Landscape More Secure Than Portrait? Zooming Into the Directionality of Digital Images With Security Implications

TL;DR

This paper investigates how image orientation affects security applications, revealing that accounting for directionality improves the robustness and performance of media security methods across various tasks.

Contribution

It systematically analyzes the causes of image directionality in acquisition pipelines and demonstrates performance improvements in security applications by addressing this factor.

Findings

Symmetrization suppresses important image properties.

Unaddressed directionality leads to overfitting and vulnerabilities.

Properly modeling directionality enhances security method performance.

Abstract

The orientation in which a source image is captured can affect the resulting security in downstream applications. One reason for this is that many state-of-the-art methods in media security assume that image statistics are similar in the horizontal and vertical directions, allowing them to reduce the number of features (or trainable weights) by merging coefficients. We show that this artificial symmetrization tends to suppress important properties of natural images and common processing operations, causing a loss of performance. We also observe the opposite problem, where unaddressed directionality causes learning-based methods to overfit to a single orientation. These are vulnerable to manipulation if an adversary chooses inputs with the less common orientation. This paper takes a comprehensive approach, identifies and systematizes causes of directionality at several stages of a typical acquisition pipeline, measures their effect, and demonstrates for three selected security applications (steganalysis, forensic source identification, and the detection of synthetic images) how the performance of state-of-the-art methods can be improved by properly accounting for directionality.