Preview

PDF 133069-A foreca...pdf | Download (2MB) | Preview

Abstract

We present a method for online detection of land coverchange based on remotely sensed time series. Change is detectedby monitoring deviations between observations and forecasts madeusing the time series historical data and similar time series in thegeographical region. This method and several others were appliedto MODIS 8-day surface reflectance data for problems of detectingsettlement expansion in Limpopo Province, South Africa, and detecting deforestation in New South Wales, Australia. The proposedmethod had significantly shorter median detection delay (DD) forequivalent rates of false alarms compared with the other evaluatedmethods. We obtained a median DD of seven samples for settlementdetection and 14 samples for deforestation detection correspondingto 56 days and 112 days, respectively. This is compared with a median DD of 224 and 544 days for the best other methods evaluated.We suggest that the proposed method is an excellent candidate forland cover change detection where rapid detection is essential.

Item Type:	Article
Authors/Creators:	Olding, WC and Olivier, JC and Salmon, BP and Kleynhans, W
Keywords:	remote sensing, change detection
Journal or Publication Title:	IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Publisher:	IEEE
ISSN:	1939-1404
DOI / ID Number:	10.1109/JSTARS.2019.2905594
Copyright Information:	Licensed under Creative Commons Attribution 3.0 Unported (CC BY 3.0) https://creativecommons.org/licenses/by/3.0/
Related URLs:	UTAS Profile: Olivier, JC Google Scholar: Salmon, BP UTAS Profile: Salmon, BP
Item Statistics:	View statistics for this item

Actions (login required)

Item Control Page