A Novel Method for Securing Medical Image Using Visual Secret Sharing Scheme

- The visual mystery sharing (VSS) plot, which starts from the visual cryptography, is an impeccable secure technique that ensures a picture by separating it into many offer pictures. It can be effortlessly remade by the human visual framework without the learning of cryptographic calculations. Security on digital medical images, faced with several security issues in today’s healthcare institution. To resolve this issue, a VSS based medical image security scheme is proposed in this paper. Here a new Color Visual Secret Sharing (CVSS)plot is connected to produce shares for input medicinal picture. The offers are made for the safe medicinal picture transmission and the picture data secrecy is kept up. At first the medicinal picture is part into various shadow pictures and the picture is uncovered when every one of these shadows are stacked together.Performance of the proposed method is evaluated by sample test medical images. The experimental results show that the CVSS scheme can provide satisfactory security level.


Introduction
In the area of medical images, concerning security is more important because of this medical image transmitting to hospitals over unsecured communication channel over open network and unauthorized person can steal and use it of his own benefits so that privileges of patient is suffered. Based on this issue, many researchers have been proposed method using Visual Cryptography method in which original image is split into different shadow images. Naor and Shamir's proposed the visual cryptography [1] which is allowed to secret sharing images without cryptographic computation and this scheme is refer as K-out-of-n VCS. Original image is given and get encrypted form of n images as follows: T  S h1  S h2  S h3 ......S hn, where⊕ is a boolean operation, S hn ∈ 1, 2, ...., n is an image which appears aswhite noise,k ≤ n, and n isthe number of noisy images. It is difficult to decipher thesecret image T using individual S hn 's [2]. The classical (2,2) VC scheme for binary image share construction is illustrated in Fig.1. In this encryption process every pixel of secret image is transformed into two pixels, and each pixel take part in corresponding share image. In the decryption process the two shares are stacked together (OR operation) to recover the secret image. A novel brand of secret offering plan is given the pet name visual secret sharing (VSS) plan [1].
VSS is an inventive procedure in which the imparted images. In this example, the image has been split into two component images. Each component image has a pair of pixels for every pixel in the original image. In visual cryptography, a secret binary image is modified into n shares of arbitrary binary prototypes. It is conceivable to translate the secret image outwardly by superimposing a qualified subset of transparencies [3].

Related Work
Researchers extended k out of n secret sharing to apply on color images. They proposed an algorithm to divide a digital color image into n number of shares where minimum k number of shares is sufficient to reconstruct the image. If k number of shares is taken then the remaining shares are (n-k). In an image if certain position of a pixel is 1, then in (n-k)+1 number of shares in that position of that pixel there will be 1. In the remaining shares in that position of the pixel there will be 0. A random number generator is used to identify those (n-k)+1 number of shares [4].
Blundo proposed Visual Cryptography schemes for gray level images [5]. Savita Patil used the concept of visual information pixel synchronization and error diffusion to attain a color visual cryptography encryption method that produces color shares [6].
Researchers enhanced the friendliness of VSS scheme by adding a simple and meaningful cover images to noise like share but the problem with this enhancement is that the recovered images are have reduced display quality Several papers investigated meaningful halftone shares [7]- [9] and emphasized the quality of the shares more than the quality of the recovered images. These studies had serious side effects in terms of pixel expansion and poor display quality for the recovered images, although the display quality of the shares was enhanced. Hence, researchers make a tradeoff between the quality of the shares and the quality of the recovered images and the pixel expansion of the image.
Hao-Kuan Tso et al. [10] have been developed Friendly VSS to secure the medical images. In this scheme, random-grid algorithm is used to construct two meaningless shadows and then a friendly pattern is apply on the shares which could make the users manage and identify them easily. Nelmiawati et al. [11] have beenproposed Pixel-Based Dispersal Scheme (PBDS) which is enhanced by combined Rabin's IDA and Shamir's SSA to secure the DICOM digital medical image. BothShamir's SSA and Rabin's IDA have been implemented and tested on multi-provider clouds in securing medical records [12]. In [13] Secret Sharing Scheme, a random multi-bit grid R with size of m by n is generated by using a seed and pseudo random number generator. Then, the random multi-bit grid R and medical images are used to construct two shared images.

Proposed Method
The color visual secret sharing scheme (CVSS) is applied on the medical image which is constructed in to different shares and able to reveal only when all of these shares are aligned together. In this scheme, medical image is read from the dataset and extracting original pixels. Meanwhile, key matrix is randomly generated which holds the values 1 through N number of shares given by user. After that key matrix index has been noted and compared it into the same index of original image, place it into corresponding share images and remaining values of shadow images are filled with 255. Now shares are constructed and impossible to reveal the original medical image with one share.
For example, original image pixels are obtained and user want to create four share images so key matrix are generated with the values from the range 1 to 4 randomly. Suppose if user want to create five share images means key matrix holds the values in the range from 1 to 5. Take the pixel "100" from the index 3 of the original image and comparing it to same index of key matrix and from that key value "4" has been noted. It tells that this value 100 move into index 3 of the fourth share and remaining values are filled 255. The above process is repeated till the pixel values are filled into the share images. Once all these shares are arranged together, there is a chance to reveal the image else no one couldn't retrieve it with single shadow image. Following algorithm is applied to construct a share from the medical image.

Color Visual Secret Sharing Scheme ( CVSS) for Medical Images
Step 1: Read original medical image (grey-scale format). Step 2: Generate key matrix randomly of same size as original image. Key matrixvalue ranges depends on the number of shadows given by the users.
Step  Step 4: T shad Step 5: T their

NPCR and UACI Analysis
NPCR is the change rate of the number of pixels in the encrypted image. The unified average changing intensity (UACI) is the measurement of the average intensity of differences between the secret image and encrypted image. It shows the high NPCR values which represents the encrypted image pixels indexes are completely changed compared with original image. Experimental results shows the estimated expectations and variance of NPCR and UACI values are very close to the theoretical values, which justify the validity of theoretical values. Hence the proposed encryption scheme is resistant against differential attacks [15]. Fully uniform image with 256 gray levels which the probability of all pixels are the same, the entropy would have its maximum value, i.e. 8, which means the most irregularities among image pixels. The proximity of the image entropy to 8 means the efficiency of proposed method in image encryption [14].

Conclusion
In this paper a novel medical image security method is proposed by using color visual secret sharing scheme. In this algorithm the pixel position is shuffled based on random matrix with traditional VSS procedure.It is very useful to enhance the medical image security by encrypting original images. The CVSS method is not only applicable for medical images and also exists for other digital images. The experimental results show that the proposed CVSS scheme gives the image confidentiality, integrity and reliability. Performance analysis proves the security, effectiveness and robustness of the proposed VSS algorithm.