Scope and topics
In recent years, satellite hyperspectral remote sensing is witnessing a boost in algorithms
development and demonstration of innovative application products, thanks to the increased data
flow and image tasking capabilities offered by the new generation of Earth Observation (EO)
missions. While in the past data analysis was mostly limited to few and temporally discontinuous
hyperspectral images, nowadays researchers can access longer and denser time series.
In this respect, institutional satellites such as Italian Space Agency (ASI)’s PRISMA and German
Aerospace Center (DLR)’s EnMAP have opened a new era, enabling the collection of more
consistent time series than it was possible before. While they remain precursor and scientific
missions, PRISMA and EnMAP are significantly helping the user community explore what
mapping missions such as European Space Agency (ESA)’s CHIME will be able to deliver in the
future. The lessons learnt are also supporting the development of new missions – e.g. PRISMA
Second Generation and the PLATiNO Hyperspectral of the IRIDE system – that are expected to
upgrade performances with respect to spatial, spectral and temporal resolution. Additionally, new
players from the space economy are investing in hyperspectral satellites, thus proving an increasing
interest also in the commercial sector.
This special session therefore aims to present state-of-the-art methods, algorithms, prototype
products and use-case demonstrations developed through satellite hyperspectral time series analysis
for a wide range of environmental applications. In addition to more consolidated fields such as
agriculture, mineralogical and geological mapping, the session encourages submissions from other
branches of environmental remote sensing where hyperspectral technologies, and in particular
multi-temporal analysis, are opening new opportunities. Examples include, but are not limited to:
water quality, air quality and pollution, emissions monitoring, forestry and mangrove mapping,
urban areas classification, green infrastructure, cultural heritage, landscape archaeology, disaster
risk management (e.g. fire damage and active fires).
Furthermore, examples of multi-modal data-based approaches leveraging images collected by
different EO missions (hyperspectral, multispectral or SAR), as well as integration with ground
truth data, are welcome.
By bringing together experts with complementary technical expertise, the session provides a forum
to discuss best practices and successful implementation, limitations of either existing satellites or
analytical approaches, and current bottlenecks, in order to identify recommendations for future
missions and outline new research avenues.
List of topics:
- Advanced algorithms and methods for satellite hyperspectral data processing, including machine learning and deep learning.
- Workflows for time series analysis for environmental monitoring.
- Multi-temporal monitoring of surface properties in natural and anthropogenic environments.
- Multi-modal data-based approaches from different satellite hyperspectral missions or
- integration with other missions.
- Challenges and solutions in combining different spaceborne spectrometers data.
- Protocols for calibration and validation.
- Open and free benchmark datasets and processing tools.
- Case studies showcasing practical societal/economic/environmental benefits from the use of
- satellite hyperspectral data.
- Lessons learnt and recommendations, as well as potential user requirements, for future missions design, applications development, and services provisions.
Organizers
Dr. Deodato Tapete, Senior Researcher, Italian Space Agency (ASI), Earth Observation Office,
Rome, Italy, deodato.tapete@asi.it
Dr. Daniele Cerra, Senior Researcher, German Aerospace Center (DLR), Earth Observation
Center (EOC), Remote Sensing Technology Institute, Oberpfaffenhofen, Germany,
Daniele.Cerra@dlr.de