Receipt Dataset for Document Forgery Detection

Beatriz Martínez Tornés, Théo Taburet, Emanuela Boros, Kais Rouis, Antoine Doucet, Petra Gomez-Krämer, Nicolas Sidere, Vincent Poulain d’Andecy

Abstract: The widespread use of unsecured digital documents by companies and administrations as supporting documents makes them vulnerable to forgeries. Moreover, image editing software and the capabilities they offer complicate the tasks of digital image forensics. Nevertheless, research in this field struggles with the lack of publicly available realistic data. In this paper, we propose a new receipt forgery detection dataset containing 988 scanned images of receipts and their transcriptions, originating from the scanned receipts OCR and information extraction (SROIE) dataset. 163 images and their transcriptions have undergone realistic fraudulent modifications and have been annotated. We describe in detail the dataset, the forgeries and their annotations and provide several baselines (image and text-based) on the fraud detection task.

External IDs:doi:10.1007/978-3-031-41682-8_28