Section: New Results

Reconstruction of macroscopic forms from images and characterization of their variability

Participants : Guillaume Cerutti, Christophe Godin, Jonathan Legrand, Katia Mirande.

• Research Axes: RA1 (Representations of forms in silico) & RA3 (Plasticity & robustness of forms)

• Key Modeling Challenges: KMC3 (Realistic integrated digital models)

To study the variability of macroscopic forms resulting from development, it is necessary to both develop digital reconstruction methods, typically based on image acquisitions, and statistical tools to define notions of distance or average between these forms. The automatic inference of computational representations of forms or organ traits from images of different types is therefore an essential step, for which the use of prior knowledge can be very beneficial. Realistic synthetic models of forms can guide the reconstruction algorithms and/or assess their performances. Computational representations of forms can then be used to analyze how forms vary at the scale of a population, of a species or between species, with potential applications in species identification and genetic or environmental robustness estimation.

Automatized characterization of 3D plant architecture. The digital reconstruction of branching and organ forms and the quantification of phenotypic traits (lengths of internodes, angles between organs, leaf shapes) is of great interest for the analysis of plant morphology at population scale. We develop an automated processing pipeline that involves the 3D reconstruction of plant architecture from RGB image acquisitions performed by a robot, and the segmentation of the reconstructed plant into organs. To provide validation data for the pipeline, we designed a generative model of Arabidopsis thaliana simulating the development of the plant architecture at organ scale. This model was used to develop the method for the measurement of angles of organs and test its accuracy. In a second phase, the model will be used to generate training data for machine learning techniques introduced in the reconstruction methods.

This work is part of the ROMI project.

Identification of plant species from morphological traits. The description of morphological traits of the various organs of the plants (leaves, bark, flowers and fruits) is essential for the characterization of a phenotype, and is highly relevant in the context of species or variety identification. In the context of tree species identification from RGB images of their organs, we study methods to represent the morphological charateristics of the plant organs, and the way to combine those different sources of information to enhance the classification performance. We demonstrated that botany-inspired descriptors of bark improves tree species classification based on leaves [13]. We also explore the possibility of using deep learning techniques to train a system to extract botanically relevant information from images [3].

This work is part of the ReVeRIES ANR project, in which the team is not directly involved.

This work has led to a publication in Ecological Informatics and to a participation at the International Worskhop on Image Analysis Methods for the Plant Sciences in Nottingham in January 2018.