Improving Bimonthly Landscape Monitoring in Morocco, North Africa, by Integrating Machine Learning with GRASS GIS DOI Creative Commons
Polina Lemenkova

Geomatics, Год журнала: 2025, Номер 5(1), С. 5 - 5

Опубликована: Янв. 20, 2025

This article presents the application of novel cartographic methods vegetation mapping with a case study Rif Mountains, northern Morocco. The area is notable for varied geomorphology and diverse landscapes. methodology includes ML modules GRASS GIS ‘r.learn.train’, ‘r.learn.predict’, ‘r.random’ algorithms supervised classification implemented from Scikit-Learn libraries Python. approach provides platform processing spatiotemporal data satellite image analysis. objective to determine robustness “DecisionTreeClassifier” “ExtraTreesClassifier” algorithms. time series images covering Morocco consists six Landsat scenes 2023 bimonthly interval. Land cover maps are produced based on processed, classified, analyzed images. results demonstrated seasonal changes in land types. validation was performed using dataset Food Agriculture Organization (FAO). contributes environmental monitoring North Africa processing. Using RS combined powerful functionality FAO-derived datasets, topographic variability, moderate-scale habitat heterogeneity, distribution types have been assessed first time.

Язык: Английский

Improving Bimonthly Landscape Monitoring in Morocco, North Africa, by Integrating Machine Learning with GRASS GIS DOI
Polina Lemenkova

SSRN Electronic Journal, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

1
Improving Bimonthly Landscape Monitoring in Morocco, North Africa, by Integrating Machine Learning with GRASS GIS DOI Creative Commons
Polina Lemenkova

Geomatics, Год журнала: 2025, Номер 5(1), С. 5 - 5

Опубликована: Янв. 20, 2025

This article presents the application of novel cartographic methods vegetation mapping with a case study Rif Mountains, northern Morocco. The area is notable for varied geomorphology and diverse landscapes. methodology includes ML modules GRASS GIS ‘r.learn.train’, ‘r.learn.predict’, ‘r.random’ algorithms supervised classification implemented from Scikit-Learn libraries Python. approach provides platform processing spatiotemporal data satellite image analysis. objective to determine robustness “DecisionTreeClassifier” “ExtraTreesClassifier” algorithms. time series images covering Morocco consists six Landsat scenes 2023 bimonthly interval. Land cover maps are produced based on processed, classified, analyzed images. results demonstrated seasonal changes in land types. validation was performed using dataset Food Agriculture Organization (FAO). contributes environmental monitoring North Africa processing. Using RS combined powerful functionality FAO-derived datasets, topographic variability, moderate-scale habitat heterogeneity, distribution types have been assessed first time.

Язык: Английский

Процитировано

1