Centre de documentation
scientifique

High spatial resolution satellite imagery, DEM derivatives, and image segmentation for the detection of mass wasting processes / J. Barlow in Photogrammetric Engineering & Remote Sensing, PERS, vol 72 n° 6 (June 2006)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 72 n° 6 (June 2006) . - pp 687 - 692 Titre : High spatial resolution satellite imagery, DEM derivatives, and image segmentation for the detection of mass wasting processes Type de document : Article/Communication Auteurs : J. Barlow, Auteur ; Steven E. Franklin, Auteur ; Y. Martin, Auteur Année de publication : 2006 Article en page(s) : pp 687 - 692 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] classification ascendante hiérarchique [Termes IGN] classification dirigée [Termes IGN] effondrement de terrain [Termes IGN] géomorphologie locale [Termes IGN] image à haute résolution [Termes IGN] image SPOT 5 [Termes IGN] interprétation automatique [Termes IGN] modèle numérique de surface [Termes IGN] réflectance du sol [Termes IGN] relief [Termes IGN] segmentation d'image Résumé : (Auteur) An automated approach to identifying landslides using a combination of high-resolution satellite imagery and digital elevation derivatives is offered as an alternative to aerial photographic interpretation. Previous research has demonstrated that per pixel spectral response patterns are ineffective in discriminating mass movements. This technique utilizes image segmentation and digital elevation data in order to identify mass movements based not only on their reflectance but also on their shape properties and their geomorphic context. Dividing the classification by process into debris slides, debris flows, and rock slides makes the method far more useful than methods that group all mass movements together. A hierarchical classification scheme is utilized to eliminate areas that are not of interest and to identify areas where mass movements are probable. A supervised classification is then conducted using spectral, shape, and textural properties to identify failures that were greater than 1 ha in area. The resulting accuracy was 90 percent for debris slides, 60 percent for debris flows, and 80 percent for rock slides. Copyright ASPRS Numéro de notice : A2006-232 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.72.6.687 En ligne : https://doi.org/10.14358/PERS.72.6.687 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=27959 [article]