Secure grayscale image communication using significant visual cryptography scheme in real time applications

An increase in information sharing and growth of Cloud Computing, Internet of Things (IoT), Machine Learning and mobile applications have brought human life in the digital world. Many times, financial transactions, medical communications that involve the transfer of images impose threats to our Privacy, Personal Identity and Personal Health Information (PHI). Such image data are being stolen while in transfer and made the processing of digital images vulnerable in World Wide Web (WWW). Protecting the image data from intruders becomes an essential demand. Various techniques such as Encryption, Steganography and Watermarking are in place to protect image data. Such techniques imply complex computations for encryption and decryption and the quality of image decrypted is reduced. Visual Cryptography (VC) is a new encryption method uses combinational techniques to encode the image into ‘n’ shares and decodes by stacking the shares without complex traditional cryptographic algorithms. The Signific Visual Cryptography (SVC) aims to transfer the real time images securely without compromising the quality. SVC is implemented for (k, n) scheme and (n, n) scheme of VC and supports several real time images such as natural images, medical images, and Quick Response (QR) images. To develop SVC scheme, a new Error Abatement Technique (EAT) is proceeded with two filtering strategies namely, Value Discretization Filtering (VDF) and Reduced Error Filtering (REF) to provide significant meaning to pixel values. The performance analysis proves that (k, n) SVC improves the quality of the reconstructed secret image with increased Peak Signal-to-Noise Ratio (PSNR) up to 21% and Pixel error is calculated as Mean Square Error (MSE) and it is reduced up to 77%. (n, n) SVC maintains integrity of the secret image with zero MSE without complex computations.