryption criterion ( AES ) algorithm.

Layer 2: Image Cleavage is done by utilizing adaptative cleavage.

Layer 3: Random choice of appropriate byte of every pel of cover-image is selected based on colour features to implant secret informations.

Encoding of informations: The procedure of encoding is the first bed for security. Advanced Encryption criterion algorithm provides high encoding to the information that is to be encrypted [ 1 ] .

Adaptive cleavage of the cover-image: The 2nd bed of security is the adaptative cleavage. The image taken as the screen image is the bitmap image and that cover image is segmented indiscriminately of regular or irregular sections based on the watchword given or any other designation given by the proprietor. Irregular cleavage gives more security for giving the input. A technique known as lossless compaction is applied for the electronic image stego image for directing big files. To retrace the original image precisely lossless compaction technique is used [ 1 ] .

Pixel Selection Style: The 3rd bed of security is the pixel choice. The Cover-image pels are selected indiscriminately to put the proper byte at its matching pel to implant the secret informations based on the colour features of the cover-image. The choice of the pels is done by presenting the construct called chief instances and sub-cases, which strength the decrease of noise in a stego-image [ 1 ] .

### Block Diagram for the Steganography Algorithm

1. Datas concealing

2. Data extraction.

The transmitter hides the information and the receiving system extracts the information which is sent by the transmitter.

The two parts are developed and applied to fulfill the demands mentioned below:

1. The statistical sensing chances must be reduced.

2. Even though different image uses are performed robustness must be provided by the algorithm

3. Stegno image should non incorporate deformation artefacts.

4. Implanting capacity must non be given-up by the algorithm.

Hiding Data into bitmap image is based on the below process [ 1 ] :

1. Sender sends the encoding watchword.

2. The maximal figure of bytes that are taken by the screen image is calculated.

3. Compaction technique is performed on the file that is to be kept secret to increase the capacity of informations to be hidden.

4. Encoding is performed of the file to be hidden.

5. A bitmap image is taken as the screen image. For that screen image Adaptive cleavage and adaptative filtering is done.

a. Based on the watchword given adaptative cleavage is performed.

B. To minimise the sum of colourss of the screen image adaptative filtering is done by utilizing minimal average square mistake.

6. By taking the features of the image, choice of an disposed pel is done and that information is hidden at that selected pel.

7. The information is hidden based on the colour features of the screen image.

Extraction of the encrypted information at the receiver terminal [ 1 ] :

### The information encrypted is extracted on verification based on the process below:

1. The watchword given for encoding.

2. The pels are scanned based on watchword.

3. The information is extracted.

4. The extracted information is decrypted based on watchword.

Megahertz

The corresponding SC

Set Description of SC

1

{ SC3, SC5, SC6 }

{ ? SC ?C=SELcolor ? CP & A ; RCi ? CP ? I =1,2 ? C ? RCi, }

2-15

{ SC1, SC3, SC5, SC6 }

{ ?SC ?C=SELcolor ?CP and RCi ? CP? i=1,2 ?C ? RCi | C ? RCi }

16

{ SC1, SC3 }

{ ?SC? C=SELcolor ?CP and RCi ? CP? i= 1, 2 ?C ? RCi }

### Table 1: Megahertz with their corresponding SC [ 1 ]

### 3. EXPECTATION MAXIMIZATION SEGMENTATION

### Color image cleavage

Color image cleavage its really interesting and intensive subject in image processing.This can see as a extension of grey degree image cleavage, there are assorted methods which can be categorized they are:

1. Cluster based cleavage

2. Contour sensing based cleavage

3. Area extraction based cleavage

4. Probabilistic theoretical accounts based cleavage

There are different algorithms which are used to make the image cleavage.There are few on a regular basis used cleavage they are:

1. Average displacement algorithm

2. Markov random field theoretical account

3. Gaussian mixture theoretical accounts ( GMM ) based on Extraction Maximization ( EM ) .

4. GMM ‘s kernel version K-EM.

### Point Estimation Algorithms

Statisticss is the scientific discipline of roll uping, analysing and showing informations.There are many statistical techniques that are used to execute informations mining.These techniques besides involves different things like point appraisal, interval appraisal, arrested development and batch more. The population whose distribution is known but depends on one or more unknown considerations, point appraisal predicts the value of the unknown considerations.point appraisal techniques and algorithms are shown below. The classical methods and algorithms are besides explained.many other utile methods such as robust sentiment methods and re-sampling methods have been obtained and on-going research goes on to the progress appraisal techniques.

### Motivation

Point appraisal is known and computationally functional tool to larn the demands of the information excavation theoretical account. This can be used in any informations excavation applications such as summarisation and clip anticipation.summarization is the procedure used for pull outing or deducing required information of the informations.In time-series anticipation, point appraisal is used to gauge one or more values looking after in a sequence by ciphering considerations for a sample.

We discuss the methods of point appraisal, with the method of minutes, maximal likeliness appraisal, and the EM algorithm. Standard to mensurate the public presentation of appraisal method, with prejudice, means squared mistake, standard mistakes, efficiency and consistence. Methods of point appraisal

There are different methods for obtaining point appraisal this includes least squares, the method of minutes, maximal likeliness appraisal, Bayes calculators and robust appraisal. This EM algorithm for happening maximal likeliness estimations.

The point Estimation in done two ways one is prognostic and descriptive informations excavation.

Three classical point appraisal methods

1. The method of minutes

2. Maximal likeliness appraisal

3. Expectation maximization algorithm

The Expectation Maximization Algorithm

The outlook maximization algorithm is most normally used for statistical appraisal issue work outing which is partial informations or issues related to mixture appraisal. The outlook maximization algorithm is used in different countries like gesture appraisal frameworks [ 7, 8,11,12 ] .

The outlook maximization algorithm is a capable method for working out the maximal likeliness ( ML ) . Maximal Likelihood is used for gauging losing or hidden informations. Expectation Maximization Algorithm includes two stairss for each and every loop.

They are

1. The E-step ( Expectation )

2. The M-step ( Maximization )

The E-step:

In this method the lost information is estimated by the utilizing experimental informations and bing estimation of the theoretical account parametric quantities. This can be obtained by doing usage of the restrictive outlook. The appraisal of the lost information is used in stead of the original lost informations [ 11 ] .

EM-E-step

The M-step: In this measure the likeliness map is maximized in such a manner that the false information is the lost information is identified [ 11 ] .

EM-M-step

HIDDING AN IMAGE IN ANOTHER IMAGE FLOW CHART

HIDDING TEXT IN AN Image

FLOW CHART

RESULTS AND DISCUSSION

RESULTS FOR HIDING IMAGE IN AN Image

S.NO

Master

Image

Size

Input signal

Image

Size

EMBEDDED

Image

OUTPUT IMAGE

1

1600?1200

437?437

2

2048?1536

350?350

3

640?480

150?150

4

2592?1944

250?250

5

640?480

250?250

6

150?150

300?300

7

1600?1200

300?300

8

1600?1200

650?650

RESULTS FOR HIDING TEXT IN AN Image

S.NO

ORIGINAL IMAGE

Size

INPUT TEXT

RESIZED Image

Size

OUTPUT IMAGE

OUTPUT TEXT

1

640?480

ABCDE

252?246

RBSDU

2

1600?1200

University

252?246

University

3

2592?1944

UNITED

KINGDOM

252?246

UNITED

KINGDOM

4

2592?1944

University OF

CENTRAL LANCASHIRE

252?246

5

3264?2448

University OF

CENTRAL LANCASHIRE

252?246

6

3264?2448

University OF

CENTRAL LANCASHIRE

252?246

University OF

CENTRAL LANCASHIRE

### APPENDIX-I

### 1. ADVANCED ENCRYPTION STANDARD

### Introduction

Advanced Encryption Standard ( AES ) is implemented utilizing MATLAB. Data Encryption Standard ( DES ) is used to implement cryptanalytic techniques since long clip, AES is the advanced engineering development for DES which is based on block cypher. MATLAB is a Matrix-oriented scheduling linguistic communication, which can be perfectly used for the matrix-based Data- Structure of AES. This can be used for encoding of indiscriminately chosen field text to chipper text and the other manner can besides be done [ 3 ] . If-then else conditions can be reduced by utilizing 128 spots of cardinal length is being introduced [ 3 ] .

### DEMO FOR ADVANCED ENCRYPTION STANDARD

As illustrated in the before subdivision the “main” plan aes_demo [ 3 ] is written to name the low-level formatting map. The low-level formatting maps provide the original encoding and decoding maps. The surrogate s box tabular arraies and reverse s box multinomial matrices and reverse multinomial matrices are called at the beginning to bring forth consecutive encoding or decipher call. Plain text encoding is defined in the following stairss which are of 16 bytes [ 3 ] . The 16 byte cypher text is obtained from the cypher encoding function.To explain the decoding process the cypher text is so passed to the decipher map which gives a plaintext block and that is compared to the chief field text [ 3 ] .

### Algorithm:

1. [ s_box, inv_s_box, tungsten, poly_mat, inv_poly_mat ] = aes_init

2. Plaintext_hex= { ‘00 ‘ ‘11 ‘ ‘22 ‘ ‘33 ‘ ‘44 ‘ ‘55 ‘ ‘66 ‘ ‘77 ‘ ‘88 ‘ ‘99 ‘ ‘aa ‘ ‘bb ‘ ‘cc ‘ ‘dd ‘ ‘ee ‘ ‘ff ‘ } ;

3. Plaintext=hex2dec ( plaintext_hex ) ;

4. Ciphertext=cipher ( plaintext, tungsten, s_box, poly_mat ) ;

5. Re_plaintext=inv_cipher ( cypher, tungsten, inv_s_box, inv_poly_mat ) ;

### AES INITIALIZATION

Before executing any original encoding or decoding the low-level formatting map is called foremost.

The AES low-level formatting [ 3 ] , produces the permutation tabular arraies and generates the s- box which provides information of the changeless vector and model key. In this measure two multinomial matrices are produced [ 3 ] .

### Algorithm

1. Function [ s_box, inv_s_box, tungsten, poly_mat, inv_poly_mat ] = aes_init

2. [ s_box, inv_s_box ] = s_box_gen ;

3. rcon = rcon_gen ;

4. key_hex = { ‘00 ‘ ‘01 ‘ ‘02 ‘ ‘03 ‘ ‘04 ‘ ‘05 ‘ ‘06 ‘ ‘07 ‘ ‘08 ‘ ‘09 ‘ ‘0a ‘ ‘0b ‘ ‘0c ‘ ‘0e ‘ ‘0f ‘ } ;

5. cardinal = hex2dec ( key_hex ) ; w= key_expansion ( cardinal, s_box, rcon ) ;

6. [ poly_mat, inv_poly_mat ] = poly_mat_gen ;

### S BOX GENERATION

The permutation of bytes one by one of the similar finite field is done by the permutation tabular arraies. They use the expanded key maps for the encoding and decoding for the replacing [ 3 ] .

To obtain high velocity in the scheduling it is bit hard to achieve the velocity by the permutation tabular arraies. The thought of implementing the s box is for comparing. S box coevals map generates the s boxes by utilizing affine transmutation for happening opposite where applicable [ 3 ] .

### Algorithm

1. map [ s_box, inv_s_box ] =s_box_gen

2. mod_pol=bin2dec ( ‘100011011 ‘ ) ;

3. opposite ( 1 ) =0 ;

4. for I = 1:255

5. opposite ( i+1 ) =find_inverse ( I, mod_pol ) ;

6. terminal

7. for i=1:256

8. s_box ( I ) =aff_trans ( opposite ( I ) ) ;

9. terminal

10. inv_s_box=s_box_inversion ( s_box ) ;

### FIND INVERSE

In the s box coevals process the initial measure is to happen the reverse generation of all the finite field elements [ 3 ] .The declaration of happen the opposite is given in the initial measure, which is the input parametric quantity. The input byte and the modulo multinomial are to be inverted harmonizing to stairss of the inversion process in the listing [ 3 ] .

### Algorithm

1. map v_inv = find_inverse ( b_in, mod_pol )

2. for 1 = 1:255

3. prod= poly_mult ( b_in, I, mod_pol ) ;

4. if prod = =1

5. b_inv=I ;

6. interruption

7. terminal

8. terminal

### 2 MATHEMATICAL REPRESENTATION OF DATA

### ARTHMETICS OF FINITE FIELD

A byte can be represented in different signifiers. A finite field has basic arithmetic ‘s which are illustrated below. A finite field can besides be named as Galois Field, which consist finitely many elements [ 9 ] .The finite field GF ( 24 ) consists of the 24=16 with different Numberss runing from ( 0…16 ) which represents 4 spots. Particular XOR and modulo-operations are used to look into whether the amount and merchandise of two finite field elements are in the similar scope of the same finite field [ 3 ] .

### DIFFERENT FORMS OF BYTE REPRESENTATION

There are four ways of representation of a finite field component. They are binary representation, denary representation, hexa-decimal representation, multinomial representation.

### BINARY REPRESENTATION

A byte is made up of 8 spots, the below illustration is in the binary representation which has an index B.

10010100b

### DECIMAL REPRESENTATION

The denary signifier representation has an index d. In the below illustration the transition from binary to decimal is performed.

1.27+0.26+0.25+1.24+0.23+1.22+0.21+0.20 =128+16+4=148d

bin2dec is a constitutional map used in MATLAB [ 3, 10 ] to execute the transition of binary figure to denary figure. In order to stand for the figure as threading individual citation Markss are used [ 3 ] .

& gt ; & gt ; bin2dec ( ‘10010100 ‘ )

Ans= 148

### HEXADECIMAL REPRESENTATION

A nibble consists of 4 spots and the scope of Numberss is 0….15. The representation of the figure are divided in two types. First set is from 0…9 which is represented usually with index h.The 2nd set of Numberss are from 10….15, whose representation is alphabets from A…F with index h. Conversion from binary to hexadecimal is done in the undermentioned method. A pass is separated into two nybbles and each nibble is denoted by its matching hexadecimal value [ 3 ] :

Measure 1: 11101001b= 1110 1001 = E9h

Measure 2: E9h = E.24+9.20 = 14.16+9.1 = 224+9 = 233d

hex2dec is a buit in map used in Matlab [ 3, 10 ] .

### POLYNOMIAL REPRESENTATION

A byte can be represented as a multinomial it is same as change overing from binary to decimal. Every figure is multiplied by ten to the corresponding coefficients [ 3 ] .

1.x7+0.x6+0.x5+1.x4+0.x3+1.x2+0.x1+0.x0= x7+x4

It is of import to observe that the multinomials can merely be 1 ‘s or 0 ‘s consequently.

### POLYNOMIAL ADDITION

The two multinomials are summed by summing the coefficients of same powers of ten.

( X9+ x4 + x3 +x + 1 ) + ( x8+x7+x ) = x9+x8+x7+x4+x3+2x+1

From the above illustration the coefficient of the ensuing multinomial are other than 0 or 1 ( 2x ) , this representation is a “classical” amount which does non denote a pass and is non the component belonging to the same finite field [ 3 ] .The XOR operation is used for the summing up, because the consequence of XOR amount of two 1 ‘s is 0 and non 2. The below is the description of the Bit-wise XOR operation of two bytes ( finite field elements ) , which leads to another byte ( component of the same finite field ) [ 3 ] .

### Bit-wise XOR operation:

1

1

0

1

0

0

1

+

0

1

0

1

1

0

0

1

Bit-XOR

0

0

1

1

0

0

0

0

The end product byte is

48d = 30h = 00110000b = x5+x4

### POLYNOMIAL MULTIPLICATION

Generation of multinomials is performed by multiplying each summand of the first multinomial by each and every summand of the 2nd multinomial and the coefficients of the similar powers are summed [ 3 ] .

( X7 + X3 + X + 1 ) . ( X5 + X4 + X2 + 1 )

1.X5 +1. X4 +1. X2 + 1 = X5+X4+X2+1

+

X.X5 +X. X4 +X. X2 +X. 1= X6+X5+X3+X

+

X3.X5 + X3. X4 + X3. X2+ X3.1= X8+X7+X5+X3

+

X7.X5 + X7. X4 + X7. X2 + X7. 1= X12+X11+X9+X7

X12+X11+X9+X8+2X7+X6+3X5+X4+2X3+X2+X+1

Classical multinomial generation

From the above equation the coefficients for some multinomials are different than 1 ‘s and 0 ‘s. To obtain the right consequences XOR operation should be performed and the consequence obtained will be

X12 + X11 + X9+ X8 + X6 + X4+ X2+ X + 1.

### Algorithm:

1. Function ab=poly_mult ( a, B, mod_pol )

2. ab=0 ;

3. for i_bit=1:8

4. if bitget ( a, i_bit )

5. b_shift=bitshift ( B, i_bit-1 ) ;

6. ab=bitxor ( ab, b_shift ) ;

7. terminal

8. terminal

9. for i_bit= 16: – 1: 9

10. if bitget ( ab, i_bit )

11. mod_pol_shift=bitshift ( mod_pol, i_bit-9 ) ;

12. ab=bitxor ( ab, mod_pol_shift ) ;

13. terminal

14. terminal

If the obtained multinomial will hold a grade higher than seven, it can non be represented in one byte so that can be rectified by executing multinomial division [ 3 ] .

### POLYNOMIAL DIVISION

Polynomial division is done in the undermentioned manner ; the highest power of the numerator is divided by the highest power of the denominator and the obtained term is multiplied by the full denominator and that is subtracted from the numerator which gives another numerator [ 3 ] . This procedure is done once more and once more until the highest power of the other numerator has become low than the highest power of the denominator. The ultimate numerator is the balance of the modulo operation [ 3 ] . By executing the “generalized XOR-rules” to the balance it gives the needed byte-confirm multinomial [ 3 ] . By executing the spot flat procedures the consequences obtained will be similar to the bit-wise displacement operations [ 3 ] .

### POLY_MULTI IMPLEMENTATION

The AES-function poly-mult [ 3 ] is performed to achieve a modular decrease by making generation of two multinomials by taking another multinomial. Low-level formatting is performed in the beginning and after that multinomial generation attains the generation and so there are two cringles of same sort for the modular reduction.The process for generation is done in the undermentioned mode every spot of the initial factor is checked and it that is at that place the following factor is moved to the left the consequence is stored by utilizing XOR [ 3 ] .

The consequence obtained by the modular decrease is treated as the numerator and the cringle begins with the most important spot and moves the identified modulo multinomial as the numerator for executing the right XOR process [ 3 ] .

### ALGORITHM FOR ESTIMATION MAZIMIZATION

### Input signal:

1. Data set distribution with unknown parametric quantities

? = { x1, …….. , x2 } ;

2. Incomplete information set X ;

3. Convergence threshold ? ;

End product:

Parameter estimations ?’= { x1 ‘ , …….. , x2 ‘ } ;

Initialize ?’0

P 0 ;

Repeat

P p+1 ;

Q ( ? , ?’n-1 ) E ( logp ( X, Y ; ? | X, ?’n-1 ) ;

Calculating the maximal likelihood estimation of ? to maximise Q ( ? , ?’n-1 ) ;

?’n arg X ? Q ( ? , ?’n-1 ) ;

until | ?’n – ?’n-1| & lt ; ?

? ‘ ?’n ;

### APPENDIX-II

### CODE FOR TEXT STEGANOGRAPHY

clc ; clear all ; shut all ;

warning off ;

% % Password for Segmentation

pw_input=input ( ‘Enter Password for Segmentation… .. : ‘ , ‘s ‘ ) ;

L=length ( pw_input ) ;

for i=1: Liter

Aw ( I ) =str2num ( pw_input ( I ) ) ;

terminal

N= ( unit of ammunition ( L/2 ) ) ;

Sv=0 ;

for i=1: Nitrogen

Sv=i^2*Aw ( I ) +Sv ;

terminal

Sh=0 ;

for i= ( N+1 ) : Liter

Sh=i*Aw ( I ) +Sh ;

terminal

% % Original image

I=imread ( ‘4.bmp ‘ ) ;

figure, imshow ( I ) , rubric ( ‘original image ‘ ) ;

I=imresize ( I, [ 256 256 ] ) ;

% figure, imshow ( I )

k1= ( unit of ammunition ( size ( I,1 ) /Sv ) ) ;

k2= ( unit of ammunition ( size ( I,2 ) /Sh ) ) ;

Row=k1*Sv ; Col=k2*Sh ;

I1=imresize ( I, [ Row Col ] ) ;

figure, imshow ( I1 ) , rubric ( ‘resized image ‘ ) ;

% Colour combination

R=I1 ( : , : ,1 ) ; G=I1 ( : , : ,2 ) ; B=I1 ( : , : ,3 ) ;

% % % Embedding method % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %

% Main Case

for i=1: Row

for j=1: Col

MCr ( I, J ) =round ( R ( I, J ) /16 ) +1 ;

MCg ( I, J ) =round ( G ( I, J ) /16 ) +1 ;

MCb ( I, J ) =round ( B ( I, J ) /16 ) +1 ;

terminal

terminal

% Sub-Case

SCr = c_case ( R ) ;

SCg = c_case ( G ) ;

SCb = c_case ( B ) ;

[ r1 ] =blocking ( R, Sv, Sh ) ;

[ g1 ] =blocking ( G, Sv, Sh ) ;

[ b1 ] =blocking ( B, Sv, Sh ) ;

% % watchword for encoding

pw1=input ( ‘Enter Password for Encryption… .. : ‘ , ‘s ‘ ) ;

% % cardinal coevals for encoding

[ s_box, inv_s_box, tungsten, poly_mat, inv_poly_mat ] = aes_init ;

% Plaintext to be embedded

f1=fopen ( ‘Document.txt ‘ , ‘r ‘ ) ;

P=fread ( f1 ) ‘ ;

ln1=length ( P ) ;

n=16 ;

sp1=P ( 1: N ) ;

sp2=P ( n+1: ln1 ) ;

% % cardinal encription

ciphertext = cypher ( sp1, tungsten, s_box, poly_mat,1 ) ;

P = [ ciphertext sp2 -1 ] ;

u=1 ; v=1 ; id1=1 ; rot1=1 ;

for i=1: ( Row/Sv )

for j=1: ( Col/Sh )

mr=MCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mg=MCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mb=MCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sr=SCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sg=SCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sb=SCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

r=r1 ( 1, U ) .data ; g=g1 ( 1, U ) .data ; b=b1 ( 1, U ) .data ; % take each barricading portion for cleavage

[ rhenium, Ge, be, P1, id1, rot1, Bk ] =embedding ( R, g, B, Mister, milligram, megabit, strontium, seaborgium, antimony, P, id1, rot1 ) ; % Embedding procedure through cleavage

r1 ( 1, U ) .data=re ; g1 ( 1, U ) .data=ge ; b1 ( 1, U ) .data=be ; % Embedded spots replace the Original spots

P=P1 ; u=u+1 ;

if ( Bk==1 )

interruption

terminal

terminal

if ( Bk==1 )

interruption

terminal

terminal

% % replace the original image

k=0 ;

for i=1: ( Row/Sv )

for j=1: ( Col/Sh )

k=k+1 ;

R ( ( ( i-1 ) *Sv+1 ) : i*Sv, ( ( j-1 ) *Sh+1 ) : j*Sh ) =r1 ( K ) .data ;

G ( ( ( i-1 ) *Sv+1 ) : i*Sv, ( ( j-1 ) *Sh+1 ) : j*Sh ) =g1 ( K ) .data ;

B ( ( ( i-1 ) *Sv+1 ) : i*Sv, ( ( j-1 ) *Sh+1 ) : j*Sh ) =b1 ( K ) .data ;

terminal

terminal

I1 ( : , : ,1 ) =R ; I1 ( : , : ,2 ) =G ; I1 ( : , : ,3 ) =B ;

figure, imshow ( I1, [ ] ) ; , rubric ( ‘Embedded image ‘ ) % % Embedded image

% % % Extracting method % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %

% % watchword for extraction

pw_input1=input ( ‘Enter Password for Extracting… .. : ‘ , ‘s ‘ ) ;

L1=length ( pw_input ) ;

for i=1: L1

Aw ( I ) =str2num ( pw_input ( I ) ) ;

terminal

N= ( unit of ammunition ( L/2 ) ) ;

Sv=0 ;

for i=1: Nitrogen

Sv=i^2*Aw ( I ) +Sv ;

terminal

Sh=0 ;

for i= ( N+1 ) : L1

Sh=i*Aw ( I ) +Sh ;

terminal

k1= ( unit of ammunition ( size ( I1,1 ) /Sv ) ) ;

k2= ( unit of ammunition ( size ( I1,2 ) /Sh ) ) ;

Row=k1*Sv ; Col=k2*Sh ;

if ( pw_input==pw_input1 )

I2=imresize ( I1, [ Row Col ] ) ;

else

I2=imresize ( I, [ Row Col ] ) ;

terminal

R=I2 ( : , : ,1 ) ; G=I2 ( : , : ,2 ) ; B=I2 ( : , : ,3 ) ;

% Main Case

for i=1: Row

for j=1: Col

MCr ( I, J ) =round ( R ( I, J ) /16 ) +1 ;

MCg ( I, J ) =round ( G ( I, J ) /16 ) +1 ;

MCb ( I, J ) =round ( B ( I, J ) /16 ) +1 ;

terminal

terminal

% Sub-Case

SCr = c_case ( R ) ;

SCg = c_case ( G ) ;

SCb = c_case ( B ) ;

[ r1 ] =blocking ( R, Sv, Sh ) ;

[ g1 ] =blocking ( G, Sv, Sh ) ;

[ b1 ] =blocking ( B, Sv, Sh ) ;

lnn=length ( P ) ;

P2=zeros ( 1, lnn ) ;

P2 ( lnn ) =-1 ;

u=1 ; v=1 ; id2=1 ; rot2=1 ;

for i=1: ( Row/Sv )

for j=1: ( Col/Sh )

mr=MCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mg=MCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mb=MCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sr=SCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sg=SCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sb=SCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

r=r1 ( 1, U ) .data ; g=g1 ( 1, U ) .data ; b=b1 ( 1, U ) .data ; % take each barricading portion for cleavage

[ P1, id2, rot2, Bk ] =extracting ( R, g, B, Mister, milligram, megabit, strontium, seaborgium, antimony, P2, id2, rot2 ) ; % Extracting procedure through cleavage

r1 ( 1, U ) .data=re ; g1 ( 1, U ) .data=ge ; b1 ( 1, U ) .data=be ; % Embedded spots replace the Original spots

P=P1 ; u=u+1 ;

if ( Bk==1 )

interruption

terminal

terminal

if ( Bk==1 )

interruption

terminal

terminal

n=16 ;

sr1=P ( 1: N ) ;

sr2=P ( n+1: length ( P2 ) ) ;

% % watchword for decoding

pw2_input=input ( ‘Enter the watchword for Decryption… .. : ‘ , ‘s ‘ ) ;

re_plaintext = inv_cipher ( sr1, tungsten, inv_s_box, inv_poly_mat,1 ) ;

Text=char ( re_plaintext )

char ( sr2 )

APPENDIX- III

CODE FOR TEXT STEGANOGRAPHY

clc ; clear all ; shut all ;

warning off ;

% % Password for Segmentation

pw_input=input ( ‘Enter Password for Segmentation… .. : ‘ , ‘s ‘ ) ;

L=length ( pw_input ) ;

for i=1: Liter

Aw ( I ) =str2num ( pw_input ( I ) ) ;

terminal

N= ( unit of ammunition ( L/2 ) ) ;

Sv=0 ;

for i=1: Nitrogen

Sv=i^2*Aw ( I ) +Sv ;

terminal

Sh=0 ;

for i= ( N+1 ) : Liter

Sh=i*Aw ( I ) +Sh ;

terminal

% % Original image

I=imread ( ‘7.bmp ‘ ) ;

% I=imresize ( I, [ 256 256 ] ) ;

figure, imshow ( I ) , rubric ( ‘original image ‘ )

k1= ( unit of ammunition ( size ( I,1 ) /Sv ) ) ;

k2= ( unit of ammunition ( size ( I,2 ) /Sh ) ) ;

Row=k1*Sv ; Col=k2*Sh ;

I1=imresize ( I, [ Row Col ] ) ;

figure, imshow ( I1 ) , rubric ( ‘cover image ‘ )

% Colour combination

R=I1 ( : , : ,1 ) ; G=I1 ( : , : ,2 ) ; B=I1 ( : , : ,3 ) ;

% % % Embedding method % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %

% Main Case

for i=1: Row

for j=1: Col

MCr ( I, J ) =round ( R ( I, J ) /16 ) +1 ;

MCg ( I, J ) =round ( G ( I, J ) /16 ) +1 ;

MCb ( I, J ) =round ( B ( I, J ) /16 ) +1 ;

terminal

terminal

% Sub-Case

SCr = sub_case ( MCr, MCg, MCb ) ;

SCg = sub_case ( MCg, MCb, MCr ) ;

SCb = sub_case ( MCb, MCr, MCg ) ;

[ r1 ] =blocking ( R, Sv, Sh ) ;

[ g1 ] =blocking ( G, Sv, Sh ) ;

[ b1 ] =blocking ( B, Sv, Sh ) ;

% % message image

img=double ( imread ( ’29.bmp ‘ ) ) ;

% % for color image

rd1=img ( : , : ,1 ) ; gd1=img ( : , : ,2 ) ; bd1=img ( : , : ,3 ) ;

img= [ rd1 ( : ) ‘ gd1 ( : ) ‘ bd1 ( : ) ‘ ] ;

P= [ img ( : ) ‘ -1 ] ;

ln1=length ( P ) ;

n=16 ;

sp1=P ( 1: N ) ;

sp2=P ( n+1: ln1 ) ;

% % watchword for encoding

pw1=input ( ‘Enter Password for Encryption… .. : ‘ , ‘s ‘ ) ;

% % cardinal coevals for encoding

[ s_box, inv_s_box, tungsten, poly_mat, inv_poly_mat ] = aes_init ;

% % cardinal encription

ciphertext = cypher ( sp1, tungsten, s_box, poly_mat,1 ) ;

P= [ ciphertext sp2 ] ;

u=1 ; v=1 ; id1=1 ; rot1=1 ;

for i=1: ( Row/Sv )

for j=1: ( Col/Sh )

mr=MCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mg=MCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mb=MCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sr=SCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sg=SCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sb=SCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

r=r1 ( 1, U ) .data ; g=g1 ( 1, U ) .data ; b=b1 ( 1, U ) .data ; % take each barricading portion for cleavage

[ rhenium, Ge, be, P1, id1, rot1, Bk ] =embedding ( R, g, B, Mister, milligram, megabit, strontium, seaborgium, antimony, P, id1, rot1 ) ; % Embedding procedure through cleavage

r1 ( 1, U ) .data=re ; g1 ( 1, U ) .data=ge ; b1 ( 1, U ) .data=be ; % Embedded spots replace the Original spots

P=P1 ; u=u+1 ;

if ( Bk==1 )

interruption

terminal

terminal

if ( Bk==1 )

interruption

terminal

terminal

% % replace the original image

k=0 ;

for i=1: ( Row/Sv )

for j=1: ( Col/Sh )

k=k+1 ;

R ( ( ( i-1 ) *Sv+1 ) : i*Sv, ( ( j-1 ) *Sh+1 ) : j*Sh ) =r1 ( K ) .data ;

G ( ( ( i-1 ) *Sv+1 ) : i*Sv, ( ( j-1 ) *Sh+1 ) : j*Sh ) =g1 ( K ) .data ;

B ( ( ( i-1 ) *Sv+1 ) : i*Sv, ( ( j-1 ) *Sh+1 ) : j*Sh ) =b1 ( K ) .data ;

terminal

terminal

I1 ( : , : ,1 ) =R ; I1 ( : , : ,2 ) =G ; I1 ( : , : ,3 ) =B ;

figure, imshow ( I1, [ ] ) , rubric ( ‘stegno image ‘ ) ; % % Embedded image

% % % Extracting method % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %

% % watchword for extraction

pw_input1=input ( ‘Enter Password for Extracting… .. : ‘ , ‘s ‘ ) ;

L1=length ( pw_input ) ;

for i=1: L1

Aw ( I ) =str2num ( pw_input ( I ) ) ;

terminal

N= ( unit of ammunition ( L/2 ) ) ;

Sv=0 ;

for i=1: Nitrogen

Sv=i^2*Aw ( I ) +Sv ;

terminal

Sh=0 ;

for i= ( N+1 ) : L1

Sh=i*Aw ( I ) +Sh ;

terminal

k1= ( unit of ammunition ( size ( I1,1 ) /Sv ) ) ;

k2= ( unit of ammunition ( size ( I1,2 ) /Sh ) ) ;

Row=k1*Sv ; Col=k2*Sh ;

if ( pw_input==pw_input1 )

I2=imresize ( I1, [ Row Col ] ) ;

else

I2=imresize ( I, [ Row Col ] ) ;

terminal

R=I2 ( : , : ,1 ) ; G=I2 ( : , : ,2 ) ; B=I2 ( : , : ,3 ) ;

% Main Case

for i=1: Row

for j=1: Col

MCr ( I, J ) =round ( R ( I, J ) /16 ) +1 ;

MCg ( I, J ) =round ( G ( I, J ) /16 ) +1 ;

MCb ( I, J ) =round ( B ( I, J ) /16 ) +1 ;

terminal

terminal

% Sub-Case

SCr = sub_case ( MCr, MCg, MCb ) ;

SCg = sub_case ( MCg, MCb, MCr ) ;

SCb = sub_case ( MCb, MCr, MCg ) ;

[ r1 ] =blocking ( R, Sv, Sh ) ;

[ g1 ] =blocking ( G, Sv, Sh ) ;

[ b1 ] =blocking ( B, Sv, Sh ) ;

ln2=length ( P ) ;

P2=zeros ( 1, ln2 ) ;

P2 ( ln2 ) =-1 ;

u=1 ; v=1 ; id2=1 ; rot2=1 ;

for i=1: ( Row/Sv )

for j=1: ( Col/Sh )

mr=MCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mg=MCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

mb=MCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sr=SCr ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sg=SCg ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

sb=SCb ( ( i-1 ) *Sv+1: i*Sv, ( j-1 ) *Sh+1: j*Sh ) ;

[ P1, id2, rot2, Bk ] =extracting ( R, g, B, Mister, milligram, megabit, strontium, seaborgium, antimony, P2, id2, rot2 ) ; % Extracting procedure through cleavage

% r1 ( 1, U ) .data=re ; g1 ( 1, U ) .data=ge ; b1 ( 1, U ) .data=be ; % Embedded spots replace the Original spots

P2=P1 ; u=u+1 ;

if ( Bk==1 )

interruption

terminal

terminal

if ( Bk==1 )

interruption

terminal

terminal

sr1=P2 ( 1: N ) ;

sr2=P2 ( n+1: length ( P2 ) ) ;

% % watchword for decoding

pw2_input=input ( ‘Enter the watchword for Decryption… .. : ‘ , ‘s ‘ ) ;

if ( length ( pw2_input ) ==length ( pw1 ) )

if ( pw2_input~=pw1 )

w=zeros ( 44,4 ) ;

sr2=round ( 256*rand ( size ( sr2 ) ) ) ;

terminal

else

w=zeros ( 44,4 ) ;

sr2=round ( 256*rand ( size ( sr2 ) ) ) ;

terminal

% % Decoding

re_plaintext = inv_cipher ( sr1, tungsten, inv_s_box, inv_poly_mat,1 ) ;

Sr= [ re_plaintext sr2 ] ;

ln2=length ( Sr ) -1 ;

% % % % % for grey scale image

% sz=sqrt ( ln2 ) ;

% Sr=reshape ( Sr ( 1: ln2 ) , [ sz sz ] ) ;

% figure, imshow ( uint8 ( Sr ) ) ;

% % % % % for colour image

sz=ln2/3 ;

S_Rd=Sr ( 1: sz ) ;

S_Gn=Sr ( sz+1:2*sz ) ;

S_Bl=Sr ( 2*sz+1: ln2 ) ;

sz1=sqrt ( sz ) ;

Sr1 ( : , : ,1 ) =reshape ( S_Rd, [ sz1 sz1 ] ) ;

Sr1 ( : , : ,2 ) =reshape ( S_Gn, [ sz1 sz1 ] ) ;

Sr1 ( : , : ,3 ) =reshape ( S_Bl, [ sz1 sz1 ] ) ;

figure, imshow ( uint8 ( Sr1 ) ) , rubric ( ‘decrypted message image ‘ ) ;

add_round_key

map state_out = add_round_key ( state_in, round_key )

% ADD_ROUND_KEY Add ( XOR ) the unit of ammunition key to the province.

%

% STATE_OUT = ADD_ROUND_KEY ( STATE_IN, ROUND_KEY )

% adds the current unit of ammunition key matrix ROUND_KEY

% to the current province matrix STATE_IN.

% Adding in GF ( 2^8 ) is performed via bitwise XOR.

% Add province ( matrix ) and round key ( matrix ) via bitwise XOR

state_out = bitxor ( state_in, round_key ) ;

AES_INIT

map [ s_box, inv_s_box, tungsten, poly_mat, inv_poly_mat ] = aes_init

clc

% Create the S-box and the reverse S-box

[ s_box, inv_s_box ] = s_box_gen ( 1 ) ;

% Create the unit of ammunition changeless array

rcon = rcon_gen ( 1 ) ;

% Define an arbitrary 16-byte cypher key in hexadecimal ( threading ) representation

% The undermentioned two specific keys are used as illustrations

% in the AES-Specification ( bill of exchange )

key_hex = { ’00 ‘ ’01 ‘ ’02 ‘ ’03 ‘ ’04 ‘ ’05 ‘ ’06 ‘ ’07 ‘ …

’08 ‘ ’09 ‘ ‘0a ‘ ‘0b ‘ ‘0c ‘ ‘0d ‘ ‘0e ‘ ‘0f ‘ } ;

% key_hex = { ‘2b ‘ ‘7e ‘ ’15 ‘ ’16 ‘ ’28 ‘ ‘ae ‘ ‘d2 ‘ ‘a6 ‘ …

% ‘ab ‘ ‘f7 ‘ ’15 ‘ ’88 ‘ ’09 ‘ ‘cf ‘ ‘4f ‘ ‘3c ‘ } ;

% Convert the cypher key from hexadecimal ( threading ) to decimal representation

cardinal = hex2dec ( key_hex ) ;

% Create the expanded key ( agenda )

tungsten = key_expansion ( cardinal, s_box, rcon, 1 ) ;

% Create the multinomial transmutation matrix and the reverse multinomial matrix

% to be used in MIX_COLUMNS

[ poly_mat, inv_poly_mat ] = poly_mat_gen ( 1 ) ;

Aff_trans

map b_out = aff_trans ( b_in )

% AFF_TRANS Apply an affine transmutation over GF ( 2^8 ) .

%

% B_OUT = AFF_TRANS ( B_IN )

% applies an affine transmutation to the input byte B_IN.

%

% The transmutation consists of

% 1. a multinomial modulo generation

% by a predefined generation multinomial

% utilizing a predefined modulo multinomial over GF ( 2^8 ) and

% 2. the add-on ( XOR ) of a predefined add-on multinomial

% Define the modulo multinomial

% to be used in the modulo operation in poly_mult

mod_pol = bin2dec ( ‘100000001 ‘ ) ;

% Define the generation multinomial

% In the Rijndael AES Proposal

% they say they use the multinomial ‘11110001 ‘ ,

% which is incorrect.

mult_pol = bin2dec ( ‘00011111 ‘ ) ;

% Define the add-on multinomial

add_pol = bin2dec ( ‘01100011 ‘ ) ;

% Modular multinomial generation

% of the input byte and the fixed generation multinomial

temp = poly_mult ( b_in, mult_pol, mod_pol ) ;

% Add ( XOR ) the changeless ( add-on multinomial )

b_out = bitxor ( temp, add_pol ) ;

Blocking

map [ x R ] =blocking ( y, xw, yw )

% Design Of Butterworth Low Pass Filter

xmax=size ( y,1 ) ;

ymax=size ( y,2 ) ;

imax=xmax/xw ;

jmax=ymax/yw ;

k=0 ;

r=zeros ( xmax, ymax ) ;

for i=1: imax

for j=1: jmax

k=k+1 ;

ten ( K ) .data=y ( ( ( i-1 ) *xw+1 ) : i*xw, ( ( j-1 ) *yw+1 ) : j*yw ) ;

R ( ( ( i-1 ) *xw+1 ) : i*xw, ( ( j-1 ) *yw+1 ) : j*yw ) =x ( K ) .data ;

terminal

terminal

return

c_case

map mask=c_case ( ima )

ima=double ( ima ) ;

copy=ima ; % make a transcript

ima=ima ( : ) ; % vectorize ima

mi=min ( ima ) ; % trade with negative

ima=ima-mi+1 ; % and zero values

m=max ( ima ) ;

s=length ( ima ) ;

% create image histogram

h=histogram ( ima ) ; % Call histogram map

x=find ( H ) ; % Return the places

h=h ( x ) ;

x=x ( : ) ; h=h ( : ) ; % changes 2D vector inti 1D

k=6 ; % % % % No. of Clusters.

% novice parametric quantities

mu= ( 1: K ) *m/ ( k+1 ) ; % Random Mean values

v=ones ( 1, K ) *m ;

p=ones ( 1, K ) *1/k ;

% start procedure

sml = mean ( diff ( x ) ) /1000 ;

while ( 1 )

% Expectation

prb = distribution ( mu, V, P, x ) ; % Calls distribution map

scal = amount ( prb,2 ) +eps ;

loglik=sum ( h.*log ( scal ) ) ;

% Maximizarion

for j=1: K

pp=h. *prb ( : ,j ) ./scal ;

P ( J ) = amount ( pp ) ;

mu ( J ) = amount ( x. *pp ) /p ( J ) ;

vr = ( x-mu ( J ) ) ;

V ( J ) =sum ( vr.*vr. *pp ) /p ( J ) +sml ;

terminal

P = P + 1e-3 ;

P = p/sum ( P ) ;

% Exit status

prb = distribution ( mu, V, P, x ) ; % Calls distribution map

scal = amount ( prb,2 ) +eps ;

nloglik=sum ( h.*log ( scal ) ) ;

if ( ( nloglik-loglik ) & lt ; 0.0001 ) interruption ; terminal ;

% clf

% secret plan ( x, h ) ;

% clasp on

% secret plan ( x, prb, ‘g — ‘ )

% secret plan ( x, sum ( prb,2 ) , ‘r ‘ )

% drawnow

terminal

% calculate mask

mu=mu+mi-1 ; % recover existent scope

s=size ( transcript ) ;

mask=zeros ( s ) ;

for i=1: s ( 1 ) ,

for j=1: s ( 2 ) ,

for n=1: K

degree Celsius ( n ) =distribution ( mu ( n ) , v ( N ) , p ( N ) , transcript ( I, J ) ) ; % Calls distribution merriment.

terminal

a=find ( c==max ( degree Celsius ) ) ;

mask ( one, J ) =a ( 1 ) ;

terminal

terminal

Cipher text

map ciphertext = cypher ( plaintext, w, s_box, poly_mat, vargin )

% CIPHER Convert 16 bytes of plaintext to 16 bytes of ciphertext.

%

% CIPHERTEXT = CIPHER ( PLAINTEXT, W, S_BOX, POLY_MAT )

% converts PLAINTEXT to CIPHERTEXT,

% utilizing the expanded cypher key W,

% the byte permutation tabular array S_BOX, and

% the transmutation matrix POLY_MAT.

%

% CIPHERTEXT = CIPHER ( PLAINTEXT, W, S_BOX, POLY_MAT, 1 )

% switches verbose manner on, which displays intermediate consequences.

%

% PLAINTEXT has to be a vector of 16 bytes ( 0 & lt ; = PLAINTEXT ( I ) & lt ; = 255 ) .

% If there is an optional “ long-winded manner ” statement

if nargin & gt ; 4

% Switch the long-winded manner flag on

verbose_mode = 1 ;

% If there is no optional “ long-winded manner ” statement

else

% Switch the long-winded manner flag off

verbose_mode = 0 ;

terminal

% If the input vector is a cell array or does non hold 16 elements

if iscell ( plaintext ) | goad ( size ( plaintext ) ) ~= 16

% Inform user and abort

mistake ( ‘Plaintext has to be a vector ( non a cell array ) with 16 elements. ‘ )

terminal

% If any component of the input vector can non be represented by 8 spots

if any ( plaintext & lt ; 0 | plaintext & gt ; 255 )

% Inform user and abort

mistake ( ‘Elements of plaintext vector have to be bytes ( 0 & lt ; = plaintext ( I ) & lt ; = 255 ) . ‘ )

terminal

% If the expanded cardinal array is a cell arrray or does non hold the right size

if iscell ( tungsten ) | any ( size ( tungsten ) ~= [ 44, 4 ] )

% Inform user and abort

mistake ( ‘w has to be an array ( non a cell array ) with [ 44 x 4 ] elements. ‘ )

terminal

% If any component of the expanded cardinal array can non be represented by 8 spots

if any ( tungsten & lt ; 0 | tungsten & gt ; 255 )

% Inform user and abort

mistake ( ‘Elements of cardinal array tungstens have to be bytes ( 0 & lt ; = w ( I, J ) & lt ; = 255 ) . ‘ )

terminal

% Display headline if requested

if verbose_mode

disp ( ‘ ‘ )

disp ( ‘******************************************** ‘ )

disp ( ‘* * ‘ )

disp ( ‘* C I P H E R * ‘ )

disp ( ‘* * ‘ )

disp ( ‘******************************************** ‘ )

disp ( ‘ ‘ )

terminal

% Copy the 16 elements of the input vector

% column-wise into the 4 ten 4 province matrix

province = reshape ( plaintext, 4, 4 ) ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘Initial province: ‘ , province )

terminal

% Copy the first 4 rows ( 4 x 4 elements ) of the expanded key

% into the current unit of ammunition key.

% Transpose to do this column-wise

round_key = ( tungsten ( 1:4, : ) ) ‘ ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘Initial unit of ammunition key: ‘ , round_key )

terminal

% Add ( xor ) the current unit of ammunition key ( matrix ) to the province ( matrix )

province = add_round_key ( province, round_key ) ;

% Loop over 9 unit of ammunitions

for i_round = 1: 9

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( [ ‘State at start of unit of ammunition ‘ , num2str ( i_round ) , ‘ : ‘ ] , province )

terminal

% Substitute all 16 elements of the province matrix

% by jostling them through the S-box

province = sub_bytes ( province, s_box ) ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘After sub_bytes: ‘ , province )

terminal

% Cyclically shift the last three rows of the province matrix

province = shift_rows ( province ) ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘After shift_rows: ‘ , province )

terminal

% Transform the columns of the province matrix via a four-term multinomial

province = mix_columns ( province, poly_mat ) ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘After mix_columns: ‘ , province )

terminal

% Extract the current unit of ammunition key ( 4 x 4 matrix ) from the expanded key

round_key = ( tungsten ( ( 1:4 ) + 4*i_round, : ) ) ‘ ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘Round key: ‘ , round_key )

terminal

% Add ( XOR ) the current unit of ammunition key ( matrix ) to the province ( matrix )

province = add_round_key ( province, round_key ) ;

terminal

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘State at start of concluding unit of ammunition: ‘ , province )

terminal

% Substitute all 16 elements of the province matrix

% by jostling them through the S-box

province = sub_bytes ( province, s_box ) ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘After sub_bytes: ‘ , province )

terminal

% Cyclically shift the last three rows of the province matrix

province = shift_rows ( province ) ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘After shift_rows: ‘ , province )

terminal

% Extract the last unit of ammunition key ( 4 x 4 matrix ) from the expanded key

round_key = ( tungsten ( 41:44, : ) ) ‘ ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘Round key: ‘ , round_key )

terminal

% Add ( xor ) the current unit of ammunition key ( matrix ) to the province ( matrix )

province = add_round_key ( province, round_key ) ;

% Display intermediate consequence if requested

if verbose_mode

disp_hex ( ‘Final province: ‘ , province )

terminal

% reshape the 4 ten 4 province matrix into a 16 component row vector

ciphertext = reshape ( province, 1, 16 ) ;

condn1

map [ CP1, NP1, d2 ] = condn1 ( CP, NP, d1 )

DB=dec2bin ( d1,8 ) ;

DB=bitget ( uint8 ( DB ) ,1 ) ;

b1= [ DB ( 5 ) DB ( 6 ) DB ( 7 ) DB ( 8 ) ] ;

CP=bitset ( CP,2, b1 ( 1 ) ) ; CP=bitset ( CP,1, b1 ( 2 ) ) ; % % hide 2 spots in Current Pixel

NP=bitset ( NP,2, b1 ( 3 ) ) ; NP=bitset ( NP,1, b1 ( 4 ) ) ; % % hide 2 spots in Following Pixel

CP1=CP ; NP1=NP ;

RT= [ DB ( 5 ) , DB ( 6 ) , DB ( 7 ) , DB ( 8 ) , DB ( 1 ) , DB ( 2 ) , DB ( 3 ) , DB ( 4 ) ] ;

RT=dec2bin ( RT ) ;

RT=RT ‘ ;

d2=bin2dec ( RT ) ;

rhythm

map matrix_out = rhythm ( matrix_in, way )

% SHIFT_ROWS Cyclically switch the rows of the province matrix.

%

% MATRIX_OUT = CYCLE ( MATRIX_IN, ‘left ‘ )

% cyclically shifts the last three rows of the input matrix to the left.

% The first row is non shifted: [ 1 2 3 4 ]

% The 2nd row is cyclically shifted one time to the left: [ 2 3 4 1 ]

% The 3rd row is cyclically shifted twice to the left: [ 3 4 1 2 ]

% The 4th row is cyclically shifted thrice to the left: [ 4 1 2 3 ]

% MATRIX_OUT = CYCLE ( MATRIX_IN, ‘right ‘ )

% cyclically shifts the last three rows of the input matrix to the right.

% The first row is non shifted: [ 1 2 3 4 ]

% The 2nd row is cyclically shifted one time to the right: [ 4 1 2 3 ]

% The 3rd row is cyclically shifted twice to the right: [ 3 4 1 2 ]

% The 4th row is cyclically shifted thrice to the right: [ 2 3 4 1 ]

% If the matrix has to be shifted to the left,

if strcmp ( way, ‘left ‘ )

% generate the column vector [ 0 5 10 15 ] ‘

gap = ( 0: 5: 15 ) ‘ ;

% If the matrix has to be shifted to the right,

else

% generate the column vector [ 16 13 10 7 ] ‘

gap = ( 16: -3: 7 ) ‘ ;

terminal

% Generate the row vector [ 0 4 8 12 ]

row = 0: 4: 12 ;

% Repeat the column to make the matrix [ 0 0 0 0 ] ( left displacement )

% [ 5 5 5 5 ]

% [ 10 10 10 10 ]

% [ 15 15 15 15 ]

gaps = repmat ( gap, 1, 4 ) ;

% Repeat the row to make the matrix [ 0 4 8 12 ]

% [ 0 4 8 12 ]

% [ 0 4 8 12 ]

% [ 0 4 8 12 ]

rows = repmat ( row, 4, 1 ) ;

% Add both matrices,

% fold back into the 0… 15 sphere,

% and add 1, because Matlab indices do get down with 1

% [ 1 5 9 13 ]

% [ 6 10 14 2 ]

% [ 11 15 3 7 ]

% [ 16 4 8 12 ]

ind_mat = mod ( rows + gaps, 16 ) + 1 ;

% Apply the merely created index matrix to the input matrix.

% Elementss of the index matrix are additive ( column-wise ) indices.

matrix_out = matrix_in ( ind_mat ) ;

Disp_hex

map disp_hex ( twine, hex_array )

% DISP_HEX Display an array in hexadecimal signifier.

%

% DISP_HEX ( STRING, HEX_ARRAY )

% displays the hexadecimal representation

% of the array ( matrix ) HEX_ARRAY

% exposing String as the array name ( rubric ) .

%

% This is merely a quick-and-dirty execution

% without any sophisticated mistake direction.

% Find the Numberss of rows and columns of the array to be displayed

[ n_hex_array, m_hex_array ] = size ( hex_array ) ;

% Find the length of the “ rubric ” twine

n_string = length ( threading ) ;

% Create an empty twine of the length of the rubric twine

empty_string = 1s ( 1, n_string ) * ‘ ‘ ;

% Loop over every row of the array

for one = 1: n_hex_array

% If we are speaking about the first row,

if one == 1

% display the rubric twine ( subsequently )

line = twine ;

% If we are non speaking about the first row,

else

% do non expose the rubric twine ( subsequently )

line = empty_string ;

terminal

% Loop over every column of the array

for J = 1: m_hex_array

% Append the hexadecimal representation of the current array component

line = [ line, lower ( dec2hex ( hex_array ( one, J ) ,2 ) ) , ‘ ‘ ] ;

terminal

% Display the assembled line

disp ( line )

terminal

% Display an empty ( dividing ) line

disp ( ‘ ‘ )

Distribution

map y=distribution ( m, V, g, x ) % Returns the end product Y

x=x ( : ) ;

m=m ( : ) ;

v=v ( : ) ;

g=g ( : ) ;

for i=1: size ( m,1 )

500 = x-m ( one ) ;

A = g ( I ) /sqrt ( 2*pi*v ( I ) ) ;

Y ( : ,i ) = amp*exp ( -0.5 * ( d.*d ) /v ( I ) ) ;

terminal

### EMBEDDING

map [ R, G, B, plaintext, ind, putrefaction, Break ] = embedding ( R, G, B, MCr, MCg, MCb, SCr, SCg, SCb, plaintext, ind, putrefaction )

[ sz1 sz2 ] =size ( R ) ;

em_bits=0 ; Break=0 ;

for i=1: sz1

for j=1: sz2

if ( plaintext ( ind ) ~=-1 )

m= [ MCr ( I, J ) MCg ( I, J ) MCb ( I, J ) ] ;

s= [ SCr ( I, J ) SCg ( I, J ) SCb ( I, J ) ] ;

Cmc=min ( m ) ;

sel=find ( Cmc==m ) ;

sel_color=m ( sel ) ;

SC=s ( sel ) ;

% % First method

if ( Cmc & gt ; =2 & A ; & A ; SC ( 1 ) ==1 )

if ( sel ( 1 ) ==1 ) % represents the colour of ruddy

CP=R ( I, J ) ; NP=R ( I, j+1 ) ; % CP=Current Pixel, NP=