Reference : UPR3346-NADMAA-071
Workplace : FUTUROSCOPE
Date of publication : Thursday, July 21, 2022
Type of Contract : FTC Scientist
Contract Period : 24 months
Expected date of employment : 1 September 2022
Proportion of work : Full time
Remuneration : Between 2663 € and 3069 € grossmonthly according to experience
Desired level of education : PhD
Experience required : Indifferent
The CNRS, within the Pprime Institute, on the Futuroscope site, is recruiting a researcher in partnership with the company IKOMIA as part of the government's recovery plan on a project entitled ODOLMA (Optimisation and Development of Software Tools for Optical Metrology and Analysis of Living Things). The company IKOMIA is developing a multi-OS software platform that includes numerous image processing and artificial intelligence tools and that can be used freely in which these new tools will be integrated.
The candidate main task will be to develop and perfect a Lagrangian tracking technique for measuring particle velocities. This volumetric technique based on laser or diode volume illumination and several cameras will allow trajectories to be tracked with good resolution.
The objective of the work to be carried out by the candidate is to reconstruct a three-dimensional displacement field of particles from recordings of these particles at short time intervals and in different points of view (typically 3). The concentration of particles shall be as high as possible.
The candidate will rely on existing camera calibration and triangulation libraries developed in C++. The candidate will have to develop a particle matching algorithm on several successive instants in order to obtain particle trajectories. The candidate will have to introduce regularisation constraints on the displacements to reduce the uncertainties on the location of the particles and allow the verification of the continuity equation. The candidate will also have to take into account the particles that enter or leave the measurement volume. The algorithms developed will be validated on synthetic data and applied to experimental data already obtained elsewhere. A comparison may be made with commercial software.
The second part of the work consists, from the 3D Lagrangian tracking developed in the field of fluid mechanics and by integrating the company's skills in Artificial Intelligence, in developing solutions for predicting fish movement. The analysed fish trajectories will be related to the hydraulic characteristics of the flows in order to predict displacements on other configurations.
The candidate will have to carry out the following activities:
- Developing a Lagrangian tracking technique with existing tools and libraries in the team
- Validating the algorithms on synthetic data
- Applying the method to to 4D measurements
- Predicting fish trajectories from Lagrangian tracking and artificial intelligence
- Validating the work in conferences and scientific journals
The candidate should have strong skills in :
- applied mathematics
- fluid mechanics
- object-oriented programming, preferably in C++,
- optimization and image processing.
The candidate should have the following knowledge:
- Written and oral presentation skills
- English language: B1 to B2 (Common European Framework of Reference for Languages)
- Respect the conditions of use of experimental devices
- Lead a meeting
- Apply the rules of hygiene and safety
- Work in a team
- Sense of initiative
- Critical thinking
- Sense of organization
The CNRS (Centre national de la recherche scientifique) is a public multidisciplinary research organization under the supervision of the Ministry of Higher Education, Research and Innovation, which aims to advance knowledge for the benefit of society.
The Pprime Institute is a UPR (Unité Propre de Recherche) of the CNRS in the fields of Physical Sciences and Engineering Sciences. Its activities cover a wide scientific spectrum, from the physics of materials to fluid mechanics, solids, mechanical engineering and energy.
The HYDEE team develops research activities within the DFTC department in the field of hydrodynamics and sedimentary transport of complex fluids.
The company Ikomia, a recent start-up, has developed a multi-OS software platform that brings together numerous image processing and artificial intelligence tools and that can be used freely.
In this ODOLMA collaborative project with IKOMIA, we are implementing new Lagrangian Tracking (LPT) and enhanced learning tools to predict fish movements.
We talk about it on Twitter!