Post-doctoral position


  Semantic Techniques for the management of Earth Observation Data

Key words: geographic data, image metadata, ontologies, semantic
indexing, semantic search


      Context: H2020 CANDELA project (2018-2020)

http://www.candela-h2020.eu/

With the advent of the Copernicus program, a wealth of Earth Observation
(EO) data will be made available. The applications and services
developed on this data are increasingly adopting big data technologies
for an efficient data storage and processing, and to support data
analytics and application development. The main objective of the CANDELA
project is to allow the creation of value from Copernicus data through
the provisioning of modelling and analytics tools (data collection,
processing, storage and access will be provided by the Copernicus Data
and Information Access Service (DIAS). The implementation starts by
putting in place a set of powerful tools that drastically lowers the
cost of getting familiar with the data and creating new services. These
modules will provide machine learning, data mining, data fusion and web
semantics solutions, combining the Copernicus data and information with
other non-EO data sources to derive novel applications and services.
CANDELA will demonstrate the breadth of project capabilities by means of
two reference scenario, a first one in macro-economics and agriculture;
and a second one for forest health monitoring (FHM).


      Position Description

Within task 2.2 of CANDELA, the postdoctoral work aims at designing and
implementing semantic search capabilities over the Earth observation
image collections and their descriptive semantic data. The descriptive
semantic data will include a semantic representation of various types of
data: (i) the image metadata (type of sensor, distance from the ground,
image quality, geolocalization ...), (ii) data extracted from images or
image series and (iii) open data that provide additional information of
interest. They will make it easier to search for relevant images in
these collections. To this end, we will use ontologies and external
resources like LOD (Linked Open Data). This work will be jointly carried
out with data producers and partners involved in use-case definitions
for one target application field, agriculture. Main tasks and expected
output:

  * to set up an annotation model and process to populate relevant
    domain ontologies that will model metadata and geographic data in
    the selected application domain (possibly agriculture);
  * to build one or several open ontologies for at least one case study;
  * to produce and publish semantic metadata of the Earth observation
    image collections, based on the standard metadata and the
    ontologies;
  * to design a search tool to identify images combining spatial,
    temporal and other kinds of dimensions.

Research issues to be investigated in this project include ontology
engineering, semantic data for image annotation and search, web as a
source of formal vocabularies reusable to annotate images, and semantic
data alignment. The candidate will define a vocabulary for the
integration of geolocalized data, an annotation model as well as methods
and support tools for populating geolocalized ontologies for image
indexing, and then a semantic search tool to query these annotated
images. Methods and tools will be adapted to at least one of the
application fields (agriculture or forestry).


      Requirements for this position

Applicants are required to have a PhD in Computer Science, a strong
background in semantic web technologies, ontology engineering, linked
data management and query. Fluency in written / spoken English is
required too. Experience on geographical data, a good publication record
and strong programing skills, as well as fluency in French language will
be a plus.


      Work environment

/Host lab/: Institut de Recherche en informatique de Toulouse (IRIT) -
UPS, 118 Route de Narbonne F-31062 Toulouse Cedex - France

/Duration/: 12 months - starting on Sept 2018, 1st

/Host group/: MELODI http://www.irit.fr/-Equipe-MELODI-
The candidate will work with three permanent academic researchers, and a
PhD student, and will collaborate with the partner companies in the
project, mainly Thalès Aliena Space, Terranis and ATOS.

/Income/: between 1800 and 2700 euros free of taxes ("net") monthly 
according to past experience


      How to apply?

Applications will be accepted until the position is closed. Applicants 
should send a full Curriculum including a complete list of publications, 
a cover letter indicating their research interests, achievements to date 
and vision for the future, as well as either support letters or the name 
of 2 persons that have worked with them.

Contact:
N. Aussenac-Gilles (aussenac@irit.fr) +33 5 61 55 82 93 or C. Trojahn
Cassia.Trojahn@irit.fr