Data Augmentation for JPEG Steganalysis

Data Augmentation for JPEG Steganalysis
复制标题

DOI:
10.1109/wifs53200.2021.9648390
复制
发表时间:
2021-12
期刊:
2021 IEEE International Workshop on Information Forensics and Security (WIFS)
影响因子:
--
通讯作者:
T. Itzhaki;Yassine Yousfi;J. Fridrich
T. Itzhaki;Yassine Yousfi;J. Fridrich
中科院分区:
其他
文献类型:
--
作者:
T. Itzhaki;Yassine Yousfi;J. Fridrich

文献摘要

相似文献

深度卷积神经网络(CNN)在JPEG隐写分析中表现得非常好。然而,它们严重依赖于大型数据集,以避免过度拟合。数据增强是一种流行的技术,可以在不收集新图像的情况下膨胀可用的数据集。对于JPEG隐写分析,研究人员主要使用的增强仅限于旋转和翻转(D4增强)。这是因为计算机视觉中使用的大多数增强都会删除隐写信号。在本文中,我们系统地综述了大量的其他增强技术,并评估了它们在JPEG隐写分析中的优势。
Deep Convolutional Neural Networks (CNNs) have performed remarkably well in JPEG steganalysis. However, they heavily rely on large datasets to avoid overfitting. Data augmentation is a popular technique to inflate the datasets available without collecting new images. For JPEG steganalysis, the augmentations predominantly used by researchers are limited to rotations and flips (D4 augmentations). This is due to the fact that the stego signal is erased by most augmentations used in computer vision. In this paper, we systematically survey a large number of other augmentation techniques and assess their benefit in JPEG steganalysis.