Meriam Bayoudh
Temporary Lecteur (ATER) at IUT Robert Schuman, Strasbourg university since Septembre 2015. I am a member of SDC team at ICube laboratory.
Research
Keywords: Remote sensing, machine learning, Inductive Logic Programming (ILP) object-based image analysis, segmentation, supervised classification, , Land cover/use maps, satellite images.
PhD Thesis
Title: Automatic learning of structural knowledge from geographic information for updating land cover maps
Thesis Supervisor: Richard Nock (Full Professor, CEREGMIA, UAG)
Thesis Co-Supervisor: Gilles Richard (Professor, IRIT, UPS)
Co-advisor: Emmanuel Roux (Reasearcher, SIC team , IRD)
Funding: FEDER, French Guiana region.
Summary: Classical methods for satellite image analysis appear inadequate for the current bulky data flow. Thus, making the interpretation of such images automatic becomes crucial for the analysis and management of phenomena changing in time and space, observable by satellite. Consequently, this work aims to contribute to the dynamic land cover cartography from satellite images, by expressive and easily interpretable mechanisms, and by explicitly taking into account structural aspects of geographic information. It is part of the \textit{object-based} image analysis framework, and assumes that it is possible to extract useful contextual knowledge from existing maps. Thus, a supervised parameterization method of an image segmentation algorithm is proposed, taking a segmentation derived from a land cover map as reference. Secondly, a supervised classification of geographical objects is presented. It combines machine learning by Inductive Logic Programming and the \textit{Multi-class Rule Set Intersection} approach. Finally, prediction confidence indexes are defined to assist interpretation. These approaches are applied to the French Guiana coastline cartography. The results demonstrate the feasibility of the segmentation parameterization, but also its variability as a function of the reference map classes and of the input data. Nevertheless, methodological developments allow to consider an operational implementation of such an approach. The results concerning the object supervised classification show that it is possible to induce expressive classification rules that convey consistent and structural information in a given application context and lead to reliable predictions, with overall accuracy and Kappa values equal to, respectively, 84.6\% and 0.7. In conclusion, this work contributes to the automation of the dynamic cartography from remotely sensed images and proposes original and promising perspectives.
Teaching
Besides of my actual experience at strasbourg university, i teached at Angers university. Below the main teached themes :
* Algorithmic and pascal development * Object oriented design with c# * C2i Certification * Web development: HTML,CSS, Java script, php * Operating Systems Development * Administration and networks
Supervison
* Blot Kévin (Licence 3 pro) * Florian Biteau (Licence 3 pro) * Blot Kévin, Marzin Kévin, Leroux Pierre-Antoine (Licence 3 pro) * Bertrand Antoine (Master 2)
Publications
- Revue Communications
- Meriam Bayoudh, Emmanuel Roux, Gilles Richard, Richard Nock: Structural knowledge learning from maps for supervised land cover/use classification: Application to the monitoring of land cover/use maps in French Guiana. Computers & Geosciences 76: 31-40 (2015)
- Meriam Bayoudh, Henri Prade, Gilles Richard:Evaluation of analogical proportions through Kolmogorov Complexity. Knowledge Based System 29: 30-30 (2012)
- International Communications with proceedings
- Meriam Bayoudh, Henri Prade, Gilles Richard: A Kolmogorov Complexity View of Analogy: From Logical Modeling to Experimentations. SGAI Conf. 2010: 93-106
- International Communications without proceedings
Automatic learning of structural knowledge from geographic information for updating land cover maps, Symposium of the Latin American Society for Remote Sensing and Spatial Information Systems (SELPER), 2012.
- National Communications without proceedings
- Meriam Bayoudh, Gilles Richard, Emmanuel Roux, Richard Nock: Apprentissage de connaissances structurelles à partir de cartes et classification multi-classes: Application à la mise à jour de cartes d’occupation du sol, PFIA, RJCIA, 2013
Contact
Meriam Bayoudh ICube Laboratory Télécom Physique Strasbourg 300 bd Sébastien Brant - CS 10413 F - 67412 Illkirch cedex
Office: C320 Phone: +33 (0)3 68 85 44 11 Email: bayoudh@unistra.fr