English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The abundance of gaps in satellite image time series often complicates theapplication of deep learning models such as convolutional neural networks forspatiotemporal modeling. Based on previous work in computer vision on imageinpainting, this paper shows how three-dimensional spatiotemporal partialconvolutions can be used as layers in neural networks to fill gaps in satelliteimage time series. To evaluate the approach, we apply a U-Net-like model onincomplete image time series of quasi-global carbon monoxide observations fromthe Sentinel-5P satellite. Prediction errors were comparable to two consideredstatistical approaches while computation times for predictions were up to threeorders of magnitude faster, making the approach applicable to process largeamounts of satellite data. Partial convolutions can be added as layers to othertypes of neural networks, making it relatively easy to integrate with existingdeep learning models. However, the approach does not quantify prediction errorsand further research is needed to understand and improve model transferability.The implementation of spatiotemporal partial convolutions and the U-Net-likemodel is available as open-source software.

Efficient data-driven gap filling of satellite image time series using deep neural networks with partial convolutions