Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : CReSTIC lab, University of Reims Champagne-Ardenne
Durée : 18 months, extendabl
Contact : valeriu.vrabie@univ-reims.fr
Date limite de publication : 2023-01-30

Contexte :
Within the framework of a collaboration between the Comité de Champagne, Segula Technologies company, and the University of Reims, we are looking to optimize the detection of the presence of vine diseases by developing and integrating imaging and machine learning concepts. This project includes several axes in which 2 PhD students participate and several master internships together with 3 (associate) professors. In this context we are recruiting a PostDoc for 18 months, extendable to 36 months, to complete the team.

Sujet :
The flavescence dorée, a serious and epidemic disease, is one of the two grapevines yellow diseases that might cause a rapid decay in Champagne and other wine regions, being considered as the new phylloxera of the vineyard. To date, the detection approach of the yellows is to collectively explore the vineyard on foot every year to identify affected vines and to perform biomolecular tests by approved laboratories. As the survey is not precise nor optimal enough for a large-scale monitoring, the development and integration of detection solutions based on imagery appear necessary.
We conducted several acquisition campaigns between 2019 and 2022 under controlled and in situ conditions. New CNN-based hierarchical architectures developed by our lab have proven to be efficient on separate data sets composed of multispectral images. But the generalization capabilities of these models are not sufficient to compensate for the variability induced by endogenous and exogenous factors. The objective is to propose new detection strategies for grapevine yellows that are more robust to this variability. Several tracks could be explored, including, but not limited to:
• Fusion of multispectral images through vegetation indices to be defined and used within adapted Deep Learning architectures
• Fusion of features extracted with heterogeneous Deep Learning models from multispectral images
• Identification of an optimal subset of multispectral images and development of reinforcement or continuous learning approaches
It might also be interesting to design semi-supervised approaches to take advantage of the possibility of acquiring many unlabeled images during new acquisition campaigns to which the Postdoc will be associated.

Profil du candidat :
• Self-motivated scientist seeking to pursue a scientific career, holding a Ph.D. or in the process of completing it, in a relevant field of machine/deep learning or other relevant fields

Formation et compétences requises :
• Excellent knowledge and skills in AI learning, machine learning and data science with hand-on skill and experience
• Excellent experience, knowledge, and skills in programming languages, especially Python (environment Tensorflow, Pytorch, Keras, Pandas, Scikit-learn, etc.)
• Deep understand of digital image processing; prior experience in working with image analysis projects (industrial or academic) will be a plus
• Independent and passionate about data science projects, however good team player, able to undertake research projects together with other team members
• Excellent scientific/technical writing skills and communication capability; ability to present research achievements at internal/external seminars, conferences and journals

Adresse d’emploi :
The PostDoc will be based at the CReSTIC lab of the University of Reims Champagne-Ardenne on the Moulin de la Housse campus in Reims. Within the framework of the project, he/she may be required to intervene to the project partners, the Comité de Champagne in Epernay or Segula Technologies in Reims downtown.