Digital images are rich in data, but in many instances they are so complex
as to require spatial filtering to distinguish the structures in them and f
acilitate interpretion. The filtering can be done geostatistically by krigi
ng analysis. It proceeds in two stages. The first involves modelling the co
rrelation structure in the imagery by decomposing the variogram into indepe
ndent spatial components. The second takes each component in turn and krige
s it, thereby filtering it from the others. The paper describes the theory
and illustrates it with an example of an analysis of a SPOT image in a fore
sted landscape of the south-eastern United States.
Variograms of the three wavebands, originally recorded as digital numbers a
nd for the red and infrared transformed to the logarithms, revealed spatial
variation on two distinct scales with effective ranges of 300m and 3 km. T
hese variograms and that of the Normalized Difference Vegetation Index (NDV
I) were fitted by nested (double) exponential models. The two spatial compo
nents in the scene were then estimated separately by kriging analysis and m
apped. The maps of NDVI are displayed and compared with data from ground su
rvey. The short-range component represents an intricate pattern of dissecti
on and its associated vegetation. The long-range component is that of the m
ajor landform units and associated ground cover classes.