Seminar: Deep Learning-Based Fusion of Optical and Sar Imagery for Canopy Height Mapping in Brazil

Join us at 10 a.m. on Wednesday, Aug. 12 in-person or on Zoom for this seminar with Matheus Pinheiro Ferreira on forest monitoring and applied remote sensing projects in Brazil.

This talk presents a deep learning framework that fuses Sentinel-2 optical imagery with C-band Sentinel-1 and L-band ALOS PALSAR data to map forest canopy height across São Paulo state, Brazil, using a comprehensive 2024 statewide airborne LiDAR campaign. 

The presentation will detail the multi-sensor fusion architecture, designed to preserve the native resolution of each data source, and discuss comparative experiments involving dry- and rainy-season composites.

About the Speaker

Matheus Pinheiro Ferreira graduated in forestry engineering from the Federal University of Paraná (UFPR) in 2010, with additional training at the University of Freiburg, Germany. He obtained his master's degree in remote sensing in 2012 and his Ph.D. in 2017 from the National Institute for Space Research (INPE). His current research focuses on monitoring and quantifying forest resources through remote sensing and forest inventories. 

Over the past years, he has been dedicated to developing artificial intelligence methods for mapping tree species, quantifying floristic diversity, and monitoring carbon stocks and sequestration in tropical forests using data obtained from active and passive remote sensors. He has expertise in hyperspectral remote sensing and radiative transfer modeling. 

From 2018 to 2024, he served as an adjunct professor in the Cartographic Engineering Section at the Military Institute of Engineering (IME). Currently, he is a professor at the University of São Paulo (USP), affiliated with the Department of Forest Sciences at the 'Luiz de Queiroz' College of Agriculture (ESALQ).

Location: 4600 River Road, room 325 or Zoom. 

For the Zoom link, please visit the GEOG Department Calendar.