Long-term forest mapping from classification of MODIS time series: best practices - Université de Rennes Accéder directement au contenu
Article Dans Une Revue Journal of applied remote sensing Année : 2019

Long-term forest mapping from classification of MODIS time series: best practices

Résumé

Coarse spatial resolution (CSR) time series have been successfully used at regional scale to produce homogeneous and up-to-date forest cover maps. This study aims to classify CSR time series using a nomenclature as detailed as national forest inventories nomenclatures. To identify best practices for classifying time series, three algorithms are compared: maximum likelihood, support vector machine, and random forest. For each algorithm, training, temporal compositing, and selection of input features have been optimized Results establish a clear improvement in classification accuracy when red, near-infrared, and short-wave infrared spectral bands are used instead of vegetation indices. Temporal compositing has a major impact when the whole phenological cycle is used for 3 consecutive years. Random forest produces the best classification, support vector machine appears to be sensitive to overtraining, and maximum likelihood is unable to deal with the complex characteristics of forest and natural vegetation classes.
Fichier non déposé

Dates et versions

hal-02420467 , version 1 (19-12-2019)

Identifiants

Citer

Jean-Philippe Denux, Emmanuelle Cano, Laurence Hubert-Moy, Marie Parrens, Véronique Cheret. Long-term forest mapping from classification of MODIS time series: best practices. Journal of applied remote sensing, 2019, 14 (02), pp.1. ⟨10.1117/1.JRS.14.022208⟩. ⟨hal-02420467⟩
81 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More