Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural Networks
The wide availability of multispectral satellite imagery through projects such as Landsat and Sentinel, combined with the introduction of deep learning in general and Convolutional Neural Networks (CNNs) in particular, has allowed for the rapid and effective analysis of multiple classes of problems...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-03-01
|
| Series: | Land |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-445X/9/3/82 |
| _version_ | 1818550910524063744 |
|---|---|
| author | Stanisław Krysiak Elżbieta Papińska Anna Majchrowska Maciej Adamiak Mikołaj Koziarkiewicz |
| author_facet | Stanisław Krysiak Elżbieta Papińska Anna Majchrowska Maciej Adamiak Mikołaj Koziarkiewicz |
| author_sort | Stanisław Krysiak |
| collection | DOAJ |
| description | The wide availability of multispectral satellite imagery through projects such as Landsat and Sentinel, combined with the introduction of deep learning in general and Convolutional Neural Networks (CNNs) in particular, has allowed for the rapid and effective analysis of multiple classes of problems pertaining to land coverage. Taking advantage of the two phenomena, we propose a machine learning model for the classification of land abandonment. We designed a Convolutional Neural Network architecture that outputs a classification probability for the presence of land abandonment in a given 15−25 ha grid element by using multispectral imaging data obtained through Sentinel Hub. For both the training and validation of the model, we used imagery of the Łódź Voivodeship in central Poland. The main source of truth was a 2009 orthophoto study available from the WMS (Web Map Service) of the Geoportal site. The model achieved 0.855 auc (area under curve), 0.47 loss, and 0.78 accuracy for the test dataset. Using the classification results and the Getis−Ord Gi* statistic, we prepared a map of cold- and hotspots with individual areas that exceed 50 km<sup>2</sup>. This thresholded heatmap allowed for an analysis of contributing factors for both low and intense land abandonment, demonstrating that common trends are identifiable through the interpretation of the classification results of the chosen model. We additionally performed a comparative field study on two selected cold- and hotspots. The study, along with the high-accuracy results of the model’s validation, confirms that CNN-type models are an effective tool for the automatic detection of land abandonment. |
| first_indexed | 2024-12-12T08:52:50Z |
| format | Article |
| id | doaj.art-78cfc7f48e1349f186df363e76649d79 |
| institution | Directory Open Access Journal |
| issn | 2073-445X |
| language | English |
| last_indexed | 2024-12-12T08:52:50Z |
| publishDate | 2020-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Land |
| spelling | doaj.art-78cfc7f48e1349f186df363e76649d792022-12-22T00:30:07ZengMDPI AGLand2073-445X2020-03-01938210.3390/land9030082land9030082Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural NetworksStanisław Krysiak0Elżbieta Papińska1Anna Majchrowska2Maciej Adamiak3Mikołaj Koziarkiewicz4Department of Physical Geography, Faculty of Geographical Sciences, University of Lodz, 90-139 Łódź, PolandDepartment of Physical Geography, Faculty of Geographical Sciences, University of Lodz, 90-139 Łódź, PolandDepartment of Physical Geography, Faculty of Geographical Sciences, University of Lodz, 90-139 Łódź, PolandSoftwareMill, 02-791 Warsaw, PolandSoftwareMill, 02-791 Warsaw, PolandThe wide availability of multispectral satellite imagery through projects such as Landsat and Sentinel, combined with the introduction of deep learning in general and Convolutional Neural Networks (CNNs) in particular, has allowed for the rapid and effective analysis of multiple classes of problems pertaining to land coverage. Taking advantage of the two phenomena, we propose a machine learning model for the classification of land abandonment. We designed a Convolutional Neural Network architecture that outputs a classification probability for the presence of land abandonment in a given 15−25 ha grid element by using multispectral imaging data obtained through Sentinel Hub. For both the training and validation of the model, we used imagery of the Łódź Voivodeship in central Poland. The main source of truth was a 2009 orthophoto study available from the WMS (Web Map Service) of the Geoportal site. The model achieved 0.855 auc (area under curve), 0.47 loss, and 0.78 accuracy for the test dataset. Using the classification results and the Getis−Ord Gi* statistic, we prepared a map of cold- and hotspots with individual areas that exceed 50 km<sup>2</sup>. This thresholded heatmap allowed for an analysis of contributing factors for both low and intense land abandonment, demonstrating that common trends are identifiable through the interpretation of the classification results of the chosen model. We additionally performed a comparative field study on two selected cold- and hotspots. The study, along with the high-accuracy results of the model’s validation, confirms that CNN-type models are an effective tool for the automatic detection of land abandonment.https://www.mdpi.com/2073-445X/9/3/82land covergismachine learningconvolutional neural networksłódź voivodeship |
| spellingShingle | Stanisław Krysiak Elżbieta Papińska Anna Majchrowska Maciej Adamiak Mikołaj Koziarkiewicz Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural Networks Land land cover gis machine learning convolutional neural networks łódź voivodeship |
| title | Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural Networks |
| title_full | Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural Networks |
| title_fullStr | Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural Networks |
| title_full_unstemmed | Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural Networks |
| title_short | Detecting Land Abandonment in Łódź Voivodeship Using Convolutional Neural Networks |
| title_sort | detecting land abandonment in lodz voivodeship using convolutional neural networks |
| topic | land cover gis machine learning convolutional neural networks łódź voivodeship |
| url | https://www.mdpi.com/2073-445X/9/3/82 |
| work_keys_str_mv | AT stanisławkrysiak detectinglandabandonmentinłodzvoivodeshipusingconvolutionalneuralnetworks AT elzbietapapinska detectinglandabandonmentinłodzvoivodeshipusingconvolutionalneuralnetworks AT annamajchrowska detectinglandabandonmentinłodzvoivodeshipusingconvolutionalneuralnetworks AT maciejadamiak detectinglandabandonmentinłodzvoivodeshipusingconvolutionalneuralnetworks AT mikołajkoziarkiewicz detectinglandabandonmentinłodzvoivodeshipusingconvolutionalneuralnetworks |