Many onboard vision tasks for spacecraft navigation require high-quality remote-sensing images with clearly decipherable features. However, design constraints and the operational and environmental conditions limit their quality. Enhancing images through postprocessing is a cost-efficient solution. Current deep learning methods that enhance low-resolution images through super-resolution do not quantify network uncertainty of predictions and are trained at a single scale, which hinders practical integration in image-acquisition pipelines. This work proposes performing multiscale super-resolution using a deep normalizing flow network for uncertainty-quantified and Monte Carlo estimates so that image enhancement for spacecraft vision tasks may be more robust and predictable. The proposed network architecture outperforms state-of-the-art super-resolution models on in-orbit lunar imagery data. Simulations demonstrate its viability on task-based evaluations for landmark identification.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Multiscale Super-Resolution Remote Imaging via Deep Conditional Normalizing Flows


    Beteiligte:
    Heintz, Aneesh M. (Autor:in) / Peck, Mason (Autor:in) / Mackey, Ian (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    2023-04-28


    Format / Umfang :

    16 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Multi-Scale, Super-Resolution Remote Imaging Via Deep Conditional Normalizing Flows

    Heintz, Aneesh / Peck, Mason A. / Mackey, Ian | TIBKAT | 2022


    Multi-Scale, Super-Resolution Remote Imaging via Deep Conditional Normalizing Flows

    Heintz, Aneesh / Peck, Mason A. / Mackey, Ian | AIAA | 2022


    Inverse Design Under Uncertainty Using Conditional Normalizing Flows

    Tsilifis, Panagiotis / Ghosh, Sayan / Andreoli, Valeria | TIBKAT | 2022


    Inverse Design under Uncertainty using Conditional Normalizing Flows

    Tsilifis, Panagiotis / Ghosh, Sayan / Andreoli, Valeria | AIAA | 2022


    Inverse Design Under Uncertainty Using Conditional Normalizing Flows

    Tsilifis, Panagiotis / Ghosh, Sayan / Andreoli, Valeria | TIBKAT | 2022