Comparer des méthodes

	Classification globale par télédétection	Autocorrélation spatiale
Domaine	Analyse spatiale	Analyse spatiale
Famille	Regression model	Regression model
Année d'origine	1970s–1980s (pixel-based global classifiers); global land-cover products 1990s–2000s	1950
Auteur d'origine	Rosenfeld & Kak; Jensen; Campbell & Wynne (textbook codifications)	P. A. P. Moran (global measure, 1950); Roy Geary (Geary's C, 1954); Luc Anselin (LISA, 1995)
Type	Supervised / unsupervised image classification	Spatial statistic / exploratory spatial data analysis
Source fondatrice	Campbell, J. B., & Wynne, R. H. (2011). Introduction to Remote Sensing (5th ed.). Guilford Press. ISBN: 978-1609181765	Moran, P. A. P. (1950). Notes on continuous stochastic phenomena. Biometrika, 37(1/2), 17–23.
Alias	global pixel-based classification, global image classification, wall-to-wall remote sensing classification, global land cover classification	spatial dependence, geographic autocorrelation, spatial clustering measure, SA
Apparentées	3	5
Résumé	Global Remote Sensing Classification assigns every pixel across an entire image or worldwide dataset to a discrete land-cover or thematic class. Treating the scene uniformly — rather than adapting to local subregions — this wall-to-wall approach underpins continental and global land-cover products such as GlobCover, FROM-GLC, and ESA CCI Land Cover.	Spatial autocorrelation quantifies the degree to which a variable's values at nearby locations resemble each other more (positive autocorrelation) or less (negative autocorrelation) than expected by chance. Global indices such as Moran's I summarise the pattern across the entire study area, while local variants reveal clusters and outliers at the level of individual observations.