Research on image steganography based on conditional Invertible Neural Network

To improve the security of image steganography, an image steganography method based on conditional Invertible Neural Network is proposed in this paper. First, we design a conditional Invertible Neural Network to obtain high-quality stego images with rich high-level semantic information and clear spatial details. Based on the conditional directivity of conditional Invertible Neural Network, we can accurately adjust the semantic information of stego image and ensure the controllability of stego image content. We introduce a dual cross-attention module into the network structure. The integration of dual cross-attention modules enhances feature extraction and captures complex image details to improve steganographic accuracy. In addition, the introduction of the convolutional block attention module in the convolutional layer direct the model's focus to key image regions, refining stego image quality. We increase the number of convolutional blocks, which improves the efficiency of feature extraction and reuse. A large number of experiments are carried out on the datasets. For cover and stego image pairs, the PSNR value reached 43.62dB, and for secret and recovery image pairs, the PSNR value reached 46.48dB. Experimental results show that the image quality and security of this method are better than other state-of-the-art image steganography methods.