A Secure Image Watermarking Using Visual Cryptography and Discrete Fractional Fourier Transform

2014 ◽  
Vol 577 ◽  
pp. 754-757
Author(s):  
Hong Yang Wang

In this paper we proposed a secure image watermarking algorithm for digital image using the visual cryptography (VC) and discrete fractional Fourier transform (DFRFT). We use the visual secret sharing scheme to construct two shares, namely, master share and ownership share. Features of the original image are extracted using Non-Negative Matrix Factorization (NMF), and are used to generate the master share. Ownership share is generated with the help of secret image (watermark) and the master share, using VC technique. In case of any dispute, the master shares and ownership shares can be stacked together to give the copyright image verifying the ownership about the host image. In order to achieve the robustness and security, the properties of VC, DFRFT and NMF are used in our scheme. The experimental results show that the proposed scheme is strong enough to resist various signal processing operations.

Today organizations face a challenge while recruiting candidates, who provide forged mark sheets in order to get a job. To prevent wrong hiring a detailed and thorough approach is needed to verify the authentication of both the candidate and the marks obtained by him/her. There are so many modern cryptographic protocols available which can be used for authenticating individual’s academic achievement certificates. Visual Cryptography is a simple and secure way to allow the secret sharing of images without any cryptographic computations or the use of encryption or decryption keys. The novelty of the visual secret sharing scheme is in its decryption process where human visual system (HVS) is employed for decryption of secret shares. In this paper we have discussed (3, 3) visual cryptography scheme which can be used to generate shares and distributes them among three parties, i.e. the Job Seeker, Certificate Issuance Authority and the Organization conducting Job interview. Secret message can be decrypted only if all the three shares are available. Every certificate carries a unique number which is encrypted using visual cryptography and without handshaking of all the parties it is impossible to decrypt, thus ensuring full proof authentication.


2010 ◽  
Vol 2010 ◽  
pp. 1-18 ◽  
Author(s):  
Mustafa Ulutas

This paper presents a new scheme for hiding two halftone secret images into two meaningful shares created from halftone cover images. Meaningful shares are more desirable than noise-like (meaningless) shares in Visual Secret Sharing because they look natural and do not attract eavesdroppers' attention. Previous works in the field focus on either increasing number of secrets or creating meaningful shares for one secret image. The method outlined in this paper both increases the number of secrets and creates meaningful shares at the same time. While the contrast ratio of shares is equal to that of Extended Visual Cryptography, two secrets are encoded into two shares as opposed to one secret in the Extended Visual Cryptography. Any two natural-looking images can be used as cover unlike the Halftone Visual Cryptography method where one cover should be the negative of the other cover image and can only encode one secret. Effectiveness of the proposed method is verified by an experiment.


2019 ◽  
Vol 6 (1) ◽  
pp. 18-42
Author(s):  
Ram Chandra Barik ◽  
Suvamoy Changder ◽  
Sitanshu Sekhar Sahu

Mapping of image-based object textures to ASCII characters can be a new modification towards visual cryptography. Naor and Shamir proposed a new dimension of Information security as visual cryptography which is a secret sharing scheme among N number of participants with pixel expansion. Later on, many researchers extended the visual secret sharing scheme with no expansion of pixel regions in binary and color images. By stacking k shares the secret can be decoded using normal vision. In this paper the authors have proposed a modification towards visual cryptography by converting the message in the form of printable ASCII character-based numerical encoding patterns in a binary host image. The encoding of the message is represented as ASCII numeric and a texture of those numeric are arranged to form a binary host image. Then, N numbers of shares are built up but after stacking all the shares the decoding of the message is achieved by converting ASCII numeric to the secret.


Sign in / Sign up

Export Citation Format

Share Document