Header image

Forest big data, deep learning - Session 2

Tracks
Breakout 2
Wednesday, September 11, 2024
12:00 PM - 1:00 PM
WSP Room

Speaker

Mr John Brandt
Senior Data Scientist
World Resources Institute

A Sub-meter Canopy Height map of the Earth generated by a Foundational Vision Transformer

12:00 PM - 12:15 PM

Abstract

Biography

John Brandt is a Senior Data Science Associate n the Restoration team at the World Resources Institute. He works on developing novel datasets to monitor forest and landscape restoration activities with remote sensing. His teams work analyzes petabytes of optical and radar satellite imagery with computer vision and deep learning approaches to map large-scale tree cover change.
Dr Qian Song
Technical University of Munich

Dominant Leaf Type Classification using Sentinel-2 time series

12:15 PM - 12:30 PM

Abstract

Biography

She received the B.E. degree from the School of Information Science and Technology, East China Normal University, Shanghai, China, in 2015, and the Ph.D. degree (Hons.) from Fudan University, Shanghai, China, in 2020. From 2020 to 2022, she was a post-doctoral fellow with the Remote Sensing Technology Institute (IMF), German Aerospace Center (DLR), Wessling, Germany. Since 2023, she is a post-doctoral fellow with the Data Science in Earth Observation, Technical University of Munich (TUM), Germany. She was awarded as the URSI (International Union of Radio Science) Young Scientist Award in 2020. Her research interests include advanced deep learning technologies and their applications in synthetic aperture radar image interpretation, and forest monitoring.
Dr Fabian Döweler
Geospatial Ecologist
Dragonfly Data Science

Using Machine Learning to Upscale ALS derived Forest Structural information

12:30 PM - 12:45 PM

Abstract

Biography

Fabian is a geospatial analyst and ecologist focused on studying large-scale changes in ecosystems using advanced imaging, machine learning, and mapping techniques. He earned his PhD from Auckland University of Technology, where his research investigated the abrupt treeline patterns in the Southern Alps using drone imagery, plant physiology, and agent-based modelling. At Dragonfly Data Science, Fabian is using remote sensing and machine learning applications to study ecological changes. His work includes developing deep learning solutions for native bird song annotation or qualitative and quantitative assessments of terrestrial landscapes. Fabian is committed to the use of open source solutions, enhancing accessibility and innovation in ecological research.
Mr Kevan Cote
Data Scientist
Eco-index

Multiband Sandwiches - Using Multispectral Imagery with Neural Networks to find Kahikatea trees across New Zealand

12:45 PM - 1:00 PM

Abstract

Biography

Kevan Cote is a contractor to the Eco-index programme focusing on the development of ecosystem detectors by combining remote sensing data with artificial intelligence. Kevan has a cross-disciplinary background in production data analysis and product development engineering. He has worked on projects ranging from using AI to assist in animal welfare measurement automation to measuring indices for human well-being remotely as well as developing an AI framework for environmental monitoring for an indigenous audience. His diverse expertise promotes agility and a fresh perspective on practical application of AI techniques.
Dr Jonathan Batchelor
Postdoctoral Scholar
University Of Washington

Chairperson

Biography

Ms Pratima Khatri-Chhetri
Graduate Student
University of Washington

Chairperson

Biography

loading