Using eCognition Definiens for automated detection of snow avalanches from optical imagery

Abstract

Detection of avalanches from remotely collected optical imagery has been tested through analysis of image properties such as brightness, contrast, and different measures of texture. There have been few publications on the subject, providing an excellent opportunity for new developments. The work conducted at NGI in 2011 aimed at detecting fresh snow avalanches from very-high resolution (VHR) optical imagery. The research presented in this Technical Note has been supported by the Ministry of Petroleum and Energy (OED) through the Norwegian Water Resources and Energy Directorate (NVE).