Science Service System
You are here : Home : Proposals_Summary

Summary of Proposal LAN2932

TitleClassification of vegetation types for biosphere modeling
Investigator Lipping, Tarmo - Tampere University of Technology, Information Technology, Pori
Team MemberNo team members defined
SummaryThe purpose of the research work supported by this proposal is to develop methods for integrating data series obtained using various spaceborne and airborne imaging technologies into a unified framework for environmental monitoring purposes. The research stems from the practical need to monitor the distribution and dynamics of the biotopes in the vicinity of industrial sites of environmental impact such as mines, nuclear power plants and the spent nuclear fuel repository, for example. This research effort will be focused on two main sites: the Olkiluoto island and the lakes of Kivijärvi and Poosjärvi in North Satakunta. From these sites reference data are available to our research group from previous research collaboration. Also, we have SAR-X data acquired in 2009, 2010 and 2013 from Olkiluoto and in 2014 from the Kivijärvi /Poosjärvi area. Airborne RGB+CIR data is also available from both sites. We propose acquisition of 6 HighRes SpotLight (HS) SAR images (Dualmode, HH & VV); 3 images of the Olkiluoto island and 3 images of the Poosjärvi / Kivijärvi area. The acquisition times should cover thephenological cycle so that one image of each site is acquired in latespring, one at the end of July or beginning of August and one in latesummer (October) The main objectives of the research are to study – at what phase of the phenological cycle the data contains most information from the point of view of biotope classification at sites similar to Olkiluoto and North-Satakunta – what is the contribution of SAR-X data to the classification performance and how these data should be integrated in the most effective way for environmental monitoring. Data analysis will concentrate on the techniques of data fusion as well as on extraction of target information using state of the art classifiers. It is expected that the research will yield the following results: – a scientifically justified data acquisition plan for environmental monitoring at sites similar to the above mentioned test sites – a set of algorithmic tools for remote sensing data fusion and classification for environmental monitoring purposes.

Back to list of proposals

© DLR 2004-2016