Sorry, you need to enable JavaScript to visit this website.

A study on the invariance in security whatever the dimension of images for the steganalysis by deep-learning

Citation Author(s):
Kévin Planolles, Marc Chaumont, Frédéric Comby
Submitted by:
Marc CHAUMONT
Last updated:
31 May 2023 - 4:01pm
Document Type:
Presentation Slides
Document Year:
2023
Event:
Presenters:
CHAUMONT
Paper Code:
Paper n° 906 IFS-P4: Watermarking, Data Hiding and Human Factors in Security
 

In this paper, we study the performance invariance of convolutional neural networks when confronted with variable image sizes in the context of a more ”wild steganalysis”. First, we propose two algorithms and definitions for a fine experimental protocol with datasets owning ”similar difficulty” and ”similar security”. The ”smart crop 2” algorithm allows the introduction of the Nearly Nested Image Datasets (NNID) that ensure ”a similar difficulty” between various datasets, and a dichotomous research algorithm allows a ”similar security”. Second, we show that invariance does not exist in state-of-the-art architectures. We also exhibit a difference in behavior depending on whether we test on images larger or smaller than the training images. Finally, based on the experiments, we propose to use the dilated convolution which leads to an improvement of a state-of-the-art architecture

up
0 users have voted: