Tribhuvan University Institute of Science and

Technology Himalaya College of Engineering

[Code: CSC-404]

A
Final Year Project Report
On
“Image Steganography”

by
Aakash Shrestha [7212/072]
Sandhya Khadka [7251/072]
Simran Dhoju [7255/072]
Supriya Amatya [7259/072]

Submitted to Department of Computer Science and Information

Technology in partial fulfillment of the requirement for the
Bachelor’s degree in Computer Science and Information Technology

Department of Computer Science and Information Technology

August 5, 2019
 DECLARATION BY CANDIDATES
We, the final year (seventh semester) B.Sc. CSIT students of Himalaya College of
Engineering, hereby declare that the project report entitled “Image Steganography”
is carried by us under the guidance and supervision of Mr. Bhupendra Ram Luhar
(Lecturer, Tribhuwan University)

We assure that this project work embedded the result of the original work and the
contents of the project have not submitted to anybody else for the award of degree.
This project is purely of academic interest.

We have followed the guide lines provided by the university in writing the report.

Signature of the candidates

Mr. Aakash Shrestha

Date: August 5, 2019 Miss Sandhya Khadka

Miss Simran Dhoju

Miss Supriya Amatya

 SUPERVISOR’S CERTIFICATE
This is to certify that the project entitled “Image Steganography” is done by Aakash
Shrestha (7212), Sandhya Khadka (7251), Simran Dhoju (7255) and Supriya Amatya
(7259) of fourth year B.Sc. CSIT under the guidance and supervision of Mr.
Bhupendra Ram Luhar, Lecturer, Himalaya College of Engineering, towards the
partial fulfillment of bachelor’s degree of Computer Science and Information
Technology.

_________________________________

Signature of Supervisor

Mr. Bhupendra Ram Luhar

Lecturer

Himalaya College of Engineering

Chyasal, Lalitpur

Date:

Seal:
 EVALUATION COMMITTEE
This is to certify that the project entitled “Image Steganography” is done by Aakash
Shrestha (7212), Sandhya Khadka (7251), Simran Dhoju (7255) and Supriya Amatya
(7259) of fourth year B.Sc. CSIT under the guidance and supervision of Mr.
Bhupendra Ram Luhar, Lecturer, Himalaya College of Engineering, towards the
partial fulfillment of bachelor’s degree of Computer Science and Information
Technology.

___________________________ ___________________________

Signature of HOD Signature of Supervisor

Er. Himal Chand Thapa Head of Mr. Bhupendra Luhar

Department (CSIT) Himalaya Lecturer

College of Engineering Chyasal, Himalaya College of Engineering

Lalitpur Chyasal, Lalitpur

_______________________

External Examiner
 ACKNOWLEDGEMENT
Firstly, we would like to thank Himalaya College of Engineering for providing us this
platform to gain knowledge about different aspects of Computer Science and
Information Technology.

We would also like to thank our Head of Department Er. Himal Chand Thapa who
played an important role in providing us advice and suggestion towards the selection
of our project topic as well as for providing us with necessary content and classes
regarding the project. We are also grateful for the care and support from our
supervisor Mr. Bhupendra Ram Luhar and for providing us all necessary
suggestions and corrections he provided, for our project.

In addition, we also want to thank our friends and classmates for helping us with the
guidelines we needed during the completion of our project.

V
 ABSTRACT
Steganography is the art of hiding the fact that communication is taking place, by
hiding information in other information. Cryptography is a technique associated with
the process of converting ordinary plain text into unintelligible text and vice-versa. In
contrast to cryptography, steganography is not to keep others from knowing the
hidden information but it is to keep others from thinking that the information even
exists. Together cryptography and steganography can provide a powerful basis for
data security.

The main purpose of this project is to produce a security tool based on steganography
and cryptography techniques for sending and receiving sensitive information over the
internet. The program first encrypts the message data using AES algorithm and then
embeds the result of encrypted data in the provided image file using steganography
technique. The system also provides the feature for extracting the hidden data from
the corresponding image file and decrypting the extracted data for eventually finding
the original message. The embedding process follows image’s LSB replacement
algorithm. To obtain the hidden message, the process is reversed. This project also
provides a client/server module for stego image transmission. The client-server
connection is established using TCP socket programming on a user given port.

Using the system produced, messages were successfully encrypted and hidden inside
an image file. Also, using the generated stego image, the hidden message was
extracted and decrypted successfully.

Keyword: Steganography, Cryptography, LSB, AES.

VI
 Table of Contents
ACKNOWLEDGEMENT ............................................................................................ V

ABSTRACT ................................................................................................................. VI

List of Figures .............................................................................................................. IX

List of Table .................................................................................................................. X

Abbreviations ............................................................................................................... XI

Chapter 1. Introduction .................................................................................................. 1

1.1 Introduction to Image Steganography .................................................................. 1

1.2 Problem Statement ............................................................................................... 2

1.3 Objective .............................................................................................................. 2

1.4 Scope and Limitations .......................................................................................... 3

1.5 Report Organization ............................................................................................. 3

Chapter 2. Literature Review ......................................................................................... 4

Chapter 3. System Analysis and Feasibility Study ........................................................ 5

3.1 Requirement Analysis .......................................................................................... 5

3.1.1 Functional Requirements .............................................................................. 5

3.1.2 Non-Functional Requirements ...................................................................... 6

3.2 Feasibility Study ................................................................................................... 7

3.2.1 Economic Feasibility .................................................................................... 7

3.2.2 Technical Feasibility ..................................................................................... 7

3.2.3 Operational Feasibility .................................................................................. 7

3.3 Process Model ...................................................................................................... 8

3.3.1 DFD Diagram................................................................................................ 8

Chapter 4. System Design ............................................................................................ 10

4.1 System Architecture ........................................................................................... 10

VII
4.2 Class Diagram .................................................................................................... 11

4.3 Activity Diagram ................................................................................................ 13

4.4 Algorithm ........................................................................................................... 14

Least Significant Bit (LSB) Replacement Method .............................................. 14

Message embedding procedure: ........................................................................... 15

Message extraction procedure: ............................................................................ 15

Chapter 5. Implementation and Testing ....................................................................... 16

5.1 Implementation................................................................................................... 16

5.1.1 Tools Used .................................................................................................. 16

5.1.2 Implementation of Encryption and Embedding Process ............................. 16

5.1.3 Implementation of Extraction and Decryption Process .............................. 17

5.1.4 Implementation of Client-Server Model ..................................................... 18

5.2 Testing ................................................................................................................ 19

5.2.1 Unit Testing ................................................................................................ 19

5.2.2 Integration Testing ...................................................................................... 22

5.2.3 System Testing ............................................................................................ 27

Screenshots .......................................................................................................... 28

5.3 Analysis .............................................................................................................. 34

5.3.1 Computational Analysis .............................................................................. 34

5.3.2 Statistical Analysis ...................................................................................... 35

Chapter 6. Conclusion and Future Enhancement ......................................................... 37

6.1 Conclusion .......................................................................................................... 37

6.2 Future Enhancement ........................................................................................... 37

References .................................................................................................................... 38

VIII
 List of Figures

Figure 3-1 Use-Case Diagram for Image Steganography..............................................5

Figure 3-2 Level 0 DFD for Image Steganography.......................................................8
Figure 3-3 Level 1 DFD for Image Steganography.......................................................9
Figure 4-1 System Architecture for Image Steganography......................................... 10
Figure 4-2 Class Diagram for Image Steganography.................................................. 11
Figure 4-3 Activity Diagram for Image Steganography..............................................13
Figure 4-4 LSB Replacement.......................................................................................14
Figure 5-1 Message embedding procedure time computation.....................................34

IX
 List of Table
Table 5-1 Histogram Analysis of cover and stego images.......................................... 35

X
 Abbreviations
AES: Advanced Encryption Standard
API: Application Program Interface
DFD: Data Flow Diagram
ICT: Information and Communication Technology
IDE: Integrated Development Environmen
IP: Internet Protocol
JPG: Joint Photographic Experts Group
LSB: Least Significant Bit
PNG: Portable Network Graphics
SMTP: Simple Mail Transfer Protocol
TCP: Transmission Control Protocol
UI: User Interface

XI
 Chapter 1. Introduction
1.1 Introduction to Image Steganography

Due to advances in ICT, most information is kept electronically. Consequently, the

security of information has become a fundamental issue. Maintaining the integrity and
confidentiality of sensitive information, blocking the access of sophisticated hackers
is very important. Steganography is a technique of hiding information in digital media
[1]. Steganography is the art of hiding the fact that communication is taking place, by
hiding information in other information. Many different carrier file formats can be used, but
digital images are the most popular ones because of their frequency on the internet. Similarly,
Cryptography is the science of protecting information by transforming it into a secure format.
In contrast to cryptography, steganography is not to keep others from knowing the hidden
information but it is to keep others from thinking that the information even exists.

Both Steganography and Cryptography are data hiding techniques with different but
related fields of applications. Steganography prevents the detection of hidden
messages whereas cryptography converts ordinary plain text into encrypted
unintelligible text and vice-versa by applying encryption tools. Generally,
steganography is used to supplement encryption. An encrypted file or message may
also be hidden using steganography, so even if the encoded stego image is decoded,
the hidden message is still not understandable. Thus, together cryptography and
steganography can provide a powerful basis for data security.

This project intends to produce a security tool based on steganography and

cryptography techniques for sending and receiving sensitive information over a
network. The program first encrypts the message data using AES encryption and then
embeds the result of encrypted data in a provided image file using LSB replacement
technique for steganography. The system also provides the feature for extracting the
hidden data from the corresponding image file’s LSBs and decrypting the extracted
data using AES decryption for eventually finding the original message. The generated
stego image can either be saved and sent over a user-chosen platform or can also be
sent directly using email sending feature of the system. This project also provides a

1
 client/server module for stego image transmission. The client-server connection is
established using TCP socket programming on a user given port. The sender (server
or client) can encrypt-embed message onto a cover image and send the generated
stego image to the receiver (client or server) through this connection. On receiver’s
side, the received image can be used to obtain the original message.

1.2 Problem Statement

With growing digitization in every field, digital security has become a fundamental
aspect. Also, because of development in the internet technology, digital media can be
transmitted conveniently over the network. This calls for security over the internet.
Throughout history steganography and cryptography have been used to secretly
communicate information between people. In the past, means of cryptography and
steganography were carried out using traditional methods of pen and paper, using
invisible ink, etc.

Therefore, with the increasing use of communication of information over digital

medium, security for digital methods are to be developed. Steganography hides data
onto a stego medium. Steganography along with cryptographic tools can ensure even
more data security.

1.3 Objective

The main objectives of this project are:

• To produce security tool based on steganography and cryptography techniques

combined.
• To avoid drawing suspicion to the existence of a hidden message.

2
1.4 Scope and Limitations

The scope of this project is to limit unauthorized access and provide better security
during message transmission. To meet the requirements, simple and basic LSB
approach of steganography had been used. The program first encrypts the message
data using AES algorithm and embeds the result of encrypted data in the provided
image file using steganography technique for sending over the network. The system
also provides the feature for extracting the hidden data from the corresponding image
file and decrypting the extracted data for eventually finding the original message.
Further, the system provides a client-server section for image transmission. The client
server connection is established using TCP socket programming.

Steganography means hiding data into another data. It can be used to hide data such
as text, image, audio, video etc. within a cover image, video etc. While the system
program provides a way to hide text data into a cover image file, it is limited only to
data of the mentioned types i.e., text data onto image data.

1.5 Report Organization

Chapter 1 includes subtopics as introduction to image steganography, problem

statement, objectives, scope and limitations.

Chapter 2 contains literature review.

Chapter 3 comprises of requirement analysis, feasibility study and process models of

the system. The requirements analysis further consists of functional and non-
functional requirements. Economic, technical and operational feasibilities are some
listed feasibilities. Process model includes DFD diagrams.

Chapter 4 consists of system architecture, class diagram activity diagram and

algorithm used.

Chapter 5 includes tools used and testing. Testing comprises of unit testing,
integrated testing and system testing.

Chapter 6 includes conclusion and future enhancements.

3
 Chapter 2. Literature Review
The word steganography is derived from the Greek words stegos meaning cover and
grafia meaning writing [2]. In Image steganography the information is hidden
exclusively in images. Steganography is the art and science of secret communication.
It is the practice of encoding/embedding secret information in a manner such that the
existence of the information is invisible. The actual files can be referred to as cover
text and the cover image. After inserting the secret message, it is referred to as stego-
medium. A stego-key is used for hiding encoding process to restrict detection or
extraction of the embedded data [3].

For more secure data transfer, cryptography is used along with steganography. In
cryptography the message is encrypted using encryption algorithm along with secret
key and transferred it to the other end, then receiver can decrypt it and get original
message by using decryption algorithm. The grouping of these two approaches
enhances the security of data. The combination of these two methods will satisfy the
requirement such as capacity and security for data transmission over an open channel.
If the attacker were able to detect the steganography technique, they would still have
to require the cryptographic decoding way to de-cipher the encrypted message and
vice versa [4].

The Least Significant Bit Replacement Algorithm is a commonly used straightforward

steganographic algorithm used to embed secret information inside a cover medium. In
this method, the least significant bits of the original data in the cover medium are altered
based on the secret message. In the case of digital images, the alteration is done only at
the least significant bits of the original image so as to reduce the effect of degradation of
the original image. By inserting the secret message only at the least significant bits, the
perceptibility of the original image is not much affected [5].

Currently, AES is regarded as the most popular symmetric cryptographic algorithm.

In 1997, NIST advocated making the AES to replace DES to meet the need of
message encryption in 21st century [6].

This system implements the image steganography technique of LSB replacement

algorithm along with AES encryption for cryptography.

4
 Chapter 3. System Analysis and Feasibility Study
3.1 Requirement Analysis

3.1.1 Functional Requirements

Figure 3-1 Use-Case Diagram for Image Steganography

Figure 3-1 shows the use-case diagram of the system. Sender and receiver are the primary
actors. They provide required information and data to the system and receive the required
output. Sender’s inputs include cover image file, text message (directly or via a txt file) to
be hidden and a secret key for message encryption. Upon receiving all valid inputs, the
system generates a stego-image which consist of the text message embedded inside it. The
sender can then either email the stego-image to the receiver or save the stego image to
transfer it over a different media. On receiver’s side, the inputs

5
include the received stego image and associated secret key. The system processes the
provided inputs and generates the hidden message in readable text. If needed, the
receiver can save the generated message as text file.

In client/Server model of the system, both server and client can act as sender or
receiver. They can send stego-images to one another and save or extract hidden
messages upon receiving them.

3.1.2 Non-Functional Requirements

Reliability: The system program is reliable as whenever provided with correct

information/input, it always gives the corresponding image (on sender’s side) or the
hidden message (on receiver’s side).

Performance: The system performs well as when provided with the correct inputs, it
efficiently provides correct output within seconds.

Usability: The system is quite easy to use as anyone with simple skills regarding
using a windows computer can use it to create a stego image or extract the messaged
hidden on a received stego image.

Maintenance: The system doesn’t really need to be maintained regularly but is open
to any future enhancement, whenever required.

6
3.2 Feasibility Study

Feasibility Study is a test of the system according to its workability, impact of the
organization, ability to meet user needs and effective use of the resources. A
Feasibility Study is generic in nature and can be applied to any type of project, be it
for systems and software development, making an acquisition, or any other project.
We can test the system by different type of the feasibilities.

3.2.1 Economic Feasibility

The system is economically feasible as all the tools and resources required are either
cheap or free.

3.2.2 Technical Feasibility

The technical requirements (for e.g. hardware and software specifications of device
for operating the system) is easily available and the system can be operated by users
with simple knowledge regarding the required technologies.

3.2.3 Operational Feasibility

The system is user friendly as it is easy to use and operating the system doesn’t
require too complex skills.

7
3.3 Process Model

3.3.1 DFD Diagram

Level 0 DFD

Figure 3-2 Level 0 DFD for Image Steganography

Figure 3-2 explains the Level 0 DFD of the system. The system takes the input as
cover image (original image) and a secret text message (directly or via a txt file) from
the sender. This application also uses a secret pre-shared key for encryption. With
these inputs provided by the sender, the application generates a stego image which is
sent to the receiver. The system then takes the associated key from the receiver and
displays the hidden message to the receiver. The client server section uses TCP
socketing for sending and receiving the stego images generated by using similar
techniques as mentioned above.

8
Level 1 DFD

Figure 3-3 Level 1 DFD for Image Steganography

Figure 3-3 represents the Level 1 DFD of the system. In this level, more detailed
process of this system is explained. The main task described in this level is that the
text message is encrypted using AES encryption and that encrypted message is
embedded to the image by replacing least significant bits of the image using LSB
technique. Then the generated stego image is sent to receiver through a data
transmission channel. The image can be transmitted using the email feature or on
client-server section, over the established TCP connection. This system then takes the
stego image and shared key from the receiver and first extracts the cipher text
message (encrypted hidden message) from the image and with the provided key it
decrypts the cipher message and displays the original hidden message to the receiver.

9
 Chapter 4. System Design
4.1 System Architecture

Figure 4-1 System Architecture for Image Steganography

Figure 4-1 shows the basic architecture of the system. User can interact with the
system to receive expected outputs.

The system is divided into two parts: normal mode and client/server mode. When
starting the program normally, the user needs to provided necessary input to get
necessary output which is either stego-image (when embedding a cover image) or
extracted message (when extracting a stego-image for hidden message).

When starting the program in client/server model, firstly a TCP server socket needs to
be started. The user provides port number to be associated to the server. When a client
connects to the server socket on particular port, the client-server connection starts.
Then the client and server can communicate with each other through associated port
and IP as the communication medium. For communicating sensitive information,
stego-image can also be used. The sender side can generate a stego-image containing
a hidden message and a key shared by both ends. Then the receiver can use the shared
key to extract the hidden message from the received stego-image.

1
0
4.2 Class Diagram

Figure 4-2 Class Diagram for Image Steganography

Figure 4-2 shows the class diagram for the system. Different classes are involved to
obtain the preferred result the system. Different classes’ instances are created and
used whenever necessary. Some major classes involved in the system process are:

ImageSteganography: This is the main class of the system which is accessed first
when starting the system. Then, preferred mode of the program can be chosen.

EncryptionPanel: This is the class that contains the panel for encrypting and
embedding the cover image file. It takes cover image file, secret message (directly or
via a txt file) and shared key as inputs, then calls necessary functions from different
classes to reach the goal.

11
Encode: This class contains the necessary functions for encrypting the given input
string and embedding the encrypted string onto provided cover image. It can be
accessed from the EncryptionPanel class, which may create and use instance of this
class whenever stego-image is to be created.

DecryptionPanel: This is the class that contains the panel for extracting and
decrypting the hidden message from the stego image file. It takes the stego image file
and shared secret key as inputs, then calls necessary functions from different classes
to reach the goal.

Decode: This class contains the necessary functions for extracting the encrypted text
from the stego image as well as decrypting the encrypted text to receive the original
message. It can be accessed from the DecryptionPanel class, which may create and
use instance of this class whenever hidden message is to be extracted.

SendMail: This class is responsible for sending the stego-image given by the Encode
class, via email. It contains the frame consisting of the user interface.

12
4.3 Activity Diagram

Figure 4-3 Activity Diagram for Image Steganography

13
Figure 4-3 explains basic activity of the system. The program can either be started in
regular mode or client server mode. Either of the modes will support image
steganography procedures.

The ‘encrypt and embed’ option takes a cover image, secret key (directly or via a txt
file) and the message to be hidden as input. Then it provides a stego-image as output.
The ‘extract and decrypt’ option takes the generated stego image and associated key
then extracts the hidden message. The message can then be saved as a .txt file.

In client server mode, firstly a TCP server socket needs to be started. The user
provides port number to be associated to the server. When a client connects to the
server socket on particular port, the client-server connection starts. Then the client
and server can communicate with each other through associated port and IP as the
communication medium. Sender loads the image, provides message (directly or via a
txt file) and key. Then stego image is generated. The stego image can be saved or sent
to the receiver. Receiver loads the stego image sent by the sender then enters
associated key to extract the hidden message. Finally, the hidden message is extracted
from the received stego image which can also be saved as a file.

4.4 Algorithm

Least Significant Bit (LSB) Replacement Method

LSB based technique is simple approach in which message bits are embedded in the
least significant bits of cover image [7]. In this technique, the least significant bit of
cover image is used to hide the secret message. The least significant bit of each pixel
of an image is altered to a bit of a message that is to be hidden [8].

Figure 4-4 LSB Replacement

1
4
 Message embedding procedure:

Input: Cover image, secret key and message.

Output: Stego image.

1. Read the cover image and secret text information which is to be embedded
into the image.
2. Convert the secret information into encrypted text by using AES encryption
(for java, Base64Encoder can also be used) and secret key shared by receiver and sender.
3. Add the value of encrypted text’s length at the beginning of the text along with
a ‘/’ character (to mark the actual cipher length during decryption).
4. Convert encrypted text message into binary form – which will give the text
message’s characters’ bits.
5. Find LSBs of each RGB pixels of the cover image.
6. Embed the bits obtained on step 4 into LSBs of RGB pixels of step 5.
7. Continue the procedure until the secret information is fully hidden in cover
image file.

Message extraction procedure:

Input: Stego Image and associated secret key.
Output: Hidden message.

1. Read the stego image and secret key.

2. Retrieve LSBs of each RGB pixels of the stego image.
3. Convert binary strings formed by every 8 RGB pixels of step 2 to character
and append the characters to a string builder.
4. Upon finding the first ‘/’ character from string builder of step 3, save its
previous characters as text length and discard all characters till that index.
5. Continue the process for more (text length obtained on step 4 * 8) times to
fully extract the hidden encrypted text.
6. Using the secret key and AES decryption (Base64Decoder for java), decrypt
secret information obtained on step 4 to get original information.

1
5
 Chapter 5. Implementation and Testing
5.1 Implementation

5.1.1 Tools Used

While many additional software tools were used when developing the project, there
was no need for any special hardware requirement. The system was built on a 64-bit
computer running with Windows 10. Various Software tools used include:

• IDE: NetBeans 8.0.2.

• Programming Language: Java programming language.
• Swing- GUI widget toolkit for java: Used for creating GUI.
• Java AWT packages: For event handling and working with files and images.
• Java Mail bean: Used to send stego image via Email.
• Java Base64 Encoder: Used to generate cipher text of given message.
• Java Base64 Decoder: Used to decrypt extracted cipher text.
• Java Socket Programming: Used for client-server connection.
• Online Image Histogram Generator (sisik.eu/histo): For image histogram.
• Additionally, various java packages and libraries were also used.

5.1.2 Implementation of Encryption and Embedding Process

The implementation of encoding a message onto a cover image is completed in two steps.
The process takes in an image file (JPG, PNG or JPEG), a message string (directly or by
uploading a .txt file consisting the message string) and a key string (up to 16 characters)
as inputs. Firstly, the message is encrypted using AES 128-bit encryption. Java Base64
encoder API has been used. The encryption process takes in a 128-bit key string and the
required message as inputs. If the user-entered key is not 128-bit i.e., 16 characters, the
provided key string is right-padded with the letter “a” in order to reach the required length
of 16 characters. However, if the entered key string exceeds the required length of 16
characters, error is encountered. The key string is then converted to the required key
datatype. The AES cipher instance is taken which is initialized with ENCRYPT_MODE
and given key. The message is then encrypted and cipher string is generated. After the
encryption process of given message, the length of the generated

1
6
cipher text multiplied by 2 along with a “/” character, is added to the beginning of that
cipher text. At this point, the program checks for whether or not the image is sufficient to
hold all the character bits of the whole cipher string. If the size of given image fails to
meet this criterion, a dialog box indicating this error is displayed. If not, the LSB
replacement process of the image with the whole string is done. The LSBs of each RGB
pixel is replaced with the character bits of cipher text until all the cipher text bits are
embedded onto the image. The rest of the image bits are left unchanged. However, if the
total number of bits in the cipher text is greater than the total number of image’s RGB
bits, a dialog box, indicating that a larger image is required is displayed. After the
embedding process completes, the resulting stego image is displayed on the screen. The
stego image can then be sent via email or saved to the operating computer. The emailing
process takes in required inputs and sends the email using Gmail SMTP host. If any input
like sender’s credentials are invalid, an error message will be displayed. By default, the
image is saved as a JPG image file. If the user explicitly provides a different extension
like PNG, JPEG or BMP, the image is saved accordingly. However, if some other
unsupported file extension is entered, the image will still be saved as JPG file.

5.1.3 Implementation of Extraction and Decryption Process

The implementation of extraction process of the hidden message from a stego image is
done in two steps, as well. The process takes in a stego image file (generated by the
program), and the associated key string as inputs. Firstly, the cipher text embedded onto
the stego image is extracted. Starting from the beginning of the image i.e., the very first
pixel, the LSBs of every 8 bits (one pixel consisting of 3 bits- RGB) are taken and
converted to characters. Upon finding the first “/” character, the process is stopped and
the string made by characters leading up to the “/” is separated and taken as integer and
divided by 2, indicating the total length of embedded cipher. This length is taken as
reference for stopping the extraction process. Then the rest of the characters until the
cipher length are extracted in the same way as mentioned earlier. After the cipher length
extraction process, the cipher text thus obtained is decrypted using 128-bit AES
decryption. Java Base64 decoder API has been used. The decryption process takes the
associated 128-bit key string as input. If the user-entered key is not 128-bit i.e., 16
characters, the provided key string is right-padded with the letter “a” in order to reach

1
7
the required length of 16 characters. However, if the entered key string is invalid or
the provided stego image is not a stego image generated by the same program, a
dialog box indicating input error is displayed. The key string is then converted to the
required key datatype. The AES cipher instance is taken which is initialized with
DECRYPT_MODE and given key. The message is then decrypted and the original
message is obtained and displayed on the screen. The user can then save the displayed
message as text file on the operating computer’s memory or discard it. The text file is
saved as a .txt file by default.

5.1.4 Implementation of Client-Server Model

The client-server section of the program is implemented using java socket programming.
The server thread is started on a user-entered port number using ServerSocket tag. If the
server couldn’t be started on given port number, a dialog box indicating the failure is
displayed. Whenever a client wishes to establish a connection with the server, the client-
side thread is started and the Socket tag is used along with server’s port number (entered
by user) and the server’s IP address (localhost by default). However, if a server at given
port and IP is not found, a dialog box indicating the error is displayed. The Socket tag is
used to accept the client’s connection request, on server side. After the connection has
been established, both server and client can send and receive stego images from each
other. The maximum number of clients that can be handled by server at once is one. The
sender side can generate and send stego images to receiver’s side by providing the
required inputs. On receiver’s side, whenever an incoming image is detected, an image
viewer window is prompted automatically. The receiver is then provided with options to
either extract the hidden message from the image or to save the image file. The extraction
process requires the associated key. Further, after providing the correct key, as used by
the sender and extracting the hidden message, the receiver can save the extracted message
on operating computer’s memory. Once the client side closes the connection to the server,
the server will be open for new client requests. Also, at this point, if the server tries to
send a stego image, a message indicating that the program is unable to send that image is
displayed. Similarly, if the client tries to send an image, after server has been closed, error
is detected and the message indicating that the server is inactive is displayed.

1
8
 5.2 Testing

5.2.1 Unit Testing

• Message encryption and embedding module:

Test Case 1: String Encryption (using 128-bit AES encryption)

Input message string: “This is a test string.”
Input key string: “hello12”
Expected output: Encrypted
string. Obtained output:

Output 1: Encrypted text

Remarks: String Encrypted successfully.

Test Case 2: Image’s LSB replacement

Input image: 800 x 500 jpg image

Input message string: “test”

Expected output: Image with LSBs changed with message string’s

19
 Output: 800 x 500 PNG image

Output 2: Stego-image

Output 3: Image and message bits

Remarks: Message embedded in image successfully.

• Message extraction and decryption module:

Test Case 3-1: String Decryption (using 128-bit AES decryption)

Input message string: Output of test case 1
(vyoeY0tNi5amaMLoPoU5OMrn7xCV3vQjyZMGsCRzWOw=)
Input key string: “hello”
Expected output: Decrypted string.

20
Obtained output:

Output 4: Decryption error

Remarks: Invalid Key Error.

Test Case 3-2: String Decryption (using 128-bit AES decryption)

Input message string: Output of test case 1
(vyoeY0tNi5amaMLoPoU5OMrn7xCV3vQjyZMGsCRzWOw=)
Input key string: “hello12”
Expected output: Decrypted string.
Obtained Output:

Output 5: Decrypted text

Remarks: String decrypted successfully.

Test Case 4: String extraction from image (Up to first 32 RGB bits)
Input image: Output from Test Case 2 (800 x 500 PNG image)
Expected output: Extracted hidden string/message.
Obtained Output:

Output 6: String extracted from image

Remarks: String extracted successfully.

21
• Client-Server module

Test Case 5: Starting Server Socket

Input port number: 80
Expected output: Server Started and waiting for client
message. Obtained Output:

Output 7: Server started

Remarks: Server socket started successfully.

Test Case 6: Client-Server connection

Input port number: 80
Input Server IP: localhost (127.0.0.1)
Expected output: Client-Server connection established.
Obtained Output:

Output 8: Server connected to client Output 9: Client connected to server

Remarks: Client-Server connection established successfully.

5.2.2 Integration Testing

The test cases 1 and 2 were integrated together and tested for output. An image file, a
message and a key were provided as inputs. After the encryption process of given
message, the length of the generated cipher text along with a “/” character, was added
to the beginning of that cipher text and LSB replacement process of the whole string
was done. The generated stego image was saved for further testing.

22
Test Case 7: Encryption and LSB replacement process integration.

Input message: A .txt file containing String- “This is a test.”

Input Key String: “hello12”
Input Image: 484 x 319 PNG image

Expected output: Stego image

Obtained Output: 484 x 319 PNG image

Output 10: Stego-image

Remarks: String extracted successfully.

Also, the message extraction and decryption process, as referenced on test cases 3 and
4, were tested as a whole. At first, a random key string and stego image produced by
test case 7 were provided as inputs, which produced an error. Later the stego image
produced by test case 7 and the correct associated key were provided as inputs, then
the hidden message was extracted from the image and displayed on the screen.

23
Test Case 8-1: Message Extraction and Decryption process integration.
Input Key String: “hello”
Input Image: Stego image generated on test case 7

Expected output: Hidden message in decrypted form

Obtained Output: Input error.

Output 11: Input error

Remarks: Invalid Key.

24
Test Case 8-2: Encryption and LSB replacement process integration.
Input Key String: “hello12”
Input Image: Stego image generated on test case 7

Expected output: Hidden message in decrypted form.

Obtained Output: Hidden message in decrypted form.

Output 12: Extracted and decrypted message

Remarks: Hidden message extracted successfully.

Test Case 9: Creating Client-Server connection and sending stego images

Input port number (on both client and server side): 80
Input message (on server side): “Client-Server test”
Input Key String (on server side): “hello”
Input image (on server side): 800 x 500 jpg image

25
Expected output: Image sent from server, received on client side and extracted message
Obtained Output: Image sent from server, received on client side and extracted message

Output 13: Image sent from server Output 14: Image received on client side

Output 15: Obtained image on client side

Output 16: Extracted and decrypted message on client side

Remarks: Stego image sent, received and message extracted successfully.

26
5.2.3 System Testing

On program’s normal mode; firstly, to test the message hiding process, the encrypt
and embed option was chosen. A cover image file, a key and a message to be hidden
were provided as inputs. After being provided with valid inputs, the system was
successfully able to generate a stego image. After receiving the stego image file, it
was saved to the hard drive. Sending the generated stego image file over email was
also successfully tested. Sending the image over the internet required inputs such as
sender’s email and password as well as the receiver’s email and password. The
provided email attributes need to be correct and the sender’s id should have
permission to send emails via windows applications.

After the successful generation of the stego image, it was tested for message
extraction. To test the image sent over the email, the image was firstly downloaded
using the option provided by Gmail. After choosing the extract and decrypt option in
the program, the stego image and associated key were provided as inputs. The
program was able to extract the hidden message from the stego image in decrypted
form. The hidden message was displayed on the screen. After receiving the hidden
message, it was saved to the hard drive as text file.

On the client-server mode of the system; firstly, server was started on a random port.
On client side, the same port number was provided with local host as default server
IP. After the connection was successfully established, a stego image was generated on
server’s side by providing necessary inputs. The stego image was then sent to the
client where an automatic image viewing frame was prompted. Then by selecting the
“extract and decrypt” option provided on the frame, the message hidden on the server
side was successfully displayed on the screen.

2
7
Screenshots

Screenshot 1: Program mode selection

Screenshot 2: Normal mode user Interface

28
Screenshot 3: “Encrypt and Embed” screen

Screenshot 4: “Extract and Decrypt” screen

29
 Screenshot 5: Email window

Screenshot 6: Incorrect emailing credentials

Screenshot 7: Email sent successfully

30
Screenshot 8: Client/Server mode: User Interface

Screenshot 9: Server side

31
Screenshot 10: Encryption window at Sender (server) side

Screenshot 11: Client side

32
 Screenshot 12: Automatic image viewer at receiver (client) side

Screenshot 13: Hidden message extraction at receiver side

33
5.3 Analysis

The algorithms implemented in this project are analyzed from various dimensions.
Computational analysis based on message embedding and message extracting
procedure time and statistical analysis based on histogram analysis of original image
and stego image has been done.

5.3.1 Computational Analysis

Computational analysis is an approach to measure the computational time required for

the execution of an algorithms. The value of message embedding and message
extracting procedure time are measured for computational performance for different
set of images in milliseconds and is observed from the bar graphs given below.

Message embedding procedure time (ms)

180

160

140

120

100

80

60

40

20

0
Image 1(png) Image 2(png) Image 3(jpg) Image 4(jpg)

Message 1(17 B) Message 2(1.55 KB)

Figure 5-1 Message embedding procedure time computation

Figure 5-1 shows the computational time for message embedding procedure. For the
analysis, different images of different file format with different sizes are taken. Each
image is embedded with two different messages kept on two different text files- one
message being short and the other longer. For better result, each unit was executed 5
times and the average execution times of each were taken. It is seen that as the image
sizes increases, the time consumed for the execution of this algorithm increases as well.
3
4
Similarly, the computational time for message extraction from stego images generated
during message embedding procedure’s time computation analysis were taken and the
hidden messages were extracted. Execution of each stego image was done 5 times and
the average execution times were taken. The process showed that the time consumed
by the message extraction procedure was nearly same (0 millisecond) for all stego
images regardless of the image or message sizes.

5.3.2 Statistical Analysis

Statistical analysis has been carried out with the help of histogram analysis. An online
histogram generator tool has been used. The histogram of an image normally refers to
a histogram of the pixel intensity values. It is the most commonly used graph to show
frequency distributions. An image histogram is a type of histogram that acts as a
graphical representation of the tonal distribution in a digital image.

The tool basically creates 256 bins for each color (red, green, blue) and greyscale (luma)
intensity. The number of bins is shown on the horizontal axis. The tool then loops
through every image pixel and counts the occurrence of each intensity. The counts of
occurrences in each bin are then displayed on vertical axis. Counts for each pixel
intensity are normalized to range 0 to 255 before they are displayed on the graph.

Table 5-1 Histogram Analysis of cover and stego images

Before hiding message After hiding message

Cover Image Stego Image

3
5
 Histogram of cover image in red plane Histogram of stego image in red plane

Histogram of stego image in green

Histogram of cover image in green plane plane

Histogram of cover image in blue plane Histogram of stego image in blue plane

Histogram of cover image in greyscale Histogram of stego image in greyscale

(luma) plane (luma) plane

Table 5-1 shows that the histograms of cover image and stego image on color or
greyscale plane do not have any significant difference. This shows that the change in
LSBs of fixed number of pixels do not have any strong effect on the image’s pixel
intensity values in red, green, blue or greyscale plane.

3
6
 Chapter 6. Conclusion and Future Enhancement
6.1 Conclusion

The final product of the project is a system that can take a cover image file, hidden
message and a secret key as input and provide a stego-image as output as well as can
extract the secret message hidden in the stego-image when provided with a key and the
corresponding stego image. On sender’s side, the hidden message is first encrypted with
AES technique with the help of provided secret key and the encrypted text is then
embedded onto the cover image file to produce stego-medium. The embedding process
follows image’s LSB replacement algorithm. On receiver’s side the process includes
extraction of the encrypted message then its decryption with original key for obtaining the
original message. During extraction of hidden cipher text, the bits at LSB are extracted
and finally through a series of steps, converted to the encrypted text.

6.2 Future Enhancement

If needed, the system will be open for future enhancements. The future enhancements
might include audio/video steganography with improved algorithms.

37
 References

[1] S. Singh, Literature Review On Digital Image Steganography and Cryptography

Algorithms, 2015.

[2] A. Soni, J. Jain and R. Roshan, "Image steganography using discrete fractional
Fourier transform," Intelligent Systems and Signal Processing (ISSP), 2013.

[3] Akhtar, N.; Johri, P.; Khan, S., "Enhancing the Security and Quality of LSB
Based Image Steganography," Computational Intelligence and Communication Networks
(CICN), 2013.

[4] K. Joshi and R. Yadav, "A New LSB-S Image Steganography Method Blend with
Cryptography for Secret Communication," in Third International Conference on Image
Infonnation Processing, 2015.

[5] S. Sugathan, "An Improved LSB Embedding Technique for Image

Steganography," in 2nd International Conference on Applied and Theoretical Computing and
Communication Technology (iCATccT), 2016.

[6] L. Xiaoqin, L. Wei, C. Xiuxin, Z. Xiaoli and D. Zhengang, "Application of the

Advanced Encryption Standard and DM642 in the image transmission system," in 7th
International Conference on Computer Science & Education (ICCSE), Melbourne, 2012.

[7] M. R. Garg, "Comparison Of Lsb & Msb Based Steganography In Gray-Scale

Images," International Journal of Engineering Research and Technology (IJERT), vol. 1, no.
8, 2012.

[8] E. Walia, P. Jain and Navdeep, "“An Analysis of LSB & DCT based
Steganography," Global Journal of Computer Science and Technology, vol. 10,
no. 1, 2010.

38