Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine Learning
Habitat mapping is essential for the management and monitoring of Natura 2000 sites. Time-consuming field surveys are still the most frequently used solution for the implementation of the European Habitats Directive, but the use of remote sensing tools for this is becoming more common. The high temp...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/5/1388 |
| _version_ | 1797614447889481728 |
|---|---|
| author | Adriana Marcinkowska-Ochtyra Adrian Ochtyra Edwin Raczko Dominik Kopeć |
| author_facet | Adriana Marcinkowska-Ochtyra Adrian Ochtyra Edwin Raczko Dominik Kopeć |
| author_sort | Adriana Marcinkowska-Ochtyra |
| collection | DOAJ |
| description | Habitat mapping is essential for the management and monitoring of Natura 2000 sites. Time-consuming field surveys are still the most frequently used solution for the implementation of the European Habitats Directive, but the use of remote sensing tools for this is becoming more common. The high temporal resolution of Sentinel-2 data, registering the visible, near, and shortwave infrared ranges of the electromagnetic spectrum, makes them valuable material in this context. In this study, we aimed to use multitemporal Sentinel-2 data for mapping three grassland Natura 2000 habitats in Poland. We performed the classification based on spectro-temporal features extracted from data collected from eight different terms within the year 2017 using Convolutional Neural Networks (CNNs), and we also tested other widely used machine learning algorithms for comparison, such as Random Forests (RFs) and Support Vector Machines (SVMs). Based on ground truth data, we randomly selected training and validation polygons and then performed the evaluation iteratively (100 times). The best resulting median F1 accuracies that we obtained for habitats were as follows: 6210, 0.85; 6410, 0.80; and 6510, 0.84 (with SVM). Finally, we concluded that the accuracy of the results was comparable, but we obtained the best results using SVM (median OA = 88%, with 86% for RF and 84% for CNNs). In this work, we confirmed the usefulness of the spectral dimension of Sentinel-2 time series data for mapping grassland habitats, and researchers of future work can further develop the use of CNNs for this purpose. |
| first_indexed | 2024-03-11T07:11:29Z |
| format | Article |
| id | doaj.art-7d0a11d6066245bd9de77bb28da42cc1 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T07:11:29Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-7d0a11d6066245bd9de77bb28da42cc12023-11-17T08:32:25ZengMDPI AGRemote Sensing2072-42922023-03-01155138810.3390/rs15051388Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine LearningAdriana Marcinkowska-Ochtyra0Adrian Ochtyra1Edwin Raczko2Dominik Kopeć3Department of Geoinformatics, Cartography and Remote Sensing, Chair of Geomatics and Information Systems, Faculty of Geography and Regional Studies, University of Warsaw, 00-927 Warsaw, PolandDepartment of Geoinformatics, Cartography and Remote Sensing, Chair of Geomatics and Information Systems, Faculty of Geography and Regional Studies, University of Warsaw, 00-927 Warsaw, PolandDepartment of Geoinformatics, Cartography and Remote Sensing, Chair of Geomatics and Information Systems, Faculty of Geography and Regional Studies, University of Warsaw, 00-927 Warsaw, PolandMGGP Aero Sp. z o.o., 33-100 Tarnów, PolandHabitat mapping is essential for the management and monitoring of Natura 2000 sites. Time-consuming field surveys are still the most frequently used solution for the implementation of the European Habitats Directive, but the use of remote sensing tools for this is becoming more common. The high temporal resolution of Sentinel-2 data, registering the visible, near, and shortwave infrared ranges of the electromagnetic spectrum, makes them valuable material in this context. In this study, we aimed to use multitemporal Sentinel-2 data for mapping three grassland Natura 2000 habitats in Poland. We performed the classification based on spectro-temporal features extracted from data collected from eight different terms within the year 2017 using Convolutional Neural Networks (CNNs), and we also tested other widely used machine learning algorithms for comparison, such as Random Forests (RFs) and Support Vector Machines (SVMs). Based on ground truth data, we randomly selected training and validation polygons and then performed the evaluation iteratively (100 times). The best resulting median F1 accuracies that we obtained for habitats were as follows: 6210, 0.85; 6410, 0.80; and 6510, 0.84 (with SVM). Finally, we concluded that the accuracy of the results was comparable, but we obtained the best results using SVM (median OA = 88%, with 86% for RF and 84% for CNNs). In this work, we confirmed the usefulness of the spectral dimension of Sentinel-2 time series data for mapping grassland habitats, and researchers of future work can further develop the use of CNNs for this purpose.https://www.mdpi.com/2072-4292/15/5/1388grassland habitatmappingtime seriesmeadowsphenologyCNNs |
| spellingShingle | Adriana Marcinkowska-Ochtyra Adrian Ochtyra Edwin Raczko Dominik Kopeć Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine Learning Remote Sensing grassland habitat mapping time series meadows phenology CNNs |
| title | Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine Learning |
| title_full | Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine Learning |
| title_fullStr | Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine Learning |
| title_full_unstemmed | Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine Learning |
| title_short | Natura 2000 Grassland Habitats Mapping Based on Spectro-Temporal Dimension of Sentinel-2 Images with Machine Learning |
| title_sort | natura 2000 grassland habitats mapping based on spectro temporal dimension of sentinel 2 images with machine learning |
| topic | grassland habitat mapping time series meadows phenology CNNs |
| url | https://www.mdpi.com/2072-4292/15/5/1388 |
| work_keys_str_mv | AT adrianamarcinkowskaochtyra natura2000grasslandhabitatsmappingbasedonspectrotemporaldimensionofsentinel2imageswithmachinelearning AT adrianochtyra natura2000grasslandhabitatsmappingbasedonspectrotemporaldimensionofsentinel2imageswithmachinelearning AT edwinraczko natura2000grasslandhabitatsmappingbasedonspectrotemporaldimensionofsentinel2imageswithmachinelearning AT dominikkopec natura2000grasslandhabitatsmappingbasedonspectrotemporaldimensionofsentinel2imageswithmachinelearning |