We are looking for a motivated candidate with a PhD degree in the field of computer vision and machine learning. Experience in multisensor data analysis and fusion expertise is an added value.Practical and analytical skills, liking experimentation, ability to work in a team, strength of proposal and ease of writing will be appreciated qualities. The candidate must also have a good level of English.
Location: PRISME Laboratory, University of Orléans, Orléans La Source
Term of contract: 14 months with the possibility of extension
Expected starting date: February-March 2021
Context and objectives
The PRISME Laboratory and an industrial partner aim to develop a greenhouse prototype for crop production in a controlled system, with an environmentally friendly production framework. The objective is to develop an infrastructure both hardware and software enabling the acquisition of observation data from the crop eco-system and to extract relevant information useful for monitoring the crop production. The results obtained will be used to better understand the interactions between the components of nutrition, climatic factors within the greenhouse and the plant, in order to obtain the desired quality and optimal yields. Several types of sensors will be used, including multispectral and even hyperspectral imaging, which will provide interesting information on plant activity.
We aim to use data from different sensors to automatically monitor plants growth and assess their health. In this context, the work consists in developing methods and algorithms capable of generating efficient crop monitoring models, using multispectral image data, time series, ... and relying on machine learning approaches. On the other hand, it is also necessary to deploy 3D reconstruction techniques in order to provide biomass assessment with geometric information. Finally, a fusion of different information will be considered to increase the performance of the methods.
- Detailed curriculum vitae
- One-page research statement
- Names and contact details of at least two referees
(c) GdR 720 ISIS - CNRS - 2011-2020.