Autonomous image analysis to accelerate energy materials discovery and integration

Image: Armin G., IEK-13, Forschungszentrum Juelich GmbH

UTILE – that’s the short version of this project, which stands for “aUTonomous Image anaLysis to accelerate the discovery and integration of energy matErials.” But the Latin word “utile” also means useful or usable. And indeed, there is hardly anything more useful at present than a tool capable of advancing and accelerating the characterization, development and application of clean energy sources.

That, in a nutshell, is what Dr. Kourosh Malek, Department Head of Artificial Materials Intelligence at Forschungszentrum Jülich (FZJ), IEK-13, and his team intend to do with their project: to develop a data platform for energy materials that automates and thus accelerates and improves the collection, analysis and management of valuable image data from diverse characterization methods across the scientific community using artificial intelligence.

So in the future, there will no longer be employees bending over the images of fuel cells taken by X-ray microtomography, for example, but: finely adjusted algorithms. “You could also say: we train machines in such a way that they deliver faster, more accurate and more reproducible analyses than humans, even those who are very experienced in this field,” says Dr. Kourosh Malek.

To do this, Malek and his team enrich their databases with large amounts of annotated image data and associated metadata, while integrating the intuition of the human gaze: “We need to understand how a lab technician would analyze the images, how they’re thinking, what they’re looking for,” Malek says. This cognitive knowledge combined with advances in cloud computing, will be translated into pre-trained algorithms that can use deep learning to continually draw new conclusions from both existing data and data that will be added in the future.

In the end, this platform will provide the entire energy materials scientific community, particularly those within hydrogen and batteries, with an extremely powerful, highly automated AI-enabled analysis tools. That alone will relieve scientists of countless hours of repetitive analysis work – time that they can in turn invest in researching clean energy sources with human capabilities. After all, the entire process contributes to much needed task of accelerating the development of green energy materials.


Hammar, J., Grünberg, I., Kokelj, S. V., van der Sluijs, J., & Boike, J. (2023). Snow accumulation, albedo and melt patterns following road construction on permafrost, Inuvik–Tuktoyaktuk Highway, Canada. The Cryosphere, 17(12), 5357–5372. https://doi.org/10.5194/tc-17-5357-2023
Saporta, P., Alonso-Gonzalez, A., & Hajnsek, I. (2023, February). Assessing Polarimetric SAR Interferometry coherence region parameters over a permafrost landscape. POLinSAR Online Abstracts. ESA POLinSAR Workshop, Toulouse, France. https://elib.dlr.de/194470/
Saporta, P., Alonso-González, A., & Hajnsek, I. (2023). Investigating signal penetration depth in Synthetic Aperture Radar imagery over lowland permafrost. In J. M. Fernández-Fernández, J. Bonsoms, J. García-Oteyza, & M. Oliva (Eds.), Book of abstracts (p. 246). https://doi.org/10.52381/EUCOP6.abstracts.1
Hammar, J., Grünberg, I., Hendricks, S., Jutila, A., Helm, V., & Boike, J. (2023). Snow covered digital elevation model and snow depth product (2019), Trail Valley Creek, NWT, Canada. PANGAEA. https://doi.org/10.1594/PANGAEA.962552
Nill, L., Grünberg, I., Ullmann, T., Gessner, M., Boike, J., & Hostert, P. (2022). Arctic shrub expansion revealed by Landsat-derived multitemporal vegetation cover fractions in the Western Canadian Arctic. Remote Sensing of Environment, 281, 113228. https://doi.org/10.1016/j.rse.2022.113228
Saporta, P., Hajnsek, I., & Alonso-Gonzalez, A. (2022). A temporal assessment of fully polarimetric multifrequency SAR observations over the Canadian permafrost. Proceedings of the European Conference on Synthetic Aperture Radar, EUSAR. European Conference on Synthetic Aperture Radar, EUSAR, Leipzig, Germany. https://elib.dlr.de/186412/
Lange, S., Grünberg, I., Anders, K., Hartmann, J., Helm, V., & Boike, J. (2021). Airborne Laser Scanning (ALS) Point Clouds of Trail Valley Creek, NWT, Canada (2018). https://doi.org/10.1594/PANGAEA.934387
Gonstalla, E., Grünberg, I., & Boike, J. (2021). Das Eisbuch - Alles, was man wissen muss, in 50 Grafiken. oekom Verlag.

Other projects


Hyper 3D-AI

Artificial Intelligence for 3D multimodal point cloud classification

The aim is to develop an artificial intelligence that can achieve the fusion of two-dimensional data with three-dimensional information. Based on this, the software would simultaneously be able to recognise image characteristics as well as the spatial relationships between different objects.
Image: SmartPhase project partners


Fast, intelligent 3-D X-ray images for material examination

In order to be able to improve materials, it helps to take a look at their microstructures. This is because valuable information about their properties and behaviour can be found there – for example information about when which ageing processes begin. The aim of this project is to automate and accelerate access to this information with the help of a smart imaging technique.

HIT Permafrost

The Hidden Image of Thawing Permafrost

The project aims to develop a method for determining just how extensively thaw processes have already progressed in permafrost regions. The machine learning approach to be developed will be used to analyse radar images from aircraft in order to learn more about the properties of the subsurface permafrost.