Data Privacy Protection in Cloud Computing Using Visual Cryptography

Visual cryptography is a cryptographic methodology designed to secure visual information in a way that it can be effortlessly deciphered through human vision. As an emerging technology, visual cryptography capitalizes on the innate capabilities of the human visual system to encode and decode images securely. This novel encryption paradigm ensures the confidentiality of digital transmissions by permitting decryption solely through visual means, a key feature that sets it apart. In this work, an innovative visual cryptography encryption system is introduced, adept at processing confidential information while safeguarding the integrity of the source data. The process begins with the fusion of a cover image and a hidden image using advanced data embedding techniques. Subsequently, the embedded image is divided into 3 × 3 patches, forming the basis for further encryption. To enhance security, a secret key is applied to encrypt the picture. Furthermore, this system employs index1 and index2, which are trained utilizing Long Short-Term Memory (LSTM), a neural network model adept at capturing sequential patterns in data. The important visual features of LSTM, such as its ability to retain long-term dependencies and discern subtle sequential variations, are instrumental in generating an encrypted image from the data block, further fortifying the security of the information. To finalize the encryption process, the system employs ZigZag scanning and the Discrete Cosine Transform (DCT) to encrypt the secret image. These cryptographic techniques collectively bolster the confidentiality and integrity of the data, ensuring its safe transmission and protection against unauthorized access. This work attains impressive results, including a Mean Squared Error (MSE) of 1.3, a Peak Signal-to-Noise Ratio (PSNR) of 55.9, a Cross-Correlation (CC) of 0.983, a Structural Similarity Index (SSIM) of 0.98, a Root Mean Square Error (RMSE) of 0.28, and a Mean Absolute Error (MAE) of 1.36.