A5072434092- Spectroscopy and Chemometric Analyses
- Sensory Analysis and Statistical Methods
- Advanced Chemical Sensor Technologies
- Leaf Properties and Growth Measurement
- Fermentation and Sensory Analysis
- Remote Sensing in Agriculture
- Data Visualization and Analytics
- Advanced Statistical Methods and Models
- Smart Agriculture and AI
- Horticultural and Viticultural Research
- Plant and animal studies
- Species Distribution and Climate Change
- Water Quality Monitoring and Analysis
- Biochemical Analysis and Sensing Techniques
- Metabolomics and Mass Spectrometry Studies
- Semantic Web and Ontologies
- Optimal Experimental Design Methods
- Biomedical Text Mining and Ontologies
- Diet and metabolism studies
- Coral and Marine Ecosystems Studies
- Pregnancy and preeclampsia studies
- Fungal Plant Pathogen Control
- Multi-Criteria Decision Making
- Air Quality and Health Impacts
- Statistical Methods and Inference
Institut Agro Rennes-Angers
2022-2025
Université d'Angers
2022-2025
Institut National de Recherche pour l'Agriculture, l'Alimentation et l'Environnement
2023-2025
L'Institut Agro
2024
Laboratoire Angevin de Recherche en Mathématiques
2023
Institut de recherche en horticulture et semences
2021-2022
Laboratoire de Mathématiques Jean Leray
2019
Maison des Sciences de l'Homme Lyon St-Étienne
2019
Maison des Sciences de l'Homme
2018
Institut National de la Recherche Agronomique du Niger
2015-2017
The use of biopesticides represents an alternative strategy to synthetic chemical products for crop protection. To promote their adoption and effective by growers, it is crucial understand modes action the optimal conditions application in crops, including compatibility with specific varieties. Through a series greenhouse experiments, this study describes development validation robust molecular diagnostic tool enabling evaluation defence gene activation. results identified plant resistance...
ComDim analysis was designed to assess the relationships between individuals and variables within a multiblock setting where several variables, organized in blocks, are measured on same individuals. An overview of this method is presented together with some its properties. Furthermore, we discuss new extension case ( K+1 ) datasets. More precisely, aim explore response dataset K other illustration latter strategy basis study involving Time Domain ‐ Nuclear Magnetic Resonance data outlined...
In the era of machine learning-driven plant imaging, production annotated datasets is a very important contribution. this data paper, unique dataset seedling emergence kinetics proposed. It composed almost 70,000 RGB-depth frames and more than 700,000 annotations. The shown valuable for training deep learning models performing high-throughput phenotyping by imaging. ability such to generalize several species outperform state-of-the-art owing delivered demonstrated. We also discuss how raises...
A new method for the analysis of a multivariate dataset depending on several factors is proposed. It called AoV-PLS (Analysis Variance-PLS). based decomposition into main effects, interactions effects and possibly residual matrix using model akin to variance (ANOVA). Each effect considered in turn assessed through use Partial Least Square regression (PLS-regression). The connection competing methods such as ANOVA-PCA ANOVA-Simultaneous Component Analysis (ASCA) emphasized these are compared...
We propose a mathematical study of the statistics chlorophyll fluorescence indices. While most literature assumes Gaussian distributions for these indices, we demonstrate their fundamental non-Gaussian nature. Indeed, while noise in raw images can be assumed as additive, deterministic ratio between them produces nonlinear distributions. investigate states which this non-Gaussianity affect statistical estimation when wrongly approached with linear estimators. provide an...
We describe Growth Data, a software for seedling growth analysis from time lapse acquired in top view via RGB-Depth low-cost sensors. The is suited the observation of young plants during their early developmental stages. Data allows discrete detection dicotyledon stages and produces continuous curves any type seedlings. As companion paper to recent articles providing methodological innovation behind these algorithms, this article describes use GUI interface provided high-throughput general...
Tropospheric ozone (O3) is one of the pollutants that have a significant impact on human health. It can increase rate asthma crises, cause permanent lung infections and death. Predicting its concentration levels therefore important for planning atmospheric protection strategies. The aim this study to predict daily mean O3 day ahead in Grand Casablanca area Morocco using primary meteorological variables. Since available explanatory variables are multicollinear, multiple linear regressions...
Thanks to the wider spread of high-throughput experimental techniques, biologists are accumulating large amounts datasets which often mix quantitative and qualitative variables not always complete, in particular when they regard phenotypic traits. In order get a first insight into these reduce data matrices size scientists rely on multivariate analysis techniques. However such approaches easily practicable faced with mixed datasets. Moreover displaying numbers individuals leads cluttered...
In this paper, we propose a new method for parameter estimation of the probability density function photosystem II (PSII) index in chlorophyll fluorescence imaging. The PSII is modeled as ratio two normal distributions. proposed based on hierarchical Bayesian modeling, and mean field variational Bayes performed to approximate inference. approach evaluated using data acquired Arabidopsis thaliana. optimal posterior distributions are computed then used estimate parameters. preliminary results...
In order to circumvent the effects of multicollinearity on quality a multiple linear regression, new strategy analysis is proposed. It based biased estimation vector coefficients. Properties this approach are shown. Moreover, link between and existing strategies discussed, particularly Ridge Generalized regression. Illustrations basis two datasets also outlined outcomes compared those
Abstract BackgroundThanks to the wider spread of high-throughput experimental techniques, biologists are accumulating large amounts datasets which often mix quantitative and qualitative variables not always complete, in particular when they regard phenotypic traits. In order get a first insight into these reduce data matrices size scientists rely on multivariate analyses. However such approaches easily practicable faced with mixed missing values. Moreover displaying numbers individuals leads...