Głównym celem prezentowanych badań było opracowanie zautomatyzowanej metody kartowania klas pokrycia terenu występujących w przestrzeni miejskiej, na drodze integracji komplementarnych technologii, tj.: wysokorozdzielczych zobrazowań satelitarnych (GeoEye-1) oraz chmur punktów lotniczego skanowania laserowego (ALS). Cel cząstkowy polegał również na porównaniu dokładności klasyfikacji OBIA zbiorowisk roślinnych w oparciu o różne zestawy danych wejściowych, w stopniu możliwie maksymalnie zautomatyzowanym, bez stosowania jakichkolwiek pól treningowych. Jednocześnie autorzy postawili sobie za cel przedstawienie statystyk przestrzennych opisujących zieleń miejską w wymiarze 3D i zaproponowali szersze wykorzystanie danych ALS.
EN
The paper presents first results of advanced research concerning the use of integrated airborne laser scanning data and high resolution satellite images for the purpose of urban land cover mapping, particularly vegetation. Object-based image analysis was used for data processing, without any training areas and with three different approaches: A - only ALS data; B - based on GeoEye-1 satellite image only; C - based on both integrated datasets. Using integrated point clouds with spectral information stored in GeoEye-1 bands resulted in the best classification outcome (Kappa = 0.83), allowing detection of all classes that were the subject of analysis. Vertical structure assessment possibilities with the use of point cloud data were also shown in the paper.
The research presented in the paper has been aimed at mapping the basic types of land-use in the upper Raba watershed (south Poland). The maps have been prepared for a study of the influence of land-use changes within the watershed on the sediment yields introduced into the reservoir. Because the erosion models used for sediment yields prediction need only to identify the main land-use / land cover classes (arable land, meadows and pastures, forests, waters, developed areas), the maps have been based on classification of middle-resolution satellite images (Landsat TM). In the research the results of traditional pixel-based classification were compared to the ones obtained in the object based approach. Six different Landsat TM images were classified. The methodology of both classification approaches have been described in the paper. The accuracy assessment of the classification results was based on their comparison with the land use types defined by the photo interpretation of colour composite images. The assessment was done by two operators. Each of them used different set of two hundred and fifty randomly generated sample points. In most cases the pixel-based approach resulted in higher overall accuracy. However, if overall accuracy confidence intervals are taken into consideration, none of the methods can be definitely recognised as a better one.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.