Centre de documentation
scientifique

Object-based crop classification using multi-temporal SPOT-5 imagery and textural features with a Random Forest classifier / Huanxue Zhang in Geocarto international, vol 33 n° 10 (October 2018)  

Public [article]  inGeocarto international > vol 33 n° 10 (October 2018) . - pp 1017 - 1035 Titre : Object-based crop classification using multi-temporal SPOT-5 imagery and textural features with a Random Forest classifier Type de document : Article/Communication Auteurs : Huanxue Zhang, Auteur ; Qiangzi Li, Auteur ; Jiangui Liu, Auteur ; Taifeng Dong, Auteur ; Heather McNairn, Auteur Année de publication : 2018 Article en page(s) : pp 1017 - 1035 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse d'image orientée objet [Termes IGN] bande spectrale [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] corrélation par régions de niveaux de gris [Termes IGN] extraction de traits caractéristiques [Termes IGN] image SPOT 5 [Termes IGN] indice de végétation [Termes IGN] limite de terrain [Termes IGN] Ontario (Canada) [Termes IGN] réflectance spectrale [Termes IGN] segmentation d'image [Termes IGN] surface cultivée [Termes IGN] surveillance agricole [Termes IGN] texture d'image [Termes IGN] variogramme Résumé : (auteur) In this study, an object-based image analysis (OBIA) approach was developed to classify field crops using multi-temporal SPOT-5 images with a random forest (RF) classifier. A wide range of features, including the spectral reflectance, vegetation indices (VIs), textural features based on the grey-level co-occurrence matrix (GLCM) and textural features based on geostatistical semivariogram (GST) were extracted for classification, and their performance was evaluated with the RF variable importance measures. Results showed that the best segmentation quality was achieved using the SPOT image acquired in September, with a scale parameter of 40. The spectral reflectance and the GST had a stronger contribution to crop classification than the VIs and GLCM textures. A subset of 60 features was selected using the RF-based feature selection (FS) method, and in this subset, the near-infrared reflectance and the image acquired in August (jointing and heading stages) were found to be the best for crop classification. Numéro de notice : A2019-049 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/10106049.2017.1333533 Date de publication en ligne : 23/06/2017 En ligne : https://doi.org/10.1080/10106049.2017.1333533 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92063 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
059-2018041	RAB	Revue	Centre de documentation	En réserve L003	Disponible