Polina Lemenkova
From
In residence at
French Geological Survey (BRGM) - FR
Host scientists
François Tertre (BRGM) & Frédéric Ros ((PRISME – Université d’Orléans/INSA-CVL)
BIOGRAPHY
Polina Lemenkova is a data analyst in Earth sciences. Her research focuses on developing algorithms of automating workflows for extraction, interpretation, and analysis of geospatial datasets. Her expertise includes Earth system modeling, satellite image classification using machine learning, and cartographic data handling for geological applications. In her research, she utilizes methods of artificial intelligence and deep learning for remote sensing data processing. She has worked on environmental monitoring and mapping using her strong technical skills in scripting languages and applied programming (Python, R, GMT, GRASS GIS).
PROJECT
Development of digital services to improve environmental monitoring through effective management of natural resources
The joint research project involving partners of Bureau de Recherches Géologiques et Minières (BRGM), l'Institut National de Recherche pour l'Agriculture, l'Alimentation et l'Environnement (INRAE), le Centre national de la recherche scientifique (CNRS), and the Universities of Orléans and Tours focuses on integrating environmental and digital research within the regional innovation strategy. This collaboration is designed to provide valuable data and services to a diverse array of socio-economic partners, thereby enhancing regional development through targeted research initiatives. The project aims to complete the components of the developments conducted over the past two years in collaboration with three primary research teams: the PRISME research team (Laboratoire pluridisciplinaire de recherche en ingénierie des systèmes, mécanique et énergétique), which focuses on image and signal processing, robotics and automation, within the University of Orléans, the LIFAT research team specializing in Fundamental and Applied Computer Science at the University of Tours, and the BRGM teams that handle data and environmental studies. The primary objective of this project is to implement the development of geospatial datasets using Python programming language. The project also focuses on data integration and the application of machine learning techniques to process heterogeneous geospatial data. The project comprises three levels of clusters: Infrastructure focusing on data integration; contributions to development of algorithms and machine learning protocols; improvement and updating of intelligent interfaces allowing users to interact with digital twins. The project entails developing digital twins of geospatial processes by providing real-time, dynamic models in construction for operational reformulation, developing digital twins, and engaging in computing advancements related to these developments. The overall objective of the project is to develop digital twins for geospatial analysis and IOT systems.
