Abstract | ||
---|---|---|
Ukraine's big problem is the disappearance of forest cover. According to the international forest monitoring project Global Forest Watch, Ukraine lost 1.08Mha of forests from 2000 to 2020. Such sad statistics are possible only due to the lack of monitoring tools for the forest industry in Ukraine. Such a tool can be created by combining Remote Sensing and Deep Learning approaches. To implement such approach for the automatic use, we combine Optical and Synthetic Aperture Radar images of the Sentinel-2 and Sentinel-1 satellite missions on which object-detection is performed using a U-Net-based neural network trained with use of the semi-supervised learning technique. This approach is being tested and shows its effectiveness in Kyiv region and going to be implemented in the same way for the Lviv, Odessa and Zakarpatya oblasts. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/IDAACS53288.2021.9661008 | 2021 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS) |
Keywords | DocType | Volume |
Deep Learning,U-Net,Remote Sensing,deforestation,object detection | Conference | 1 |
ISSN | ISBN | Citations |
2770-4262 | 978-1-6654-2606-0 | 0 |
PageRank | References | Authors |
0.34 | 11 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Leonid Shumilo | 1 | 0 | 0.34 |
Mykola Lavreniuk | 2 | 0 | 0.34 |
Nataliia Kussul | 3 | 191 | 25.01 |
Bella Shevchuk | 4 | 0 | 0.34 |