Estimation of Mediterranean forest attributes by the application of k-NN procedures to multitemporal Landsat ETM+ images
Author(s)
Maselli, Fabio
Chirici, Gherardo
Bottai, Lorenzo
Corona, Piermaria
Marchetti, Marco
Date Issued
2005
Type
Article
Abstract
Routinely, applications of nonparametric estimation methods to satellite data for
assisting the creation of forest inventories in Northern European countries are
stimulating interest in the possible extension of these methods to more complex
Mediterranean areas. This is the subject of the current work, which presents an
experiment based on the integration of remotely sensed images and sample field
measurements aimed at producing forest attribute maps in central Italy. Testing
was carried out in an area where 370 geocoded field plots, sampled on a singlestage
cluster design, were collected to characterize wood and non-wood forest
attributes. These ground data served to apply various k-Nearest Neighbour (k-
NN) estimation procedures to multitemporal Landsat 7 ETM+ images in order
to map major forest attributes (basal area and simulated leaf area index, LAI).
More specifically, the investigation focused on evaluating the effects of using
satellite images from different periods of the growing season and spectral metrics
of increasing complexity. The results achieved by the examined methods are
finally discussed in order to provide guidelines for possible operational
utilization.
Additional information
L'articolo รจ disponibile sul sito dell'editore http://www.tandf.co.uk/journals/
Citation
Maselli, F. et al. 2005. Estimation of Mediterranean forest attributes by the application of k-NN procedures to multitemporal Landsat ETM+ images. "International Journal of Remote Sensing " 26 (17): 3781-3796.
File(s)![Thumbnail Image]()
Name
maselli_KNN_IJRS1.pdf
Size
51.14 KB
Format
Adobe PDF
Checksum (MD5)
825e6607a857d1ed448d711e86366fdb