Deep learning models based on diffusion processes have shown great potential in a range of generative tasks, such as image generation. For remote sensing applications, generative models are not that common. The question that we tried to answer is whether diffusion processes can be used to efficiently pretrain models for discriminative tasks?

Multimodal datasets for remote sensing are oftentimes limited to two data modalities, such as multispectral and SAR polarization data. In order to experiment with a much wider range of data modalities, we extended the well-known BigEarthNet dataset to includes a wide range of data modalities.

Road traffic noise is a global issue that can lead to severe health effects. Despite the ubiquity of traffic noise, its quantification or estimation is complicated and detailed road traffic maps are only available for select countries or areas. We investigate whether it is possible to train a segmentation model to esimate road traffic noise from satellite imagery.

Self-supervised learning provides a powerful means to pretrain models based on un-labeled data. Un-labeled Earth observation data are abundant: this circumstance combined with the availability of multi modal data makes Earth observation a perfect playground for self-supervised learning. Our early results are very promising…

Tracking Greenhouse Gas emissions will be increasingly important in the future, as power generation from fossil fuels is supposed to fade out. Independent tools to monitor power plants are required for this task. We present a method that uses freely available satellite imagery to estimate power generation and CO2 emission rates on a global scale.

Air pollution is a major health issue and often also contributes to climate change. Measuring air pollution is costly and therefore only available in some countries. We investigated whether freely available Earth observation data can be utilized to estimate air pollution on the surface level.

Can we identify trucks from space and estimate truck traffic rates anywhere on the planet? Yes, we can!

We developed a deep learning model that is able to distinguish between 10 different types of power plants in an effort to automatically identify and characterize industrial sites in satellite imagery. This work will help us to estimate greenhouse gas emission rates for individual industrial sites in the future.

Greenhouse gas emissions from the industrial economic sector are a major driver of the currently observed climate change. We developed a deep learning approach to identify and characterize industrial smoke plumes. In the future, we will utilize this approach to estimate greenhouse gas emissions from remote sensing data on a global scale.

Don Quixote has been considered asteroidal since its discovery in 1983, despite its comet-like orbit. We found activity in this object during its 2009 and 2018 apparitions, leading us to believe that it really is an active comet. But are there other objects like Don Quixote? I monitored a sample of ~100 asteroids that are somewhat likely to turn active, too, over a period of more than 4 years. Bottom line: Don Quixote is rather unique.

This is a toy project that turned into a real research project and a preparation for my new job as research scientist in computer vision: using machine learning techniques to identify clouds in all-sky camera images.

Based on Spitzer Space Telescope observations, we derived diameter and albedo estimates for 2132 asteroids in near-Earth space - the largest dataset of its kind. Our results and data are available online.

TESS is a satellite that observes bright stars to find brightness modulations revealing exoplanet transits - but it also observes a lot of asteroids over a long period of time, which makes it a unique asset for deriving long asteroid rotation periods.

The Gaia mission builds a unique survey of the stars in our Milky Way - but it also observes asteroids that are crossing its field of view. We derive ensemble shape information for different asteroid populations from the first batch of asteroid data from Gaia.

There is a range of asteroid thermal models that are based on different physical assumptions. In this analysis, I investigate how these assumptions affect their results and applicability to different situations.

sbpy is an astropy affiliated package for small-body planetary astronomers. We proposed this idea to NASA and were funded for developing this Python module, which is outlined here.

photometrypipeline is a Python software package for automated image registration, calibration, and extraction of photometry tailored to the needs of asteroid observer. However, pp can also be applied to other imaging observations.

Dead comets are small bodies that appear as inactive asteroids, but have a cometary origin. Some of these objects might still harbor ices and can still activate and appear comet-like. Since these objects might signifcantly contribute to the volatile reservoir of the asteroid population, it is important to understand how many of these objects there are.

With rapid-response observations using UKIRT, we are able to observe newly discovered near-Earth asteroids when they are still close to Earth and thus bright. Our observations enable a probabilistic taxonomic classification of asteroids with typically small sizes and the determination of their compositional distribution.

We observed two very small asteroids with the Spitzer Space Telescope to actually measure their physical properties for the first time. The constraints that we were able to place on their properties do not agree with the standard picture of small asteroids.

Near-Earth asteroid Don Quixote has long since been considered a good candidate for an inactive comet nucleus due to its comet-like orbit. Since its discovery in 1983, comet-like has never been observed - until we observed this object with the Spitzer Space Telescope…

My first research paper. Using Herschel Space Observatory observations, we investigated the physical properties of 18 Plutinos - small bodies at the outskirts of the Solar System that have orbital properties similar to those of Pluto. What we found? That Pluto is pretty unique…