CHAPTER 1

INTRODUCTION

The Internet is fast becoming indispensable in our daily lives. The Digitalization has the
greatest potential to alter the way we live. In today’s era of the digitalized world, security is an
important concern. Security problems start to become apparent when information is being
delivered from node to node over the network. The number of threats has been increasing at a
wider rate, therefore, strong security techniques need to be deployed. Cryptography is one of the
prime techniques for providing information security. In traditional cryptographic methods,
enormous computational power and complicated algorithms are common, taking much time and
money to encode and decode a secret message. This chapter starts with an overview of
cryptography along with an introduction to the idea of visual cryptography. This research work
major focuses on secret sharing concepts using visual cryptographic aspects. Its characteristics,
various research challenges, and practical applications have been discussed here. A number of
researchers have already dug a lot in this area, but still, some open challenging research issues
need more exploration. All these critical issues which motivated us for this research work have
been discussed along with the other key points. The problem statement and the key objectives of
this research work along with research formulation have been described in details. The scope of
this research work and organization of the complete thesis have been given at the end of the
chapter.

1.1 Cryptography
Cryptography is concerned exclusively with message confidentiality. It was first
introduced when humans need a secret language for communication. It is the practice and study
of techniques for protecting the information, or message from being read. To ensure secrecy in
the communication of military leaders, diplomats and protecting information in many types of
civilian systems, encryption techniques have been used. In recent decades, the cryptography field
has expanded beyond confidentiality concerns. It includes techniques for message integrity
checking, sender/receiver identity authentication, digital signatures and secure computation,
among several others. Advancement in the traditional method of cryptography has led to the
invention of many new distinct types of technologies such as network cryptography, visual
cryptography, etc.

Today our lives are completely digitalized and security of personal information has
become an integral part of nearly everyone’s daily life. Nearly all our confidential information
has been stored in the databases of the government, banks, health care services and many others
public parties. Cryptography is used to protect the right to privacy and the right to communicate
confidentially.

Cryptography is not only used for protection of data from hackers, but it is also utilized
for providing the user authentication during communication. Symmetric cryptography,
Asymmetric cryptography, and hash functions have been commonly popular types of
cryptography for applying it to several security problems. For referring the process of encryption
and decryption following common terms have mostly been used in literature:
 Plain Text: Plain text refers to the text which needs to be encrypted.
 Encipher / Encryption: It is the process of converting the plain text into cipher text form
by means of a predetermined system key.
 Decipher / Decryption: The process by which conversion of enciphered text to the
equivalent plain text is possible with the help of a system key.
 Cryptanalysis: It refers to the attempt made for the conversion of encrypted messages to
plain text without initial knowing the encryption algorithm and/or key employed in the
encryption.
 Key: It is usually a sequence of random binary digits used at the setup time in the
cryptosystem for the purpose of encrypting or decrypting the text. Sometimes the same
key is also used for producing another keys for the system.
In the past eras, various encryption methods of traditional cryptography have been used
to protect the information. The protection of visual information (e.g., sensitive picture,
handwritten notes, digital data and printed text, etc.) is a challenging security issue. It has
become an area of prime concern since last two decades after the digitalization of the world has
begun. Secret image sharing is an important research area that combines methods and
techniques, both from cryptography and image processing. It helps in protecting the visual
information.

1.2 Visual Cryptography

Cryptography has a long and fascinating history within the security domain. Handling of
sensitive images carrying confidential information is of prime concern in several departments
such as sharing the maps over the internet in the military and in many others commercial sectors.
To handle the security problems of sensitive images, various image secret sharing schemes have
been evolved. One of the techniques named as Visual cryptography (VC) has been developed
by Naor and Shamir [1] in the year 1995 to handle secret sharing for images.

VC is an approach in which a secret image containing confidential visible information is

encrypted in a perfectly secure way such that the decryption can be performed directly by the
human visual system (HVS), without the assistance of computers. VC allows encrypting any
visual information such as printed text, handwritten notes, and pictures. It eliminates complex
computation during decryption process, and the images can be restored by doing stacking
operation on its shares. It combines the feature of perfect ciphers creation and secret sharing in
cryptography. Secret image is usually divided into two or more pieces known as shares. When
the required number of shares, print on transparencies and then superimposed, the secret images
get recovered.

Naor et al.[1] introduced the technique of VC in which the binary image is decomposed
into n number of shares. Figure 1.1 shows an example of share creation and recovery of a secret
image using visual cryptography. In the scheme of , shares when stacked over one another
reveals the original secret image. Naor scheme is quite suitable for a binary image. The shares
created in the original image are determined by randomly selecting pairs of sub-pixel matrices for
black and white pixels [2].
 Figure 1.1: Original image, Halftone, Share 1, Share2 and Decrypted image

VC scheme suggested by Naor et al.[1] requires no computer participation in any

situation for decryption. Visual cryptography combines the notion of the perfect secret with a
random image for the purpose of secret sharing [3]. The next section describes the common
characteristics of VC schemes.

1.3 Characteristics of Visual Cryptography

Generally, visual cryptography is used to preserve the privacy of raw images. Most of the
business organizations need to protect data from disclosure[4]. As the world is more connected
by computers, most organizations are afraid to store data in a single computer. So, VC provides a
solution to distribute the data in several places and destroy the original one. As and when a need
of original data arises, it could be reconstructed from the distributed shares. The information will
not be available centralized at one point. The following characteristics made visual cryptography
very popular among researchers and academicians to utilize it in various domains of security.
 Its implementation is very easy. The person who has even less knowledge about the
scheme can implement it without any hassle.
 It does not require any intervention of computer or any other hardware or software device
in most of the situations during the restoration of the secret.
 As there is no decryption algorithm, so its decryption is very simple. When all the
required shares superimposed on each other, the original secret image is retrieved.
 Its computational cost is very low, as it does not involve any cryptographic computation.
However, on the other hand, the quality of the image is compromised in the scheme.
Visual cryptography possesses mainly three characteristics: Perfect security, secret
restoration without the aid of a computing device, and robustness against lossy compression [5]
This simple and yet secure approach made visual cryptography a popular and interesting research
field. Many researchers have explored it further and applied to diversified domains, as
highlighted in the next section.

1.4 Applications of Visual Cryptography

Aside from the obvious application of information hiding, visual cryptography can be
applied to a number of domains such as access control (opening of bank vault), threshold
signatures (wallet security through multiple devices e.g. bitcoin), copyright protection,
watermarking, visual authentication, ticket validation, and human identification, etc. Visual
cryptography applications range from the banking industry, satellite imaging to commercial
application for preserving collected biometric data.

Visual cryptography technique is very intuitive to the user. However, it is surprising that
within the last two decades since its inception by Naor and Shamir only a few suggestions have
been made to apply it to practical problems. Naor and Pinkas [6] presented a technique to use
visual cryptography in order to protect online transactions against manipulation and Chaum et al.
[7] suggested to apply it for verification of the correctness of the outcome of an election.

In order to secure online money transfers, the user gets a numbered set of transparencies,
from the transaction server. The server sends a visual message containing the transaction data to
the user's screen, and it is encoded via visual cryptography. If the user puts the transparency with
a certain number on top of the encoded image screened person can see the message contained
within the image. When the server receives the right TAN it executes the transaction, otherwise
not. In this way, one can make banking transaction more secure in nature.

Specifically, VC is popular in the use of Moiré pattern and watermarking. Moirés

patterns are generated when a revealing layer is stacked onto the top of an image with
periodically repeated shapes. Moiré pattern has been attempted by researchers to embed into VC
shares. The original secret can be revealed by superimposing the shares when the shares are
separated, the embedded image could be viewed.

Another frequently used application for VC is watermarking. Watermarking plays an

important role in information hiding and embedding. Similar to traditional VC, the implementation
of VC in watermarking is based on basis matrix and the final recovered secret is perceived by using
the contrast between white and black colors.

Similarly, Luo et al. [8] also explore the usage of watermarks within visual cryptography.
A digital image copyright scheme based on visual cryptography has been presented by Hwang
[9]. Embedding VC based watermarking into products is an effective way of preventing
cheating, especially in the domains where already get benefits from the use of watermarking.

Because of difficulties such as adjustment, size, and costs of special equipment, these
suggestions did not lead to applications that are used for serious purposes. But in the future,
further developments of the ideas presented in different approaches, as well as new ideas, could
spread practical applications of visual cryptography. VC in conjunction with modern day image
hatching techniques would allow the extension of VC into the currency domain, such as within
the banking industry. The use of shares within the secure printing industry should also be
considered. Scanning a share into a computer system and then digitally superimposing its
corresponding share could also be considered.

1.5 Challenges in Visual Cryptography

There are several drawbacks and challenges in the field of visual cryptography along with
its merits[10]. A wide variety of applications of visual cryptography would require the printing
of the shares on transparencies[10]. The scanned shares have to be superimposed in order to
reconstruct the secret image.

Pixel expansion is an important challenging issue for Visual Cryptography based

Schemes (VCS) [1] [11] [12]. Most of the research in the literature is dedicated to reducing pixel
expansion i.e., to reduce the number of sub-pixels in the shares that represent a pixel in the
original secret image. The final size of the transparencies of the VCS is affected not only by the
number of the sub-pixels, but also by the size of the sub-pixels in the transparencies[5].
However, not only pixel expansion is a major challenging issue, there are many other
challenging research issues on which various researchers have worked since its inception. These
several issues and challenges related to VC schemes are listed as follows:
 The possibility of sharing multiple secrets in the same set of shares: Numerous
schemes are presented which range from sharing just two secrets to the general case of
sharing any number of secrets. Circular shares seem to be best in terms of the secret
recovery and contrast for multiple secret supporting schemes. The scheme presented for
sharing more than two secrets using standard rectangular shares has issues with contrast.
Using a color cover image also presents an effective approach to share multiple smaller
secrets. Overall, the majority of the multiple secret sharing schemes are effectively hiding
two or more secrets within a set of shares. The schemes that support the secrets into
circular shares prove to be the most interesting and effective in terms of sharing many
secrets with very high contrast. The more secrets a user wishes to hide, the larger is the
size of the resultant shares. The final share size when many large secrets are considered
can become unmanageable.
 Schemes based on the probabilistic reconstruction of secret images: VC remains an
important research for solving pixel expansion problem. More specifically, the schemes
which minimize pixel expansion and also increase the overall contrast will certainly be
more suited. There is always a tradeoff between contrast and pixel expansion. An
approach based on coding theory helps to provide an optimal tradeoff between the
contrast and the number of sub-pixels. The size-adjustable scheme which allows the user
to specify what size of shares to generate is very interesting work. This allows for a user
defined tradeoff between quality and portability of shares. This increases the potential for
VC once again, rather than being restricted to a specific scheme which only allows for a
certain type of quality. Optimal (2, n) schemes are examined in terms of contrast related
to the Hamming distance, as well as the sub-pixels tradeoff required for these optimal
schemes [13]. So a scheme which has no pixel expansion along with high contrast
recovery of secret seems to be an ideal solution for secret sharing using visual
cryptography aspects.
 Contrast enhancement techniques: The overall contrast affects the quality of the
recovered secret in any VC scheme. Optimal contrast secret sharing schemes in visual
cryptography have been extensively researched[14][15]. A possible option for improving
the efficiency of VC is to use the other logical operators[16]. This method will not allow
traditional stacking of the shares on transparencies, but it will improve the overall quality
 of share. It can be applied to color images as well [17]. The annoying presence of the loss
of contrast makes traditional visual cryptography schemes practical only when quality is
not an issue which is relatively rare[18].
 Schemes based on the different logical operation for combining the shared images:
OR-based visual cryptographic schemes suffer from the low quality of the reconstructed
image. This is an inherent drawback of the OR based VCS. It cannot be improved beyond
a certain level. An XOR based VCS used the polarization of light where the underlying
mathematical operation is the Boolean “XOR” operation. Thus, unlike OR-based
schemes where a participant has to carry a number of transcripts to update the shares, in
an XOR-based VCS a person has to carry just one dedicated trusted device that has a
display[19]. However, using XOR based operation losses the real nature of VCs
decryption without the participation of a computer. Some other mean of improving
contrast is required to get better contrast [20].
 Color image visual cryptography: Applying visual cryptography techniques allows the
use of natural color images to secure information. In most color visual cryptography
schemes, when the shares are superimposed and the secret is recovered, the color image
gets darker. This is due to the fact that when two pixels of the same color are
superimposed, the resultant pixel gets darker. Therefore, the basic scheme is extended to
allow visual cryptography to be directly applied on grayscale and color images. Image
halftoning is employed in order to transform the original image from the grayscale or
color space into the monochrome space which has proved to be quite effective[21].
Working with different types of color model and size of the pixel is also a challenging
issue in color VC[22].
 Cheating prevention schemes: If an attacker knows the exact distribution of black and
white pixels of each of the shares, the person may be able to successfully attack and cheat
the scheme[23]. A method which prevents the attacker from obtaining this distribution
may be useful to prevent cheating. The coating process could cause damage to genuine
participants because they will accept a forged image different from the actual secret
image as authentic. Many researchers have experimented with the idea of cheating the
system and suggested the solution for its prevention also[24]. Methods for cheating the
basic VC schemes have been presented, along with techniques used for cheating
 prevented VC schemes. Lots of new VC scheme requires an additional look to check
whether the scheme is cheating immune or not.
 Alignment position for image shares: The size of the circle shares increases with the
number of secrets hidden. The expansion is twice the number of secrets to be hidden.
However, the number of secrets that are contained within the shares still remains a secret
unless supplementary lines are added to the circle shares to ease the alignment. This is
another problem with sharing multiple secrets, especially when dealing with circle shares.
Exactly knowing the correct alignment point and how many secrets are actually contained
within the shares is also a concern. If the rotation angle is small and rotation of the shares
occurs too quickly, it is possible that all secrets may not be recovered.
 Random grid techniques: The use of random grids (RG) technique of secret sharing is
quite similar to the basic scheme of VC. Random grid scheme [25] takes an input image
and transforms it into multiple grids that provide no information on the original image.
Three different algorithms have been proposed to encrypt a binary image into two cipher
grids. This scheme has the additional benefit that it does not require pixel expansion and
codebook requirement. Many research has been carried out to show the relationship
between random grids and traditional visual cryptography methods. A lot of research is
required in the direction of mapping other variants of visual cryptography based scheme
with the random grid approach.

Despite the security nature of the scheme for secret sharing, VC shares lacks in security.
Many VC schemes have the problem of the artifact itself in the generated shares.
Previous related research has observed the possibility of VC cheating through different
methods. Attackers are able to complete both cheating and modification of the VC
process without being noticed by VC participants. The various factors affect the VCS
scheme before applying for any practical application, e.g. pixel expansion, contrast, no.
of shares generated, and type of shares. The major work of this thesis is focusing towards
effectively improving the security of shares created in visual cryptography based schemes
for secret sharing. Therefore the main research question of this thesis is focusing to
improve the effectiveness of VC secure share generation along with its application for
authentication in practical usage. By raising this research question and reviewing
 previous related works, experiments of utilizing bit level based processing as well as
chaotic map features has been proposed.

1.6 Background and Research Motivation

The initial awareness of visual cryptography was raised by Naor and Shamir in 1994 [1].
Unlike commonly used security methods which tend to hide information by applying a
mathematical transformation on secrets in the format of plain text, Visual Cryptography Scheme
(VCS) is defined as an activity that a secret is stored in an image (usually black and white). In
VC, a secret image is split into several random looking images called VC shares. Different from
traditional secret sharing where each of the participants has the knowledge of part of the secret,
every piece of VC shares has no indication of the original secret image while viewers are solely
able to clearly perceive the secret by simply overlaying these shares.

Among the different kinds of VC, there are five significant VC schemes which are
commonly investigated and discussed [26] that are as follows: Traditional VC, Extended VC,
Dynamic VC, Color VC and Progressive VC. Traditional visual cryptography is the basic VC
scheme which utilizes only black and white pixels to encrypt binary images while the VC shares
are the random pictures without obviously visible and semantic information.

The process of extended visual cryptography is similar to that of traditional visual

cryptography. VC shares in extended visual cryptography are usually a picture with meaningful
cover information but have no clues about the secret. Dynamic visual cryptography indicates
the VCS where VC shares can be applied to reveal more than one secret. Color visual
cryptography is used to encrypt a color secret image. Meanwhile, progressive visual
cryptography more suited to the situation where the secret is gradually revealed[27].

Despite the distinct encryption process in different types of VC, the basic idea of VC
encryption in all types has been based on pixel expansion. Pixel expansion indicates that one
pixel of the original secret image is represented by a number of sub-pixels in its corresponding
region on VC shares [26]. One pixel is represented by a randomly selected group of sub-pixels.
Thus, same pixels in the original secret image are possibly split into sub-pixels in the different
arrangement, which subsequently leads to the randomness of the appearance of the whole share.
Moreover, the color of a specific pixel on VC shares is determined by the predefined expansion
rules of VCS.

As for the decryption process, there are two kinds of VC revealing operations, namely,
XOR and OR. The revealing result of using XOR is better than that of OR which is, however
commonly used as the simplicity of its implementation. In OR operation, VCS result is perceived by
human visual system [1] [28]. On superimposed VC secret image, light regions are represented by
groups of white and black sub-pixels, while dark regions are filled up with only black sub-pixels.
The contrast of the light regions and dark regions is easily identified by HVS [29].

The aim of the secret sharing scheme is to improve the way secret information of highly
sensitive nature can be shared with the use of information and communication technologies and
cost effective decoding technique. Therefore, all over the world governments have been devising
their security plans to make this objective successful. Lots of interest has been generated in
accomplishing the objective of secure secret sharing due to economic and environmental issues.
Few motivating factor to work in the area of VCS has been listed hereunder.
 As a powerful technique for information security, VC indicates the possibility of visually
protecting crucial secrets from the view of secret sharing [30] [26]. Therefore, VC is a
preferable security scheme for the protection of confidential documents such as bank
vault password, personal account access password, and many others.
 The main advantages of VC can be seen in three aspects. The first one is that VC secret is
revealed by using only VC shares which are convenient to carry with, unlike other
cryptography methods which require complicated computations and powerful computers
for the decryption process. VC concentrates on using images as the media for
transmission of secret information. Compared to encrypt the plain text, images appear to
be more flexible in conveying secret content and have larger information storage.
 The foremost benefit of using visual cryptography is that it saves excess efforts in
cryptanalysis on the part of the authorized users wanting access as no mathematical
computations are involved and decryption can be easily performed using the basic human
visual system.
 With all these outstanding benefits in information hiding and great potential in the
development of both theory and application, VC has become one of the most promising
 security methods in secret sharing.
 As the whole process of VC is based on VC shares, the security and protection of genuine
VC shares would be required to make sure the successful revealing of the secret.
However, based on the analysis of encryption and decryption mechanism of VC, cheating
attempts are proven to be effective in VC authentication.
 The lack of a strong authentication result in the loss of confidential material. In 2015,
Statistics produced by the Moscow-Pullman Daily News cites that Sensitive information
was compromised for somewhere between 4 million and 14 million federal employees. A
Study conducted by Gemalto, identity theft accounts for 64 percent of all data breaches
globally. The joint study by Assocham and Ernst & Young states the mobile platform
faces the greatest risk with mobile fraud slated to grow by 60 -65 percent in the next
decade in India alone.
 With such a wide range of security applications, it is going to be quite interesting and
helpful to have VCS in security domain as prime technique. There is a need for efficient
and low-cost practical application for authentication using VCS.

Due to the security weaknesses of existing VC shares, relevant solutions of VC

authentication have been raised such as transforming one VCS to another which matches
cheating prevention requirements. Many VC schemes have the problem of the artifact itself in
the generated shares. Chen et al.[31] also proposed a cheating prevention method that deals with
the cheating immune problem. Most previous work for VC authentication is introduced in
Chapter 3. Moreover, the benefits of visual cryptography can be further accentuated if the
problems of contrast reduction and pixel expansion were to be solved. Owing to the fact that
nearly none of the techniques have been benchmarked yet, it provides a wide scope for
improvement and research in this field.

1.7 Problem Statement

Many experimental results reported in the literature showed that most of the scheme
suffers from the problem of pixel expansion and low contrast output. There is a continuous need
to study various secret sharing schemes to select an appropriate method for the different domain
of security.
 The intent of this research is to study the available visual cryptography based schemes
and performance analysis of these schemes on the basis of different parameters (pixel expansion,
the number of secret images, image format and types of shares generated), with a view to
develop a new, more efficient technique to overcome the problem encountered in the existing
schemes. In the present work, we proposed to design efficient visual cryptography based scheme
for providing security. Here, we mainly concentrated on progressive sharing schemes of VC.
Progressive visual cryptography schemes are more promising than other traditional sharing
schemes due to its advantage in terms of minimal or no pixel expansion.

1.8 Objectives
The following objectives have been set to achieve the proposed goals described above in
the problem definition:
 Detailed study of various existing visual cryptography schemes for all types of images.
 Development of a new algorithm on visual cryptography based approach to solve the
existing drawbacks of different techniques along with experimental verifications.
 Development of secure and more general scheme for sharing the secret message.
 Analyze the visual similarity between the original image and the encrypted share images
so as to check the vulnerability of shares.

The objective of the work is to find an optimal method for ensuring the security of secret
message in all the generated shares, along with high contrast reconstructed secret message. The
contributions of this thesis are a set of studies, modifications to the existing algorithm and
analysis of the designed proposed algorithms.
Along with authentication, issue of generation of secure share is equally important. This
research is based on using applicable methods for VC shares security and set up the analysis of
designing a VC scheme from the view of features.

1.9 Scope of this Research Work

In today’s scenario, everything is digitalized. The rate at which data are increasing, on the
contrary, its misuse is also increasing. For protecting the digital information or content,
researchers have proposed various security techniques. Security techniques can be made more
powerful if we store the sensitive data in a distributed way. Traditional security techniques take
more computational time in comparison to visual cryptography techniques.

The research work carried out here shows the findings of the literature related to visual
cryptography, its applications, and how visual cryptography can be applied in traditional form to
ensure the security of data. Different approaches have been validated and compared on the basis
of the number and type of secret images, pixel expansion after encoding, different type of
generated shares (meaningful, random, hatched, etc.).[32].

To have more insights in this area, traditional visual cryptography and random grid
technique (identified from above work) have been applied for detecting the fraud candidate.
Further, a novel authentication technique using bit plane based encryption has also been
suggested [33]. All these three mentioned techniques have been compared for their suitability.

Specifically, applicable methods for VC secure share generation in this thesis include
chaotic theory. Past research has investigated the suitability of using chaos into VC shares for
secure share scheme. To be specific, there are several benefits of using chaos theory in
authentication [34] [35]. Firstly, different from using a pseudo-random number for encryption
matrices, multiple 1-D maps can be used into shares, thereby simplifying the process of uniform
random number generation. In addition, cheaters can hardly get the information of the secret
from the prediction of the secret in shares. Further, using map has the advantage of encoding a
large scale of information into a controllable set of shares.

A chaotic system is very sensitive to changes in input parameters. Using the VC shares
generation with chaos facilitates the process to stop the artifact of the content in the shares. The
chaotic map related part of this thesis mainly concentrates on improving the share generation
algorithm process to be more practical for VC secure share. The research of a 1-D chaotic map
for the VC scheme in this thesis is due to the similarity between the need of VC secure shares
and that of a chaotic map, which is explained in chapter 4 and 5.

Extracting features of images are always a significant analysis of image processing,

especially in the use of image matching. For the motivation of making use of VC shares’ image
as more secure, it is critical to statistically analyze these shares. The investigation of VC from the
visual aspects of the authentication process and secure VC share generation has not been
comprehensively done till now. This research will focus on the usage of these types of
features[36]. Specifically, visual features include Pixel histograms, Entropy, Mean Square Error
(MSE), peak signal-to-noise ratio (PSNR), Structural Similarity Index Measure (SSIM) etc.
Different images have different appearances with regard to these features and even a slight
change in an image will lead to the obvious difference in its features. Cryptographic features can
be employed to encrypt the visual features and use of these features as a vector for
authentication. The SSIM has been explored further to match with the reconstructed secret.

After analyzing the various traditional techniques and their applicability, now, our
research mainly focused on progressive visual secret sharing (PVSS) scheme which has added
advantage over the traditional approach of secret sharing schemes[37]. Here, the major efforts
have been done in the direction of secure shares generation with PVSS. The idea of progressive
visual secret sharing has been extended further by introducing logistic chaotic maps in the
algorithm for generating secure shares. After modification, the algorithm is named as Chaos
based PVSS (CB-PVSS) [38]. An improvement has also been suggested in existing block based
schemes and the modified block based scheme is named as Chaos based block progressive visual
secret sharing (CB-BPVSS) [39]. General block based scheme nature is (2, n). Further, Block
based scheme has been extended to support threshold based (k,n) block recovery using the
progressive paradigm [40]. Details of all this work have been explained in further chapters
mainly in chapter 3, 4 and 5.

1.10 Organization of the Thesis

This thesis aims at reporting the investigation of secret sharing schemes using visual
cryptographic aspects and its different approaches centering upon the idea of VCS. It starts off
with chapter 1 which is mainly focused towards the introductory concepts related to visual
cryptography, including its different challenging issues, motivation, objective and scope of the
research.

Chapter 2 presents the literature survey done from the beginning for VCS. It discusses the
foundation of visual cryptography and developments in this area of research till date. This
chapter briefly outlines the different schemes proposed by the researchers in this area of study.
Here, a study of implementing VC using random grids and both OR & XOR operations with
respect to both binary and grayscale images along with different performance measures in the
VCs schemes has also been presented. In this work, the main focus is on the deterministic and
probabilistic models of visual cryptography. Performance analysis of all the existing visual
cryptographic based schemes and their features has been analyzed with a view of better insight
about the scheme and its limitation.

Chapter 3 deals with the suggestion of a novel authentication scheme called Bit-level based
visual cryptography and its implementation for the verification of a person’s identity. It also
provides a comparison of the proposed scheme with the existing techniques for authentication as
well as with other visual cryptography techniques. A design of authentication system has been
proposed for both registrations and authentication phases. At the end of the chapter,
implementation of the scheme has been discussed all the results are shown as comparison with
the existing schemes, under different types of attacks and of different file formats.

Chapter 4 consists of the cryptanalysis of progressive visual cryptography and its improvement
using chaos based random number generation. Results have been shown to exemplify the
working and observations have been made during different stages of testing the algorithms. It
also describes the role of random number generator in progressive secret sharing schemes. It
states the results at the end in a tabular and graphical form highlighting the evaluation
parameters, which can then be used as performance metrics for the same.

Chapter 5 discusses the idea of block based progressive visual cryptography and its
improvement using a chaotic map to solve the secret artifact disclosure in corresponding blocks
of the scheme. Further, a novel threshold based block recovery of the image has been suggested
to extend block based progressive scheme from (2, n) to (k, n). Experimental results, showcasing
the performance and security of designing schemes have been highlighted. The last chapter 6,
i.e. the conclusion and future direction, analyze the results, previously stated in the thesis and
hence, our observations along with scope for future work in this area. Lastly, all the references
used throughout report have been listed.