GEOSCAN, résultats de la recherche

Menu GEOSCAN


TitreInfluence of sample distribution and prior probability adjustment on land cover classification
TéléchargerTéléchargements
LicenceVeuillez noter que la Licence du gouvernement ouvert - Canada remplace toutes les licences antérieures.
AuteurPouliot, D; Latifovic, R; Parkinson, W
SourceGéomatique Canada, Dossier public 23, 2016, 13 pages, https://doi.org/10.4095/297517 (Accès ouvert)
Année2016
ÉditeurRessources naturelles Canada
Documentdossier public
Lang.anglais
DOIhttps://doi.org/10.4095/297517
Mediaen ligne; numérique
Formatspdf
ProvinceAlberta
SNRC74D; 74E; 84A; 84H
Lat/Long OENS-114.0000 -110.0000 58.0000 56.0000
Sujetsétablissement de modèles; méthodes analytiques; satellite LANDSAT; classification des formes de relief; géophysique; géomathématique
Illustrationssatellite images; plots
ProgrammeScience de la télédétection spatiale, Méthodologie
Diffusé2016 01 28
Résumé(disponible en anglais seulement)
Machine learning algorithms are widely used for remote sensing land surface characterization. Successful implementation requires a representative training sample for the domain it will applied in (i.e. area of interest or validation domain). However, accessibility and cost strongly limit the acquisition of suitable training samples for large regional applications. Further, it is often desirable to use previously developed datasets where significant resources have been invested, such as data developed from extensive field survey or high resolution remotely sensed imagery. These data often only partially represent the domain of interest and can lead to various forms of sample bias (land cover distribution or class properties). Classifier spatial extension is an extreme case, where a sample is trained from one region (i.e. sample domain) and applied in another (i.e. application domain). This approach is desirable from a cost perspective, but achieving acceptable accuracy is often difficult. In this research we investigate two approaches to account for possible differences between the sample and application domain land cover distributions. The first is an iterative resampling approach to predict the application distribution and adjust the sample distribution to match. The second is the use of prior probabilities to adjust class memberships. Results reinforce the importance of the land cover distribution on accuracy for algorithms that are designed to minimize the classification error with training data. Of the adjustments methods tested resampling was superior if the application domain distribution was well known. However, if it is not then the use of prior probabilities performed similarly overall. A generic model was developed to predict if resampling or prior adjustment should be applied to enhance accuracy.
Résumé(Résumé en langage clair et simple, non publié)
Un dossier public est un document technique qui examine le rôle de l¿échantillonnage et des mécanismes de correction de l¿échantillonnage pour améliorer la cartographie de la couverture terrestre à l¿aide des données de télédétection, par l¿utilisation d¿algorithmes d¿apprentissage automatique visant à réduire au minimum les erreurs de données attribuables à l¿apprentissage.
GEOSCAN ID297517