A New High Capacity Image Steganography Method Combined With Image Elliptic Curve Cryptography and Deep Neural Network

Xintao Duan,Baoxia Li,Nao Liu,Chuan Qin,Daidou Guo,Mengxiao Gou

doi:10.1109/access.2020.2971528

Abstract

Image steganography is a technology that hides sensitive information into an image. The traditional image steganography method tends to securely embed secret information in the host image so that the payload capacity is almost ignored and the steganographic image quality needs to be improved for the Human Visual System(HVS). Therefore, in this work, we propose a new high capacity image steganography method based on deep learning. The Discrete Cosine Transform(DCT) is used to transform the secret image, and then the transformed image is encrypted by Elliptic Curve Cryptography(ECC) to improve the anti-detection property of the obtained image. To improve steganographic capacity, the SegNet Deep Neural Network with a set of Hiding and Extraction networks enables steganography and extraction of full-size images. The experimental results show that the method can effectively allocate each pixel in the image so that the relative capacity of steganography reaches 1. Besides, the image obtained using this steganography method has higher Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity Index(SSIM) values, reaching 40dB and 0.96, respectively.

Highlights

In the context of informatization, the development of network security and multimedia has brought a great convenience to people’s daily life and work, but it has exposed more and more security issues
Through some features of the host image, the secret image is hidden into the host image, which is image steganography
Traditional image steganography schemes are generally implemented by artificial design algorithms, such as the space domainbased LSB [1], [2], HUGO [3], WOW [4] hidden methods and transform domain-based DCT [5], DWT [6] and DFT [7] hidden methods

Summary

INTRODUCTION

In the context of informatization, the development of network security and multimedia has brought a great convenience to people’s daily life and work, but it has exposed more and more security issues. In [20], coverless information steganography based on Deep Convolution Generation Against Network (DCGAN) is proposed In this process, there is no need to modify the host information, and it has high security, but the steganographic capacity is still limited. In the left and right blocks of the network, the feature vectors of 64 components are mapped to the required number of categories using a convolution of 3 × 3, and the secret image and the host image are calculated using the Sigmoid activation function.In this network, stride is set to 1 and padding is set to 1, so it is guaranteed that the size of the image remains the same during the convolution process. 7) Convert the value in 6) to a plain text image

EXPERIMENTAL ANALYSIS

STRUCTURAL SIMILARITY INDEX ANALYSIS OF

CONCLUSION