statistique de Getis

OBS: statistique de Getis : terme proposé par Mme Michèle Bernier-Cardou, statisticienne, Ressources naturelles Canada, Centre de foresterie des Laurentides.
OBS: «Local Spatial Autocorrelation Characteristics of Landsat TM Imagery of a Managed Forest Area», par M. Wulder et B. Boots. [...] la mesure des indicateurs locaux d'association spatiale (LISA, «local indicators of spatial indicators» [sic : plutôt «local indicators of spatial association»]) permet d'évaluer pour chaque pixel de l'image à la fois le degré de dépendance spatiale du pixel par rapport aux pixels voisins et l'amplitude des valeurs de variation dans le voisinage du pixel. Dans cette étude, on applique une mesure LISA, la mesure Getis (G[indice i]), à des images Landsat TM d'une région forestière aménagée. [...] Cette recherche exploratoire confirme que l'information sur la dépendance spatiale telle que dérivée de la mesure de (G[indice i]) constitue une nouvelle source importante d'information spatiale pour l'évaluation des images numériques en milieu forestier.

CONT: Application of the Getis statistic to hemispheric and regional scale passive microwave derived snow water equivalent imagery. ... In this study, five day averaged Special Sensor Microwave/Imager (SSM/I) derived SWE [snow water equivalent] imagery are analyzed with the Getis statistic (Gi), a local indicator of spatial autocorrelation.
CONT: In contrast to semivariance, the Getis statistic (G[subscript i]), generates values which relate variations within patterns of spatial dependence. Thus, it has the potential to uncover discrete spatial regimes which might be overlooked by existing techniques. ... Wulder and Boots have adapted the Getis statistic for processing remotely sensed imagery. The Getis statistic yields a standardized value which indicates both the degree of autocorrelation in the values of the digital numbers centered on a given pixel and the magnitude of these values in relation to those of the entire image. The G[subscript i] values measure the extent to which a pixel is surrounded by a cluster of high or low values of a particular variable (e.g. DN). Large positive G[subscript i] values denote a cluster of high DN values; large negative G[subscript i] values indicate the spatial dependence within a tree crown or between shadow elements. High positive values generated from infrared wavelength image data indicate the presence of a tree object whereas high negative values relate a non-tree feature.
OBS: Getis: This name refers to Mr. Arthur Getis, of San Diego State University, who developed the local autocorrelation statistic named after him. [Information obtained from Mr. Frank Eichel, RPF, ADV. Dipl. GIS, Msc Forestry (Remote Sensing) candidate, University of British Columbia.]