An Adaptive Cloud-Based Image Steganography System with Fast Stego Retrieval

Image steganography is one of the common techniques used in information-hiding applications. Existing image steganography solutions generally focus on improving both the computation of data embedding and the completeness of data extraction. However, the steganography service providing adaptive matching of multiple ciphertexts and suitable image files with the support of fast stego retrieval is mostly overlooked by existing works. This paper introduces a cloud-based steganography system providing imperceptibility preservation and adaptive matching between text, file, or ciphertext within randomly selected cover images. To this end, we propose an algorithm to compute the optimal cover image for different ciphertext sizes based on the optimized calculation of the least significant bit (LSB) and file ratio. We also propose a graph-based technique to structurally model the correlation between users, ciphertext, and image files. This enables fast stego image retrieval and scalability for multi-users in a cloud setting. Finally, we conducted experiments to measure the performance of data embedding, data extraction, and stego retrieval to substantiate that our proposed system is feasible to implement.