Inpainting radar missing data regions with deep learning

Geiss, Andrew; Hardin, Joseph C.

doi:https://doi.org/10.5194/amt-14-7729-2021

Articles | Volume 14, issue 12

https://doi.org/10.5194/amt-14-7729-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-14-7729-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 14, issue 12

Research article

|

09 Dec 2021

Research article |

| 09 Dec 2021

Inpainting radar missing data regions with deep learning

Andrew Geiss and Joseph C. Hardin

Supplement

https://doi.org/10.5194/amt-14-7729-2021-supplement

Data sets

Ka ARM Zenith Radar (KAZRCFRGEQC) J. Hardin, A. Hunzinger, E. Schuman, A. Matthews, N. Bharadwaj, A. Varble, K. Johnson, and S. Giangrande https://doi.org/10.5439/1615726

Model code and software

avgeiss/radar_inpainting: AMT Supplementary Code Andrew Geiss and Joseph C. Hardin https://doi.org/10.5281/zenodo.5643624

Additional Inpainting Examples Andrew Geiss and Joseph C. Hardin https://doi.org/10.5281/zenodo.5744857

Short summary

Radars can suffer from missing or poor-quality data regions for several reasons: beam blockage, instrument failure, and near-ground blind zones, etc. Here, we demonstrate how deep convolutional neural networks can be used for filling in radar-missing data regions and that they can significantly outperform conventional approaches in terms of realism and accuracy.