Публікація:
Automated Detection of Forgeries in Documents Using Segmentation Neural Networks

Завантаження...
Зображення мініатюри

Дата

2025

Назва журналу

ISSN журналу

Назва тома

Видавництво

ХНУРЕ

Дослідницькі проекти

Організаційні підрозділи

Видання журналу

Анотація

This paper presents a neural network–based approach for automated detection of forged areas in PDF documents. The method converts document pages into image representations and applies semantic segmentation to localize manipulations such as inserted signatures or replaced text. U-Net and DeepLabV3+ architectures were trained on a dataset of authentic and synthetically forged samples. Experimental results show that segmentation models achieve high accuracy (IoU ≈ 0.91), outperforming traditional metadata-based techniques. The proposed solution can be integrated into electronic document management and cybersecurity systems for automated authenticity verification.

Опис

Ключові слова

document forgery detection

Бібліографічний опис

Lohvynenko S. R. Automated Detection of Forgeries in Documents Using Segmentation Neural Networks / S. R. Lohvynenko, V. V. Prosolov // Computer and information systems and technologies : Eighth International Scientific and Technical Conference, October 09-10, 2025. – Kharkiv : NURE, 2025. - P. 116.

DOI