Aller au contenu principal

page search

Bibliothèque Framework for monitoring the conversion of cultivated land to construction land using SAR image time series

Framework for monitoring the conversion of cultivated land to construction land using SAR image time series

Framework for monitoring the conversion of cultivated land to construction land using SAR image time series

Resource information

Date of publication
Décembre 2015
Resource Language
ISBN / Resource ID
AGRIS:US201600058175
Pages
794-803

The land cover change from cultivated land to construction land is a world issue since the urbanization process is extensively studied around the world. Chengdu, China, is a representative urbanization area, where cloud cover is very high most of the time, restricting the use of visible and near-infrared satellite data. Here, we present a novel framework for land change monitoring based on synthetic aperture radar (SAR) time series, which comprises three key components: (1) construction of pixel-level SAR image time series; (2) spatio-temporal similarity analysis based on morphological-structural characteristics of time series; and (3) iterative binary partition mean square error model analysis to ascertain change nodes. Experimental results showed that the proposed framework could effectively extract the change nodes and change pixels, with correctness of 85.82% and completeness of 84.78%, outperforming the time-series-only (non-spatial) method, as well as traditional classification methods, and the same framework using shorter Landsat TM image time series.

Share on RLBI navigator
NO

Authors and Publishers

Author(s), editor(s), contributor(s)

Wang, Yafei
Zhao, Feifei
Cheng, Liang
Yang, Kang

Publisher(s)
Data Provider
Geographical focus