Semi-supervised learning techniques are gaining popularity due to their capability of building models that are effective, even when scarce amounts of labeled data are available. In this paper, we present a framework and specific tasks for self-supervised pretraining of multichannel models, such as the fusion of multispectral and synthetic aperture radar images. We show that the proposed self-supervised approach is highly effective at learning features that correlate with the labels for land cover classification. This is enabled by an explicit design of pretraining tasks which promotes bridging the gaps between sensing modalities and exploiting the spectral characteristics of the input. In a semi-supervised setting, when limited labels are available, using the proposed self-supervised pretraining, followed by supervised finetuning for land cover classification with SAR and multispectral data, outperforms conventional approaches such as purely supervised learning, initialization from training on ImageNet and other recent self-supervised approaches.

The detection of faint, fast-moving objects such as space debris, in optical data is a major challenge due to their low signal-to-background ratio and short visibility time. This work addresses this issue by implementing the Stack-CNN algorithm, originally designed for offline analysis, on an FPGA-based platform to enable real-time triggering capabilities in constrained space hardware environments. The Stack-CNN combines a stacking method to enhance the signal-to-noise ratio of moving objects across multiple frames with a lightweight convolutional neural network optimized for embedded inference. The FPGA implementation was developed using a Xilinx Zynq Ultrascale+ platform and achieves low-latency, power-efficient inference compatible with CubeSat systems. Performance was evaluated using both a physics-based simulation framework and data acquired during outdoor experimental campaigns. The trigger maintains high detection efficiency for 10 cm-class targets up to 30–40 km distance and reliably detects real satellite tracks with signal levels as low as 1% above background. These results validate the feasibility of on-board real-time debris detection using embedded AI, and demonstrate the robustness of the algorithm under realistic operational conditions. The study was conducted in the context of a broader technology demonstration project, called DISCARD, aimed at increasing space situational awareness capabilities on small platforms.

In this article, we present cutting-edge machine learning-based techniques for the detection and reconstruction of meteors and space debris in the Mini-EUSO experiment, a detector installed on board of the International Space Station, and pointing toward the Earth. We base our approach on a recent technique, the STACKing method plus Convolutional Neural Network (STACK-CNN), originally developed as an online trigger in an orbiting remediation system to detect space debris. Our proposed method, the refined-STACKing method plus convolutional neural network (R-Stack-CNN), makes the STACKing method plus convolutional neural network (STACK-CNN) more robust, thanks to a random forest that learns the temporal development of these events in the camera. We prove the flexibility of our method by showing that it is sensitive to any space object that moves linearly in the field of view. First, we search small space debris, never observed by Mini-EUSO. Due to the limiting statistics, also in this case, no debris were found. However, since meteors produce signals similar to space debris but they are much more frequent, the R-Stack-CNN is adapted to identify such events while avoiding the numerous false positives of the Stack-CNN. Results from real data show that the R-Stack-CNN is able to find more meteors than a classical thresholding method and a new method of two neural networks. We also show that the method is also able to accurately reconstruct speed and direction of meteors with simulated data.

Hyperbolic neural networks are emerging as an effective technique to better capture hierarchical representations of many data types, from text to images and, recently, point clouds. In this paper, we extend our earlier work, that showed how to use regularizers in the hyperbolic space to improve performance of point cloud classification models, to the problem of part segmentation. This requires careful modeling of the hierarchical relationships between parts and whole point cloud to properly control the hyperbolic geometry of the feature space produced by the neural network. We show how the proposed method improves the performance of commonly used neural network architectures, reaching state-of-the-art performance on the part segmentation task.

The detection of faint, fast-moving objects such as space debris, in optical data is a major challenge due to their low signal-to-background ratio and short visibility time. This work addresses this issue by implementing the Stack-CNN algorithm, originally designed for offline analysis, on an FPGA-based platform to enable real-time triggering capabilities in constrained space hardware environments. The Stack-CNN combines a stacking method to enhance the signal-to-noise ratio of moving objects across multiple frames with a lightweight convolutional neural network optimized for embedded inference. The FPGA implementation was developed using a Xilinx Zynq Ultrascale+ platform and achieves low-latency, power-efficient inference compatible with CubeSat systems. Performance was evaluated using both a physics-based simulation framework and data acquired during outdoor experimental campaigns. The trigger maintains high detection efficiency for 10 cm-class targets up to 30-40 km distance and reliably detects real satellite tracks with signal levels as low as 1% above background. These results validate the feasibility of on-board real-time debris detection using embedded AI, and demonstrate the robustness of the algorithm under realistic operational conditions. The study was conducted in the context of a broader technology demonstration project, called DISCARD, aimed at increasing space situational awareness capabilities on small platforms.