Invisibile Image Steganography via Deep Learning

• 1、School of Cyberspace Security, Beijing University of Posts and Telecommunications, Beijing 100876

Abstract：Steganography and steganalysis are main content of information hiding, they always make constant progress in confrontation. There is large consent that measures based on deep learning have outperformed conventional approaches in steganalysis, which have shown that deep learning is very promising for the information hiding area. And in the last two years, there are also several works using deep learning to achieve the procedure of steganography. While these works still have problems in invisibility. We proposed a new steganography model based on deep learning. Our model can conceal a secret gray image into a color cover image, and can reveal the secret image out successfully. And the invisibility of our model was improved significantly. A mixed loss function is proposed for the steganography, which will spped up the training. Experiment results show that ISGAN can achieve start-of-art performance on LFW and Pascal VOC12 datasets.

Keywords： Information security technology Image steganography Convolutional neural networks Mixed loss function