Domain Generalization for Document Authentication against Practical Recapturing Attacks

TL;DR

This paper presents a domain generalization approach for document recapture detection using a Siamese network with a novel loss function, effectively handling variations across devices, substrates, and document types.

Contribution

It introduces a new Siamese network-based scheme with combined metric learning and forensic techniques, improving robustness in practical recapturing attack scenarios.

Findings

Outperforms state-of-the-art methods across various experimental settings.

Achieves less than 5% APCER and 5.56% BPCER in challenging cross-device document scenarios.

Demonstrates robustness against variations in printing, imaging, and document types.

Abstract

Recapturing attack can be employed as a simple but effective anti-forensic tool for digital document images. Inspired by the document inspection process that compares a questioned document against a reference sample, we proposed a document recapture detection scheme by employing Siamese network to compare and extract distinct features in a recapture document image. The proposed algorithm takes advantages of both metric learning and image forensic techniques. Instead of adopting Euclidean distance-based loss function, we integrate the forensic similarity function with a triplet loss and a normalized softmax loss. After training with the proposed triplet selection strategy, the resulting feature embedding clusters the genuine samples near the reference while pushes the recaptured samples apart. In the experiment, we consider practical domain generalization problems, such as the variations in printing/imaging devices, substrates, recapturing channels, and document types. To evaluate the robustness of different approaches, we benchmark some popular off-the-shelf machine learning-based approaches, a state-of-the-art document image detection scheme, and the proposed schemes with different network backbones under various experimental protocols. Experimental results show that the proposed schemes with different network backbones have consistently outperformed the state-of-the-art approaches under different experimental settings. Specifically, under the most challenging scenario in our experiment, i.e., evaluation across different types of documents that produced by different devices, we have achieved less than 5.00% APCER (Attack Presentation Classification Error Rate) and 5.56% BPCER (Bona Fide Presentation Classification Error Rate) by the proposed network with ResNeXt101 backbone at 5% BPCER decision threshold.