Land Use/cover Classification with Classification and Regression Tree Applied to MODIS Imagery
This study attempted to develop a low-cost, high-precision method, to acquire land use/cover data by combining multi-temporal and multi-spectral Moderate Resolution Imaging Spectroradiometer (MODIS). The results indicate that, Classification and Regression Tree algorithm clearly outperforms the Maxi...
Saved in:
Published in | Journal of applied sciences (Asian Network for Scientific Information) Vol. 13; no. 18; p. 3770 |
---|---|
Main Authors | , , , , |
Format | Journal Article |
Language | English |
Published |
2013
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | This study attempted to develop a low-cost, high-precision method, to acquire land use/cover data by combining multi-temporal and multi-spectral Moderate Resolution Imaging Spectroradiometer (MODIS). The results indicate that, Classification and Regression Tree algorithm clearly outperforms the Maximum Likelihood in land use/cover classification using MODIS, and the first principal component with multi-spectral MODIS image that reflected more soil information can efficiently improve the accuracy of classification based on MODIS NDVI time series. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 1812-5654 1812-5662 |
DOI: | 10.3923/jas.2013.3770.3773 |