Creation of Event-based Landslide Inventory From Panchromatic Images by Object Oriented Analysis
Year: 2012
Type: article
Abstract: There has been an increase in the application of object-based image classification techniques for fast retrieval of information after natural disasters. One of the fields wherein this method has shown its potential is the detection of landslides. However, most of the studies are confined to the application of object-based image analysis techniques to multispectral images. While this application has shown tremendous potential in the early stage of disaster response, availability of high resolution multispectral images immediately after a disaster situation as reference data for change detection is not always guaranteed. In this paper, we investigated how landslides can be detected from panchromatic images, which lack spectral information of various terrain features. An objectbased change detection technique developed by Martha et al. (2012) using archived panchromatic images was tested in an area in the high Himalayas where several landslides were triggered after a severe rainfall event in the year 2010. Cartosat-1 high resolution panchromatic images acquired by ISRO immediately after the event were used, and a total of 142 landslides in a catchment area of 10.5 km2 were identified. A landslide inventory created by this method can be used for future event-based landslide susceptibility assessment of affected area.
Authors: Tapas R. Martha, N. Kerle
Language: English
Subfield: Management, Monitoring, Policy and Law
Field: Environmental Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2157011297