We present in a descriptive way the first results of our study of the problem of document image tampering detection. We aim at helping the community by establishing certain guidelines in what refers to the categorization and targeting of this problem. We propose a categorization of the main types of forgeries performed by a direct manipulation of the document image. That applies to most of the cases we observed in real world forged documents according to our sources from external private companies. In addition, we describe a set of visual clues result of these tampering operations that can be addressed when developing automatic methods for its detection.
CITATION STYLE
Cruz, F., Sidère, N., Coustaty, M., Poulain d’Andecy, V., & Ogier, J. M. (2019). Categorization of Document Image Tampering Techniques and How to Identify Them. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11188 LNCS, pp. 117–124). Springer Verlag. https://doi.org/10.1007/978-3-030-05792-3_11
Mendeley helps you to discover research relevant for your work.