27 KB132 KB17 KB200020242002Amram, OliverFlouzat, GuyLaporterie, Florenceštevilka:1letnik:215 stranistr. 49-53ISSN:1580-3139COBISSID:16040921URN:URN:NBN:SI:doc-CFA540L5enDruštvo za stereologijo in kvantitativno analizo slike, Medicinska fakultetaImage analysis and stereologyEnvironmentImage Processing, Computer-Assistedmatematična morfologijanatural environmentOkoljepiramidni algoritmiSlika, obdelava z računalnikomUrbanizacijaUrbanizationvečresolucijska parametrizacijaThe morphological pyramid and its applications to remote sensing: multiresolution data analysis and features extraction|In remote sensing, sensors are more and more numerous, and their spatial resolution is higher and higher. Thus, the availability of a quick and accurate characterisation of the increasing amount of data is now a quite important issue. This paper deals with an approach combining a pyramidal algorithm and mathematical morphology to study the physiographic characteristics of terrestrial ecosystems. Our pyramidal strategy involves first morphological filters, then extraction at each level of resolution of well-known landscapes features. The approach is applied to a digitised aerial photograph representing an heterogeneous landscape of orchards and forests along the Garonne river (France). This example, simulating very high spatial resolution imagery, highlights the influence of the parameters of the pyramid according to the spatial properties of the studied patterns. It is shown that,the morphological pyramid approach is a promising attempt for multi-levelfeatures extraction by modelling geometrical relevant parametersTEXTznanstveno časopisjejournalsSlovenian National E-content AggregatorNational and University Library of SloveniaDruštvo za stereologijo in kvantitativno analizo slike