Publications

Detailed Information

Extension of Visual Cryptography for Color Images : 컬러 이미지를 위한 확장 시각 암호학

DC Field Value Language
dc.contributor.advisorSrinivasa Rao Satti-
dc.contributor.authorKumenitHagosAlema-
dc.date.accessioned2018-12-03T01:42:55Z-
dc.date.available2018-12-03T01:42:55Z-
dc.date.issued2018-08-
dc.identifier.other000000152501-
dc.identifier.urihttps://hdl.handle.net/10371/143844-
dc.description학위논문 (석사)-- 서울대학교 대학원 : 공과대학 컴퓨터공학부, 2018. 8. Srinivasa Rao Satti.-
dc.description.abstractComputer and communication technology has led to rapid development of digital media. Huge amount of digital data, such as text, voice, image, etc…, are transmitted and shared over the Internet. Multimedia data can be easily spread, duplicated, captured, and modified. However, the easiness of data exchange over the Internet has also brought problems such as data embezzlement, unlawful data usage and other problems regarding intellectual property rights. Protecting important data from attacks becomes key issue. Secure information sharing technique to prevent transmitted data from being accessed by unauthorized body is required.



Visual cryptography is a cryptographic technique in which a secret image can be revealed when the minimum required number of shares are stacked together. The secret image is encoded into n transparent shares. Each share constitutes some visual information of the secret image. To decode the secret image, $k$ out of $n$ transparencies should be superimposed one on top of the other. An individual share or any combination of less than these $k$ shares cannot reveal the secret image. The advantage of visual cryptography is that decryption of the visual information is done by using human visual system, i.e., it does not need any mathematical computation to decrypt the visual information.



The security of visual cryptography depends, of the many factors, on how hard it is to guess that an individual share is one of the many shares used to decode a secret visual information. A lot of articles have been published on visual cryptography. Most of them deal with grayscale images and the shares are not meaningful which help attackers to easily guess that an individual share is carrying a secret visual information. The focus of this thesis is to create a meaningful shares using cover images. Halftone technique is utilized and extended to be used with color images, which was originally proposed for grayscale images. Two colorful images are selected and preprocessed to make them ready for hiding a secret image. A colorful secret image is preprocessed in turn and encoded in these two meaningful shares. To decrypt the hidden secret image, the generated two meaningful shares are aligned one on top of the other properly, with out the need to do any cryptographic computation. The proposed scheme has two main features. First, each share is a meaningful image, as opposed to random looking share, and this makes it hard for attackers to guess that a given share is carrying an encoded secret image. Second, even if an attacker gets access to an individual share and guesses that the given share is an encoded share, he will not be able to recover the secret image unless he gets access to all the shares in which the secret image is encoded.



The main goal of this thesis is to design a visual cryptography scheme for color images. Individual shares of proposed scheme are meaningful, as opposed to random noise-like shares. Simple alignment of transparencies of these two meaningful shares is all we need to do to recover the secret image. The scheme uses 50\% of the total pixel of both the secret and cover images and is able to recover the secret image with 50\% contrast loss and with a small pixel expansion. A scheme that uses simple computation and increases the intensity of the recovered image from 50\% to 100\% is also proposed, which we believe that it is worth trading off.



An access structure ($\Gamma_{Qual}, \Gamma_{Forb}$) on a set of n participants is another cryptographic technique to encode a secret image into n share transparencies. Each participant receives a single transparency. Participants in the qualified subset can decode the secret image while participants in the forbidden subset has no information about the concealed secret image. This scheme is an enclosure of the $(k,n)$-threshold visual cryptography technique, where any $k$ or more participants can recover the secret image while less than $k$ participants cannot, even with the help of high mathematical computation. As an extension of the original proposed scheme, an access structure based cryptographic technique for color images is presented at the end of this thesis. Experimental results are presented and show the applicability of the proposed schemes.
-
dc.description.tableofcontentsChapter 1 Introduction 1

Chapter 2 Previous Works 10

2.1 Color Models and Color Superposition . . . . . . . . . . . . . . . 11

2.2 Naor and Shamir Basic Visual Cryptography . . . . . . . . . . . 13

2.3 Extended Visual Cryptography . . . . . . . . . . . . . . . . . . . 17

2.4 Visual Cryptography for Continuous-tone Digital Images . . . . . 20

2.4.1 Digital Halftoning . . . . . . . . . . . . . . . . . . . . . . 22

2.4.2 Visual Cryptography for Grayscale Images . . . . . . . . . 29

2.4.3 Color Visual Cryptography . . . . . . . . . . . . . . . . . 34

Chapter 3 Proposed Scheme 37

3.1 Preparing Input Images for Processing . . . . . . . . . . . . . . . 39

3.2 Color Image Halftoning via Error Diffusion . . . . . . . . . . . . 39

3.3 Secret Image Encoding . . . . . . . . . . . . . . . . . . . . . . . . 44

3.4 Cover Image Encoding . . . . . . . . . . . . . . . . . . . . . . . . 50

3.5 Meaningful Share Generation . . . . . . . . . . . . . . . . . . . . 52

3.6 Color Image Decoding . . . . . . . . . . . . . . . . . . . . . . . . 54

3.7 Perfect Pixel Reconstruction of Every Pixel . . . . . . . . . . . . 57

3.8 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 60

Chapter 4 General Access Structure for Color Images 63

4.1 Construction of Basis Matrices for Access Structures . . . . . . . 66

4.2 Share Generation and Decoding . . . . . . . . . . . . . . . . . . . 71

4.3 Color Cryptography with No Pixel Expansion . . . . . . . . . . . 75

4.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 77

Chapter 5 Conclusions and Future Work 82

Bibliography 85

요약 89
-
dc.formatapplication/pdf-
dc.format.mediumapplication/pdf-
dc.language.isoen-
dc.publisher서울대학교 대학원-
dc.subject.ddc621.39-
dc.titleExtension of Visual Cryptography for Color Images-
dc.title.alternative컬러 이미지를 위한 확장 시각 암호학-
dc.typeThesis-
dc.contributor.AlternativeAuthor하고스-
dc.description.degreeMaster-
dc.contributor.affiliation공과대학 컴퓨터공학부-
dc.date.awarded2018-08-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share