TITLE: Use of Sentinel-2 satellite data in agriculture, a perspective on using deep learning for land use
SPEAKER: Dr. Etienne Lord, Researcher / ScientistDigital Agronomy, AAFC/AFC

LOCATION: Zoom (link available on request, please contact Dr. Francois Tardif )
TIME: 3:00 to 4:00 PM
DATE: Thursday, March 18th,2021

Summary: The use of A.I. in agriculture for the classification of land use and land cover is still an area of active development with global implication. This seminar shall cover some differences between different deep learning paradigms such as image segmentation, image classification and distinct object identification in a landscape. We will first explore the basis of creating a deep learning model with a practical example and the use of convolutional neural networks (CNN) and the Keras API. Second, an introduction to real time distinct classes identification for drones will be discussed. Finally, we will look at some new tools facilitating the use of deep learning in land cover and land use classification.

Biography: Dr Etienne Lord’s work addresses microbiome identification, classification and evolution as well as the identification of biological network hubs. He also applied deep learning on multi-omics and drone imaging analysis to tackle complex environments evolution and sustainability. He did his postdoctoral training in machine learning and biology at the University of Montreal (UdeM) and later at Laval University in Quebec city (U. Laval). He currently works as a research scientist in digital agronomy within the Science and Technology Branch of Agriculture and Agri-Food Canada.