Young-Chang Hou

Visual cryptography for color images

Visual cryptography, an emerging cryptography technology, uses the characteristics of human vision to decrypt encrypted images. It needs neither cryptography knowledge nor complex computation. For security concerns, it also ensures that hackers cannot perceive any clues about a secret image from individual cover images. Since Naor and Shamir proposed the basic model of visual cryptography, researchers have published many related studies. Most of these studies, however, concentrate on binary images; few of them proposed methods for processing gray-level and color images. This paper proposes three methods for visual cryptography of gray-level and color images based on past studies in black-and-white visual cryptography, the halftone technology, and the color decomposition method. Our methods not only retain the advantages of black-and-white visual cryptography, which exploits the human visual system to decrypt secret images without computation, but also have the backward compatibility with the previous results in black-and-white visual cryptography, such as the t out of n threshold scheme, and can be applied to gray-level and color images easily.

Progressive Visual Cryptography With Unexpanded Shares

The basic (k, n)-threshold visual cryptography (VC) scheme is to share a secret image with n participants. The secret image can be recovered while stacking k or more shares obtained; but we will get nothing if there are less than k pieces of shares being overlapped. On the contrary, progressive VC can be utilized to recover the secret image gradually by superimposing more and more shares. If we only have a few pieces of shares, we could get an outline of the secret image; by increasing the number of the shares being stacked, the details of the hidden information can be revealed progressively. Previous research, such as Jin in 2005, and Fang and Lin in 2006, were all based upon pixel-expansion, which not only causes the waste of storage space and transmission time but also gets a poor visual quality on the stacked image. Furthermore, Fang and Lins research had a severe security problem that will disclose the secret information on each share. In this letter, we proposed a brand new sharing scheme of progressive VC to produce pixel-unexpanded shares. In our research, the possibility for either black or white pixels of the secret image to appear as black pixels on the shares is the same, which approximates to 1/n. Therefore, no one can obtain any hidden information from a single share, hence ensures the security. When superimposing k (sheets of share), the possibility for the white pixels being stacked into black pixels remains 1/n, while the possibility rises to k/n for the black pixels, which sharpens the contrast of the stacked image and the hidden information, therefore, become more and more obvious. After superimposing all of the shares, the contrast rises to (n-1)/n which is apparently better than the traditional ways that can only obtain 50% of contrast, consequently, a clearer recovered image can be achieved.

Copyright protection scheme for digital images using visual cryptography and sampling methods

A novel copyright protection scheme for digital images based on visual cryptography and statistics is proposed. The proposed method employs sampling distribution of means and visual cryptography to achieve the requirements of robustness and security. Our method can register multiple secret images without altering the host image and can identify the rightful ownership without resorting to the original image. Moreover, the proposed method enables the secret images to be of any size regardless of the size of the host image. Finally, experimental re- sults show that the proposed scheme can resist several common attacks.

Block-based progressive visual secret sharing

Block-based progressive visual secret sharing (BPVSS) is an encryption technique that utilizes the human visual system to recover a secret image. The recovery method is based on building up different image blocks step by step. In the last few years, several research projects have been devoted to the study of BPVSS. However, they have some common drawbacks. Their dispatching matrices are chaotically designed by adding some redundant 0/1s in order to maintain the size as a fixed value, which makes the methods hard to be implemented. In addition, the pixel expansion ratio grows faster when the number of participants increases and the visual quality of the restored image is quite poor. In this paper, we propose a non-expanded block-based progressive visual secret sharing scheme with noise-like and meaningful shares, respectively. Experimental results demonstrate the feasibility of this scheme. Our design has several advantages over other related methods: (1) the model is concise, easy to implement, and the number of participants need not be restricted; (2) with noise-like shares, the contrast of the restored image is 50%, which means that the hidden message can be clearly recognized by the naked eye; (3) when transparencies are transferred from noise-like patterns into a meaningful cover image, the contrast in the shares and the restored image is 25%, which is still superior to that obtained in other studies; (4) our scheme is more suitable for grayscale and color secret images; and (5) the size of the transparencies is the same as that of the secret image.

An asymmetric watermarking scheme based on visual cryptography

Digital watermarking is a very active research area for copyright protection of electronic documents and media. A visual cryptographic approach is used to generate two random shares of a watermark: one is embedded into the cover-image and another is kept as a secret key for the later watermark extraction. The watermark can be extracted by simply superimposing the key share over the stego-image. This asymmetric digital watermark is specially designed and is not easily changed or removed. But, it is very convenient to be extracted. The embedded digital watermark by this approach seems robust after several attacks are performed on the stego-image.

An optimization model for visual cryptography schemes with unexpanded shares

Visual cryptography schemes encrypt a secret image into n shares so that any qualified set of shares enables one to visually decrypt the hidden secret; whereas any forbidden set of shares cannot leak out any secret information. In the study of visual cryptography, pixel expansion and contrast are two important issues. Since pixel-expansion based methods encode a pixel to many pixels on each share, the size of the share is larger than that of the secret image. Therefore, they result in distortion of shares and consume more storage space. In this paper, we propose a method to reach better contrast without pixel expansion. The concept of probability is used to construct an optimization model for general access structures, and the solution space is searched by genetic algorithms. Experimental result shows that the proposed method can reach better contrast and blackness of black pixels in comparison with Ateniese et al.s.

Random-Grid-Based Visual Cryptography Schemes

This paper discusses a random-grid-based nonexpanded visual cryptography scheme for generating both meaningful and noise-like shares. First, the distribution of black pixels on the share images and the stack image is analyzed. A probability allocation method is then proposed that is capable of producing the best contrast in both the share images and the stack image. With our method, not only can different cover images be used to hide the secret image, but the contrast can be adjusted as needed. The most important result is the improvement of the visual quality of both the share images and the stack image to their theoretical maximum. Our meaningful visual secret sharing method is shown in experiments to be superior to past methods.

A technique for sharing a digital image

A new information hiding scheme for color images based on the concept of visual cryptography and the Boolean exclusive-or (XOR) operation is proposed. Three different schemes with noise-like, meaningful and binary shares are presented. Meaningful shares may reduce suspicion that something is concealed there. Binary shares can achieve both the benefits of smaller share size and good visual quality. Our model can be easily extended from 256 colors to 65,536 or true color images simply by expanding the block size from 3×3 to 4×4 or 5×5, respectively. A true color secret image can be concealed into two meaningful cover images.5 algorithms are proposed to share a secret color image.Our schemes can achieve both the benefits of smaller share size and good visual quality.

Dynamic programming decision path encoding of genetic algorithms for production allocation problems

Genetic algorithm is a novel optimization technique for solving constrained optimization problems. The penalty function methods are the popular approaches because of their simplicity and ease of implementation. Penalty encoding method needs more generations to get good solutions because it causes invalid chromosomes during evolution. In order to advance the performance of Genetic Algorithms for solving production allocation problems, this paper proposes a new encoding method, which applies the upper/lower bound concept of dynamic programming decision path on the chromosome encoding of genetic algorithm, that encodes constraints into chromosome to ensure that chromosomes are all valid during the process of evolution. Utilization of the implicated parallel processing characteristic of genetic algorithms to improve dynamic programming cannot guarantee to solve complex problems in the polynomial time. Additionally, a new simultaneous crossover and mutation operation is proposed to enable the new method to run correctly following the standard genetic algorithm procedures. This approach is evaluated on some test problems. Solutions obtained by this approach indicate that our new encoding genetic algorithms certainly accelerate the performance of the evolution process.

Image protection based on visual cryptography and statistical property

In this paper, a novel intellectual property protection scheme for digital images based on visual cryptography and statistical property is proposed. The result of comparing two pixels that are selected randomly from the host image determines the content of the master share. Then, the master share and the watermark are used to generate the ownership share according to the encryption rules of visual cryptography. Our method does not need to alter the original image and can identify the ownership without restoring to the original image. Besides, our method allows multiple watermarks to be registered for a single host image without causing any damage to other hidden watermarks. Moreover, it is also possible for our scheme to cast a larger watermark into a smaller host image. Finally, experimental results will show the robustness of our scheme against several common attacks.