Tree Cover Estimation in Global Drylands from Space Using Deep Learning
Accurate tree cover mapping is of paramount importance in many fields, from biodiversity conservation to carbon stock estimation, ecohydrology, erosion control, or Earth system modelling. Despite this importance, there is still uncertainty about global forest cover, particularly in drylands. Recentl...
| Príomhchruthaitheoirí: | , , , , , |
|---|---|
| Formáid: | info:eu-repo/semantics/article |
| Teanga: | English |
| Foilsithe / Cruthaithe: |
MDPI
2020
|
| Ábhair: | |
| Rochtain ar líne: | http://hdl.handle.net/10835/7706 |
| _version_ | 1789406349686210560 |
|---|---|
| author | Guirado Hernández, Emilio Alcaraz Segura, Domingo Cabello Piñar, Francisco Javier Puertas Ruíz, Sergio Herrera, Francisco Tabik, Siham |
| author_facet | Guirado Hernández, Emilio Alcaraz Segura, Domingo Cabello Piñar, Francisco Javier Puertas Ruíz, Sergio Herrera, Francisco Tabik, Siham |
| author_sort | Guirado Hernández, Emilio |
| collection | DSpace |
| description | Accurate tree cover mapping is of paramount importance in many fields, from biodiversity conservation to carbon stock estimation, ecohydrology, erosion control, or Earth system modelling. Despite this importance, there is still uncertainty about global forest cover, particularly in drylands. Recently, the Food and Agriculture Organization of the United Nations (FAO) conducted a costly global assessment of dryland forest cover through the visual interpretation of orthoimages using the Collect Earth software, involving hundreds of operators from around the world. Our study proposes a new automatic method for estimating tree cover using artificial intelligence and free orthoimages. Our results show that our tree cover classification model, based on convolutional neural networks (CNN), is 23% more accurate than the manual visual interpretation used by FAO, reaching up to 79% overall accuracy. The smallest differences between the two methods occurred in the driest regions, but disagreement increased with the percentage of tree cover. The application of CNNs could be used to improve and reduce the cost of tree cover maps from the local to the global scale, with broad implications for research and management. |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-7706 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2020 |
| publisher | MDPI |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-77062023-04-12T18:57:26Z Tree Cover Estimation in Global Drylands from Space Using Deep Learning Guirado Hernández, Emilio Alcaraz Segura, Domingo Cabello Piñar, Francisco Javier Puertas Ruíz, Sergio Herrera, Francisco Tabik, Siham convolutional neural networks data augmentation deep learning dry forest forest mapping large-scale datasets transfer learning Accurate tree cover mapping is of paramount importance in many fields, from biodiversity conservation to carbon stock estimation, ecohydrology, erosion control, or Earth system modelling. Despite this importance, there is still uncertainty about global forest cover, particularly in drylands. Recently, the Food and Agriculture Organization of the United Nations (FAO) conducted a costly global assessment of dryland forest cover through the visual interpretation of orthoimages using the Collect Earth software, involving hundreds of operators from around the world. Our study proposes a new automatic method for estimating tree cover using artificial intelligence and free orthoimages. Our results show that our tree cover classification model, based on convolutional neural networks (CNN), is 23% more accurate than the manual visual interpretation used by FAO, reaching up to 79% overall accuracy. The smallest differences between the two methods occurred in the driest regions, but disagreement increased with the percentage of tree cover. The application of CNNs could be used to improve and reduce the cost of tree cover maps from the local to the global scale, with broad implications for research and management. 2020-02-19T12:39:43Z 2020-02-19T12:39:43Z 2020-01-21 info:eu-repo/semantics/article 2072-4292 http://hdl.handle.net/10835/7706 en https://www.mdpi.com/2072-4292/12/3/343 Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess MDPI |
| spellingShingle | convolutional neural networks data augmentation deep learning dry forest forest mapping large-scale datasets transfer learning Guirado Hernández, Emilio Alcaraz Segura, Domingo Cabello Piñar, Francisco Javier Puertas Ruíz, Sergio Herrera, Francisco Tabik, Siham Tree Cover Estimation in Global Drylands from Space Using Deep Learning |
| title | Tree Cover Estimation in Global Drylands from Space Using Deep Learning |
| title_full | Tree Cover Estimation in Global Drylands from Space Using Deep Learning |
| title_fullStr | Tree Cover Estimation in Global Drylands from Space Using Deep Learning |
| title_full_unstemmed | Tree Cover Estimation in Global Drylands from Space Using Deep Learning |
| title_short | Tree Cover Estimation in Global Drylands from Space Using Deep Learning |
| title_sort | tree cover estimation in global drylands from space using deep learning |
| topic | convolutional neural networks data augmentation deep learning dry forest forest mapping large-scale datasets transfer learning |
| url | http://hdl.handle.net/10835/7706 |
| work_keys_str_mv | AT guiradohernandezemilio treecoverestimationinglobaldrylandsfromspaceusingdeeplearning AT alcarazseguradomingo treecoverestimationinglobaldrylandsfromspaceusingdeeplearning AT cabellopinarfranciscojavier treecoverestimationinglobaldrylandsfromspaceusingdeeplearning AT puertasruizsergio treecoverestimationinglobaldrylandsfromspaceusingdeeplearning AT herrerafrancisco treecoverestimationinglobaldrylandsfromspaceusingdeeplearning AT tabiksiham treecoverestimationinglobaldrylandsfromspaceusingdeeplearning |