Dealing with missing modalities at test time for land cover mapping: A case study on multi-source optical data
- 14 June 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. abs 1806 1, 621-626
Abstract
In recent years, multiple sources of remote sensing data have become increasingly available to monitor Earth’s surface phenomena. However, unlike High Spatial Resolution (HSR) data, Very High Spatial Resolution (VHSR) satellite images remain difficult to collect over large areas due to acquisition costs and a smaller swath. This often compromises the simultaneous use of both sources of data over same study areas for many applications. In this work, we investigate a land cover mapping setting in which both HSR and VHSR are available at the learning stage of a deep neural network while only the HSR data is available at inference time for model inference. We thus propose simple but effective strategies for enhancing the land cover classification in this scenario of incomplete multi-source remote sensing data when the model is deployed.Keywords
This publication has 4 references indexed in Scilit:
- Multisensor Land Cover Classification With Sparsely Annotated Data Based on Convolutional Neural Networks and Self-DistillationIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021
- Knowledge Distillation: A SurveyInternational Journal of Computer Vision, 2021
- Operational High Resolution Land Cover Map Production at the Country Scale Using Satellite Image Time SeriesRemote Sensing, 2017
- ESA's sentinel missions in support of Earth system scienceRemote Sensing of Environment, 2012