Quantum classifiers for remote sensing

Johann Maximilian Zollner

Abstract: Hybrid classic-quantum systems utilized existing quantum hardware for machine learning (ML) by running pre- and post-processing on classic hardware to overcome the limitations of today's quantum computers. In this work, hybrid systems with several pre-processing techniques and two circuit architectures are evaluated by classifying remote sensing (RS) imagery. The potential of quantum machine learning (QML) for RS is investigated and particularly autoencoder methods are found to be suitable for pre-processing. The code is published in an open repository: https://github.com/tumbgd/qc4rs.