Comparison of Segment and Pixel Based Non-Parametric Classification of Land Cover in the Amazon Region of Brazil Using Multitemporal Landsat TM/ETM+ Imagery

Comparison of Segment and Pixel Based Non-Parametric Classification of Land Cover in the Amazon Region of Brazil Using Multitemporal Landsat TM/ETM+ Imagery

This study evaluated the ability of segment-based classification paired with non-parametric methods (CART and kNN) to classify a chronosequence of Landsat TM/ETM+ imagery spanning from 1992 to 2002 within the state of Rondônia, Brazil. Pixel-based classification was also implemented for comparison...

Full description

Bibliographic Details
Main Author: Joseph, Katherine Amanda
Other Authors: Forestry
Format: Others
Published: Virginia Tech 2014
Subjects:
Amazon
multitemporal
kNN
non-parametric
CART
remote sensing
contextual classification
Brazil
Landsat
Online Access:http://hdl.handle.net/10919/32802
http://scholar.lib.vt.edu/theses/available/etd-05152005-224254/

Internet

http://hdl.handle.net/10919/32802
http://scholar.lib.vt.edu/theses/available/etd-05152005-224254/

Similar Items