Hybrid Steganography: A Multi-Layered Framework for Hybrid Text and Image Concealment Enhanced by AES and LSB Techniques

Recent advancements in steganography primarily focus on various network layers and diverse data-hiding techniques. However, these endeavors often result in image manip-ulation, distortion, and limited payload capacity. This paper introduces a novel dimension to the steganographic paradigm by merging text and image concealment techniques. Instead of employing a single-layer approach, this model conceals data through a multi-layer process, intertwining one set of data with another to create stego data, which is then securely embedded within a cover image. This innovative model facilitates the con-cealment of larger volumes of data within a single stego container associated with a specific cover image. This model employs a pair of encoders and decoders for encryption and decryption, enabling the processing of two inputs to generate a unified hybrid output. Moreover, the use of a modified 32-bit key ensures the security of user-confidential data. This model expands and enhances the security mechanisms of steganography, enabling the concealment of substantial data volumes. Additionally, the bit allocation adapts dynamically based on the length of the text to be encrypted within the host image. This approach forms a hierarchy involving two steganographic and one encryption model. Importantly, it guarantees a distortion-free process, ensuring that data retrieval after successful concealment yields high accuracy, SSIM, PSNR, and low MSE.