DocumentCode :
3260976
Title :
Online change detection: Monitoring land cover from remotely sensed data
Author :
Fang, Yi ; Ganguly, Auroop R. ; Singh, Nagendra ; Vijayaraj, Veeraraghavan ; Feierabend, Neal ; Potere, David T.
Author_Institution :
Computational Sci. & Eng. Div., Oak Ridge Nat. Lab., TN
fYear :
2006
fDate :
Dec. 2006
Firstpage :
626
Lastpage :
631
Abstract :
We present a fast and statistically principled approach for land cover change detection. The approach is illustrated with a geographic application that involves analyzing remotely sensed data to detect changes in the normalized difference vegetation index (NDVI) in near real time. We use the Wal-Mart land cover change data set as a nontraditional way to monitor and validate known cases of NDVI change. A reference distribution has been justified to fit the available data. An adaptive metric based on the exponentially weighted moving average (EWMA) of normal scores derived from p-values is tracked for new or streaming data, leading to alarms for large or sustained changes. A heuristic algorithm based on the property of the metric is proposed for change point detection. The proposed framework performed well on the validation dataset
Keywords :
geographic information systems; remote sensing; sensor fusion; change point detection; exponentially weighted moving average; land cover monitoring; normalized difference vegetation index; online change detection; remotely sensed data; Change detection algorithms; Data analysis; Data engineering; Heuristic algorithms; Laboratories; Remote monitoring; Time series analysis; US Government; Urban planning; Vegetation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Data Mining Workshops, 2006. ICDM Workshops 2006. Sixth IEEE International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
0-7695-2702-7
Type :
conf
DOI :
10.1109/ICDMW.2006.125
Filename :
4063701
Link To Document :
بازگشت