Satellites provide huge amounts of images, the number of datasets increases and the domain became more and more data-driven. Image quantization and embedding into file formats do support convolutional neural networks to estimate the labels of images, by strengthening the features.

Project to apply and benchmark deep learning models on FPGAs in the field of remote sensing using TensorFlow and VitisAI. This project is funded by the European Space Agency (ESA) and is a collaboration between the Technical University of Munich, Airbus, and OroraTech.

Deep learning can be deployed to a wide range of hardware products, such as CPUs or GPUs using libraries like TensorFlow. In order to deploy convolutional neural network to FPGAs, it is important that the framework of the fpga does support all used layers.

In this project, we created a novel approach for trajectory similarity based on Kolmogorov complexity approximated by a lossy compression of the original trajectory data using selected features compressed into a concise memory representation by means of a Bloom filter.

In this project, we created a novel approach for trajectory similarity based on Kolmogorov complexity approximated by a lossy compression of the original trajectory data using selected features compressed into a concise memory representation by means of a Bloom filter.