Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery
Multispectral satellite imagery offers a new perspective for spatial modelling, change detection and land cover classification. The increased demand for accurate classification of geographically diverse regions led to advances in object-based methods. A novel spatiotemporal method is presented for o...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/14/6648 |
| _version_ | 1797587447978131456 |
|---|---|
| author | Domen Kavran Domen Mongus Borut Žalik Niko Lukač |
| author_facet | Domen Kavran Domen Mongus Borut Žalik Niko Lukač |
| author_sort | Domen Kavran |
| collection | DOAJ |
| description | Multispectral satellite imagery offers a new perspective for spatial modelling, change detection and land cover classification. The increased demand for accurate classification of geographically diverse regions led to advances in object-based methods. A novel spatiotemporal method is presented for object-based land cover classification of satellite imagery using a Graph Neural Network. This paper introduces innovative representation of sequential satellite images as a directed graph by connecting segmented land region through time. The method’s novel modular node classification pipeline utilises the Convolutional Neural Network as a multispectral image feature extraction network, and the Graph Neural Network as a node classification model. To evaluate the performance of the proposed method, we utilised EfficientNetV2-S for feature extraction and the GraphSAGE algorithm with Long Short-Term Memory aggregation for node classification. This innovative application on Sentinel-2 L2A imagery produced complete 4-year intermonthly land cover classification maps for two regions: Graz in Austria, and the region of Portorož, Izola and Koper in Slovenia. The regions were classified with Corine Land Cover classes. In the level 2 classification of the Graz region, the method outperformed the state-of-the-art UNet model, achieving an average F1-score of 0.841 and an accuracy of 0.831, as opposed to UNet’s 0.824 and 0.818, respectively. Similarly, the method demonstrated superior performance over UNet in both regions under the level 1 classification, which contains fewer classes. Individual classes have been classified with accuracies up to 99.17%. |
| first_indexed | 2024-03-11T00:39:05Z |
| format | Article |
| id | doaj.art-6881a2c5116a4e0aac635c5ec19fe11c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T00:39:05Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6881a2c5116a4e0aac635c5ec19fe11c2023-11-18T21:20:40ZengMDPI AGSensors1424-82202023-07-012314664810.3390/s23146648Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite ImageryDomen Kavran0Domen Mongus1Borut Žalik2Niko Lukač3Faculty of Electrical Engineering and Computer Science, University of Maribor, Koroška Cesta 46, 2000 Maribor, SloveniaFaculty of Electrical Engineering and Computer Science, University of Maribor, Koroška Cesta 46, 2000 Maribor, SloveniaFaculty of Electrical Engineering and Computer Science, University of Maribor, Koroška Cesta 46, 2000 Maribor, SloveniaFaculty of Electrical Engineering and Computer Science, University of Maribor, Koroška Cesta 46, 2000 Maribor, SloveniaMultispectral satellite imagery offers a new perspective for spatial modelling, change detection and land cover classification. The increased demand for accurate classification of geographically diverse regions led to advances in object-based methods. A novel spatiotemporal method is presented for object-based land cover classification of satellite imagery using a Graph Neural Network. This paper introduces innovative representation of sequential satellite images as a directed graph by connecting segmented land region through time. The method’s novel modular node classification pipeline utilises the Convolutional Neural Network as a multispectral image feature extraction network, and the Graph Neural Network as a node classification model. To evaluate the performance of the proposed method, we utilised EfficientNetV2-S for feature extraction and the GraphSAGE algorithm with Long Short-Term Memory aggregation for node classification. This innovative application on Sentinel-2 L2A imagery produced complete 4-year intermonthly land cover classification maps for two regions: Graz in Austria, and the region of Portorož, Izola and Koper in Slovenia. The regions were classified with Corine Land Cover classes. In the level 2 classification of the Graz region, the method outperformed the state-of-the-art UNet model, achieving an average F1-score of 0.841 and an accuracy of 0.831, as opposed to UNet’s 0.824 and 0.818, respectively. Similarly, the method demonstrated superior performance over UNet in both regions under the level 1 classification, which contains fewer classes. Individual classes have been classified with accuracies up to 99.17%.https://www.mdpi.com/1424-8220/23/14/6648multispectralSentinel-2superpixelnodeEfficientNetV2GraphSAGE |
| spellingShingle | Domen Kavran Domen Mongus Borut Žalik Niko Lukač Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery Sensors multispectral Sentinel-2 superpixel node EfficientNetV2 GraphSAGE |
| title | Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery |
| title_full | Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery |
| title_fullStr | Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery |
| title_full_unstemmed | Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery |
| title_short | Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery |
| title_sort | graph neural network based method of spatiotemporal land cover mapping using satellite imagery |
| topic | multispectral Sentinel-2 superpixel node EfficientNetV2 GraphSAGE |
| url | https://www.mdpi.com/1424-8220/23/14/6648 |
| work_keys_str_mv | AT domenkavran graphneuralnetworkbasedmethodofspatiotemporallandcovermappingusingsatelliteimagery AT domenmongus graphneuralnetworkbasedmethodofspatiotemporallandcovermappingusingsatelliteimagery AT borutzalik graphneuralnetworkbasedmethodofspatiotemporallandcovermappingusingsatelliteimagery AT nikolukac graphneuralnetworkbasedmethodofspatiotemporallandcovermappingusingsatelliteimagery |