# Interpretability of Deep High-Frequency Residuals: A Case Study on SAR Splicing Localization

## TL;DR

This paper explores how deep high-frequency residuals can help identify and understand image manipulations in SAR images.

## Contribution

The study reveals that DHFRs not only detect image splicing but also reveal the nature of editing techniques used.

## Key findings

- DHFRs serve as a visual aid for identifying manipulated regions in SAR images.
- The appearance of DHFRs correlates with the high-frequency energy content in tampered zones.
- DHFRs show significant interpretability properties despite being derived from deep learning models.

## Abstract

Multimedia Forensics (MMF) investigates techniques to automatically assess the integrity of multimedia content, e.g., images, videos, or audio clips. Data-driven methodologies like Neural Networks (NNs) represent the state of the art in the field. Despite their efficacy, NNs are often considered “black boxes” due to their lack of transparency, which limits their usage in critical applications. In this work, we assess the interpretability properties of Deep High-Frequency Residuals (DHFRs), i.e., noise residuals extracted from images by NNs for forensic purposes, that nowadays represent a powerful tool for image splicing localization. Our research demonstrates that DHFRs not only serve as a visual aid in identifying manipulated regions in the image but also reveal the nature of the editing techniques applied to tamper with the sample under analysis. Through extensive experimentation on spliced amplitude Synthetic Aperture Radar (SAR) images, we establish a correlation between the appearance of the DHFRs in the tampered-with zones and their high-frequency energy content. Our findings suggest that, despite the deep learning nature of DHFRs, they possess significant interpretability properties, encouraging further exploration in other forensic applications.