In November 2020 we contributed to the online fall school AMERSIE on Advanced methods for remote sensing information extraction, jointly organized by the IEEE Geoscience and Remote Sensing Society chapters of Norway and Spain, within the framework of the ChapNet initiative. The school aimed at introducing to advanced signal processing methods for information retrieval in large scale datasets collected by multiple remote sensors.
The new developments in remote sensing technologies are moving towards smaller, specialized satellite systems, constellations involving many platforms, and exploitation of multi-sensor (multi-modal) observations. Together with the ever-increasing capabilities for higher spatial, spectral, temporal, and radiometric resolution, this results in an explosive growth in the amount of RS data available for Earth science applications. The proliferation of remote sensing data also increases the complexity of these data, like larger diversity and higher dimensionality. This development goes hand in hand with the rise of new, innovative signal processing methodologies to take advantage of the new monitoring capabilities and opportunities.
The school had the following programme:
- High performance computing for hyperspectral remote sensing information extraction by Antonio Plaza
- Extracting physical information in multichannel radar remote sensing by Carlos López-Martínez
- Applied mathematics for SAR and PolSAR information extraction by Avik Bhattacharya
- Statistical information theory for remote sensing data analysis & Best practice guidelines for publishing a scientific paper by Alejandro Frery
- Sensing soil and vegetation water with space-borne microwave radiometers by Maria Piles
- Challenges and opportunities in multimodal RS information extraction by Andrea Marinoni
The schools finished with a panel debato on the role of artificial intelligence on remote sensing data exploitation.
The school material is completely available in the IEEE Geoscience and Remote Sensing Society YouTube channel.