Postdoc at IRISA - knowlege-based inference for ID document recognition

** apologies for multiple copies **

Postdoctoral position on "knowlege-based inference for ID document 
recognition"
- in IDFRAud ANR project (Identity Document Fraud Detection and Profiling)
- at IRISA, Rennes, France

Keywords: knowledge management, inference systems, Semantic Web, 
document recognition

* Context *

The IDFRAud project aims at the automatic recognition of ID documents,
and the detection of ID frauds. Semantic Web technologies will be used
to acquire, organize, and exploit detailed knowledge about hundreds of
ID document models from all countries (e.g., French driving license,
German passport, Spanish ID card). That knowledge will be the basis
for the automatic recognition and fraud detection of ID documents from
scans taken under different wavelengths (e.g., ultraviolet). The
recognition includes identifying the ID model (e.g., French ID card),
extracting useful information (e.g, holder name), and performing
security checks (e.g., UV features).

* Subject *

The first task of the postdoctoral researcher will be to contribute
with other participants of the IDFRAud project in the development of
the knowledge base of ID models. A first version will exist at the
beginning of the postdoc.

The second and main task will be to work on the problem of the
automatic recognition of ID documents given the knowledge base of ID
models, and a collection of ID document analyzers. A first objective
is to find a way, based on the knowledge base, to coordinate calls to
those analyzers in order to infer as much as possible information
about an ID document, and in particular to detect ID fraud. A second
objective is to compile the inference process into a lightweight
executable program that can perform recognition offline, without
access to the knowledge base.

The aspects related to image analysis and information extraction will
be handled by another postdoctoral researcher working on the project.

* Profile *

Applicants should have recently obtained a PhD degree, and show strong
motivation and research potential. Applications should include a
one-page motivation letter, a detailed CV with publication lists and
participations to projects, and 2-3 significant published papers.

Mandatory requirements for applicants:
1. PhD in Computer Science
2. Experience in
    - Semantic Web technologies (RDF(S), SPARQL, triple stores)
    - inference/rule systems
3. Programming skills to ensure autonomy in completing the technical 
tasks of the project

Optional requirements:
- generation/transformation of inference rules or programs

Starting date: September 2015 - January 2016
Duration: up to 24 months
Working environment: in Rennes, at the IRISA laboratory, in the LIS team
Salary: gross salary 2600€ (net income: approximately 2100€)
Contact: Sébastien Ferré <ferre@irisa.fr>, Peggy Cellier <cellier@irisa.fr>
Web: http://www.irisa.fr/LIS/

Received on Wednesday, 15 April 2015 07:17:19 UTC