This site uses cookies to improve your user experience and to achieve audience statistics.
I acceptI refuseKnow more
fr es
fr es
All our news

Research Project – IDFRAud, an operational framework for identity document fraud detection

IDFraud project

IDFRAud is an industrial collaborative research project (ANR-14-CE28-0012) coordinated by ARIADNEXT (with Ahmad Montaser Awal as the project’s scientific coordianator). The IDFRAud project is funded by the French National Research Agency (ANR) and co-funded by the French Government Defense procurement and technology agency (DGA), under the program “Freedom and protection of citizens and residents” 2014.

IDFRAud is also labeled by the Image & Network (IR) competitiveness pole. It has established a new kind of collaboration by creating a partnership associating:

  • Industrials (ARIADNEXT)
  • Academic researchers (SemLIS and LinkMedia teams of the University of Rennes1 /Institut de Recherche en Informations et Systèmes Aléatoires (IRISA)
  • And experts in fraud detection and analysis :
    – the “Pôle Judiciaire de la Gendarmerie Nationale/ Institut de Recherche Criminelle de la Gendarmerie Nationale” (IRCGN)
    – the “École Nationale Supérieure de la Police (ENSP)/Bureau de la Fraude Documentaire/Police Aux Frontières”

The project started in February 2015 and lasted 42 months. It was granted an ANR subvention (co-financed by the DGA) of an amount of 905k€ for a global cost of 2 328k€.


Fight against document fraud


Identity-related frauds represent a major risk to the society safety given its serious consequences. These consequences may vary from small but very frequent frauds (ex. small credits) to transnational organized crimes and terrorist actions.

An increasing number of false identity documents (IDs) have been detected during the last few years, according to several official studies around the globe. Traditional investigation methods applied to IDs rely on the document authentication by an expert, which significantly reduces the chances of careful verifications in many administrative and commercial entities. In addition, existing automatic ID control tools have shown several limitations, especially, the nonexistent evolving capacities.

The main originality of IDFRAud is to propose an automatic solution for ID verification that can handle documents issued from a large set of countries thanks to an evolving knowledge base. Such an intelligent solution aims at replacing the manual (human) analysis by an automatic process, in the absence of an expert, therefore reducing a lot the processing time. Experts from national security authorities along with academic and industrial partners work side by side to propose an automatic solution for document analysis and verification.


Identity document analysis and verification


The IDFRAud project aims at proposing an operational and automatic framework for the detection and analysis of fraudulent identity documents. It relies on the inter-connectivity between three modules: ID verification, ID knowledge management, and ID fraud analysis.

The verification module relies on the classification and information extraction. The ID family (type, country and version) is first identified by a fine-classification based on both image and textual representations of the image. Textual content is then extracted thanks to preprocessing steps adapted to the document class. Document integrity verification is finally obtained from both representations of the document. The proposed approaches are based on the latest advances in the artificial intelligence and deep learning techniques.

In order to guarantee an adapted behavior at each ID analysis step, the identity document descriptions are organized by a knowledge management module. The objective of this module is to facilitate the addition of new document descriptions and the description completion of existing documents, which makes the entire system flexible and evolving. Moreover, IDFRAud system includes a data analyzer in order to detect forensic links from fake IDs datasets. This detection will be associated to clustering methods in order to better analyze and visualize fraud cases.


The results of the project


The technological advancements of the project resulted in significant improvements in ARIADNEXT’s identity document verification and analysis service, IDCHECK.IO. This service was tested during two field experiments organized by the ENSP, the DZ-PAF Ouest[1] and the DCSP[2] – DDSP[3] 35 in 2016 and 2017 and a third experiment conducted with the ENSP and the DZ-PAF Sud Est in 2018. This service has allowed ARIADNEXT to become a key player in a fast-growing market.

[1] Direction Zonale de la Police Aux Frontières
[2] Direction Centrale de la Sécurité Publique
[3] Direction Départementale de la Sécurité Publique


Follow us!