JPEG steganalysis based on ResNeXt with Gauss partial derivative filters
- PDF / 3,862,696 Bytes
- 18 Pages / 439.642 x 666.49 pts Page_size
- 99 Downloads / 186 Views
JPEG steganalysis based on ResNeXt with Gauss partial derivative filters Ante Su1,2 · Xiaolei He1,2
· Xianfeng Zhao1,2
Received: 6 November 2019 / Revised: 14 June 2020 / Accepted: 13 July 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The latest research indicates that the image steganalysis has been greatly promoted by convolutional neural networks (CNNs). This study further addresses the problem of JPEG steganalysis through proposing a novel CNN architecture in which Gauss partial derivative (GPD) filters and two constructed blocks based on ResNeXt are integrated. In the proposed network, multi-order GPD filters are designed as the pre-processing layer to generate residual images, which can effectively capture sufficient embedding disturbance in texture and edge regions. Furthermore, referring to ResNeXt, two multi-branch blocks are constructed and aggregated to fully exploit the residual images to generate image features for classification. Numerous experiments have been conducted against J-UNIWARD on the public dataset to demonstrate the effectiveness and remarkable performance of the proposed network. Experimental results prove that the proposed network makes better performance than state-of-the-art CNN-based method J-Xu-Net and SCA-GFR. Source code is available via GitHub: https://github.com/Ante-Su/RXGNet. Keywords Steganalysis · JPEG · CNNs · Gauss partial derivative
This work was supported by NSFC under 61972390, U1736214, 61872356, 61902391 and 61802393, and National Key Technology R&D Program under 2019QY0701 Xiaolei He
[email protected] Ante Su [email protected] Xianfeng Zhao [email protected] 1
State Key Laboratory of Information Security, Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093, China
2
School of Cyber Security, University of Chinese Academy of Sciences, Beijing 100093, China
Multimedia Tools and Applications
1 Introduction Steganography is a technology for covert communication, which embeds a secret message into an innocent cover. As the adversary of steganography [5], image steganalysis is committed to detecting the existence of steganographic manipulation. Over the past decades, most of modern image steganalysis methods are based on machine learning with an image feature set, such as SRM [8] and its variant max-SRM [6] for spatial domain, DCTR [11] and GFR [7, 18] for JPEG domain. Ensemble classifier [13] is usually used as the classifier for image steganalysis. Although the combination of image feature set and ensemble classifier has achieved remarkable detection performance, it remains challenging to manually devise image features that contain abundant embedding trace leaved by steganography. The design of image feature requires researchers to have a wealthy of experience, and the design process is time-consuming. Therefore, researchers are dedicated to conducting researches on the automatic learning of image features. In recent years, CNN-based steganalyzers have contributed to a great success in s
Data Loading...
