Institut national de la recherche scientifique, Centre Eau Terre Environnement,Québec, CANADA
Deadline: No particular deadline
Expected starting date: As soon as possible
This Ph.D. project aims to develop a platform for land covers mapping and monitoring at the regional and national scales from multitemporal and multisource data, e.g. optical and radar sensors, and using the state-of-the-art approaches based on AI/ML and Object-based Image Analysis (OBIA). The large-scale analysis aims to fuse data from different sensor types and apply various processing to these data in different scales benefiting from cloud computing. Applying existing methods, optimizing them, and designing new algorithms would be part of this platform. Specifically, with the rise of deep learning and its effect on large-scale data processing, designing, optimizing, deploying, and troubleshooting deep learning network would be considered as one of the research themes for this project.
As part of this research project, several potential thematic applications have been considered through collaboration with federal, provincial, and commercial partners. The main areas in which the candidate, based on his/her background, can work are:
- Croplands’ mapping and monitoring
- Wetlands’ mapping and monitoring
- Forest tree species’ mapping
- Evaluate the state-of-the-art methods for classification, detection, and fusion in the cropland/wetland/forest tree species mapping
- Develop, optimize, and deploy ML/AI algorithm on remote sensing/GIS data
- Proposing new idea for solving the research gaps in these fields
- Writing scientific paper, reports and presenting them to the remote sensing community.
The Ph.D. candidate will work within the research Group of Earth Observations Analytics by Artificial Intelligence (GEO-AI) at the Centre Eau Terre Environment of INRS.