About

In the context of rapid climate change and decreases in the use of phytosanitary products due to environmental damages, affecting temperate perennial tree functioning and yield, the selection of fruit tree genotypes able to maintain their performance under suboptimal conditions becomes unavoidable. In order to evaluate the characteristics on large variety panels, needed for varietal testing trial, it is essential to carry out measurements that can be tedious and often lacking of objectivity. In this context, faster, more precise and reproducible phenotyping methods often based on images (RGB, multi or hyperspectral) or LIDARs (3D scanner) acquisition are being developed.

Terrestrial LIDARs appear as promising tools for the reconstruction of 3D plant structures and to broaden the architectural characterization of large variety panels. The methodology is based on the calculation of distances to plant entities using a large number of laser beams to reconstruct scattered point clouds in 3D. In this context, the AGAP unit has been developing methods for automatic reconstruction of plant architecture potentially guided by the user through the PlantScan3D software (Boudon et al., 2014). In the context of the studies carried out in the particular case of the apple tree, initial works consisted in very accurately scanning trees in winter (without leaves) and in summer, and in setting up a first point cloud treatment pipeline for the reconstruction of the architecture of the varieties.

However, many challenges remain, especially on fruit trees (i) to estimate the production of fruit of a tree from point cloud or airborne imagery, (ii) to test the influence of the levels of precision of the data in the reconstruction in order to optimize acquisition protocols and / or extend protocols to low cost LIDARs potentially accessible for breeders. The engineer to be recruited will work on these axes as part of a project funded by the European Union (INVITE project).

Qualifications

The candidate must have a master's degree in Computer Science / Bioinformatics / Signal Processing / Digital Agriculture.

• He / she must have a taste for biology and agronomy.
• He / she must be curious, autonomous and able to work in a multidisciplinary context (agronomy, computer science, statistics, etc.).
• He / she must know how to program with the Python language. Knowledge in C ++ and R are a plus.

Gross salary: about 2350 euros gross / month