Land-use/cover classification in a heterogeneous coastal landscape using RapidEye imagery: evaluating the performance of random forest and support vector machines classifiers

摘要: Mapping of patterns and spatial distribution land-use/cover (LULC) has long been based on remotely sensed data. In the recent past, efforts to improve reliability LULC maps have seen a proliferation image classification techniques. Despite these efforts, derived are still often judged be insufficient quality for operational applications, due disagreement between generated reference this study we sought pursue two objectives: first, test new-generation multispectral RapidEye imagery output using machine-learning random forest (RF) support vector machines (SVM) classifiers in heterogeneous coastal landscape; second, determine importance different bands output. Accuracy thematic was assessed by computing confusion matrices classifiers’ cover with respective independent validation data sets. An overall accuracy 93.07% kappa value ...