Classification of Agricultural Crops with Random Forest and Support Vector Machine Algorithms Using Sentinel-2 and Landsat-8 Images DOI Open Access
Murat Güven TUĞAÇ, Fatih Fehmi Şimşek, Harun TORUNLAR

et al.

International Journal of Environment and Geoinformatics, Journal Year: 2024, Volume and Issue: 11(3), P. 106 - 118

Published: Sept. 14, 2024

Monitoring crop development and mapping cultivated areas are important for reducing risks to food security due climate change. Remote sensing techniques contribute significantly the efficient effective management of agricultural production. In this study, fields (sunflower, wheat, maize, oat, chickpea, sugar beet, alfalfa, onion, fallow) other (non-agricultural, pasture, lake) were identified by using Random Forest (RF) Support Vector Machines (SVM) machine learning algorithms with Sentinel-2 Landsat-8 images in area covering Polatlı, Haymana Gölbaşı districts Ankara province Multi-temporal used distinguish winter summer crops, taking into account periods. As a result classification; overall accuracy RF SVM models S2 89.5% 84.6% kappa coefficients 0.88 0.83, while L8 79% 78.1% 0.76 0.75. model was found have higher prediction than SVM. imagery has all classes compared Landsat-8, indicating that its high temporal spatial resolution is more suitable great potential pattern detection.

Language: Английский

Classification of Agricultural Crops with Random Forest and Support Vector Machine Algorithms Using Sentinel-2 and Landsat-8 Images DOI Open Access
Murat Güven TUĞAÇ, Fatih Fehmi Şimşek, Harun TORUNLAR

et al.

International Journal of Environment and Geoinformatics, Journal Year: 2024, Volume and Issue: 11(3), P. 106 - 118

Published: Sept. 14, 2024

Monitoring crop development and mapping cultivated areas are important for reducing risks to food security due climate change. Remote sensing techniques contribute significantly the efficient effective management of agricultural production. In this study, fields (sunflower, wheat, maize, oat, chickpea, sugar beet, alfalfa, onion, fallow) other (non-agricultural, pasture, lake) were identified by using Random Forest (RF) Support Vector Machines (SVM) machine learning algorithms with Sentinel-2 Landsat-8 images in area covering Polatlı, Haymana Gölbaşı districts Ankara province Multi-temporal used distinguish winter summer crops, taking into account periods. As a result classification; overall accuracy RF SVM models S2 89.5% 84.6% kappa coefficients 0.88 0.83, while L8 79% 78.1% 0.76 0.75. model was found have higher prediction than SVM. imagery has all classes compared Landsat-8, indicating that its high temporal spatial resolution is more suitable great potential pattern detection.

Language: Английский

Citations

0