Abstract
Spatiooral data sets such as satellite image series are of utmost importance for understanding global developments like climate change or urbanization. However, incompleteness of data can greatly impact usability and knowledge discovery. In fact, there are many cases where not a single data point in the set is fully observed. For filling gaps, we introduce a novel approach that utilizes Markov random fields (MRFs). We extend the probabilistic framework to also consider empirical prior information, which allows to train even on highly incomplete data. Moreover, we devise a way to make discrete MRFs predict continuous values via state superposition. Experiments on real-world remote sensing imagery suffering from cloud cover show that the proposed approach outperforms state-of-the-art gap filling techniques.
Original language | English |
---|---|
Title of host publication | Proceedings - 2020 IEEE 7th International Conference on Data Science and Advanced Analytics, DSAA 2020 |
Editors | Geoff Webb, Zhongfei Zhang, Vincent S. Tseng, Graham Williams, Michalis Vlachos, Longbing Cao |
Place of Publication | Piscataway NJ USA |
Publisher | IEEE, Institute of Electrical and Electronics Engineers |
Pages | 546-555 |
Number of pages | 10 |
ISBN (Electronic) | 9781728182063 |
ISBN (Print) | 9781728182070 |
DOIs | |
Publication status | Published - 2020 |
Event | IEEE International Conference on Data Science and Advanced Analytics 2020 - Sydney, Australia Duration: 6 Oct 2020 → 9 Oct 2020 Conference number: 7th https://ieeexplore-ieee-org.ezproxy.lib.monash.edu.au/xpl/conhome/9259989/proceeding (Publication) http://dsaa2020.dsaa.co (Website) |
Conference
Conference | IEEE International Conference on Data Science and Advanced Analytics 2020 |
---|---|
Abbreviated title | DSAA 2020 |
Country/Territory | Australia |
City | Sydney |
Period | 6/10/20 → 9/10/20 |
Internet address |
Keywords
- Gap filling
- Graphical models
- Probabilistic machine learning
- Remote sensing
- Spatiooral