Science Service System

Summary of Proposal LAN2782

Investigator Rosentreter, Johannes - Freie Universität Berlin, Remote Sensing and Geoinformatics
Team Member
Prof. Dr. Waske, Björn - Freie Universität Berlin, Institute for Geographical Sciences
M. Sc. Rosentreter, Johannes - Freie Universität Berlin, Institute for Geographical Sciences
SummaryMegacities are considered as symbol of highly dynamic global urbanization processes. To keep record of urban dynamics, dimension and complexity, earth observation plays a central role. A comprehensive review of existing state-of-the-art geoinformation products is compiled by Gamba & Herold (2009). New initiatives such as the “Global Urban Footprint” project of the DLR (Esch et al. 2012) or the “Global Human Settlement Layer” project of the JRC (Persaresi et al. 2012) reveal new potentials for the geometric resolution of global urban mapping. The thematic depth analysis of these researches are relatively small due to the low spectral resolution. The greatest potentials of hyperspectral data for urban applications have been airborne systems, such as HyMap or AVIRIS. Proportions of urban land cover were achieved and quantified by spectral decomposition (Franke et al. 2001). Recent work has demonstrated the potential of methods from the field of machine learning, which turn out to be useful for high dimensional (hyperspectral) data (Okujeni et al., 2013; Waske et al., 2009; Waske et al., 2010). Although the performance of these methods can be influenced by high dimensional hyperspectral data, they are usually more suitable than conventional analysis methods (Waske et al. 2010). Therefore we aim to create an expressive classification legend, and implement latest machine learning methods using multi-scaled hyperspectral and X-band data as additional value. The methodological development will initially take place in a very well researched German test area in order to keep the number of unknowns as low as possible. Selected Publications: Esch T., H. Taubenböck , A. Roth., W. Heldens, A. Felbier, M. Thiel, M. Schmidt, A. Müller & S. Dech (2012): TanDEM-X mission: New perspectives for the inventory and monitoring of global settlement patterns. – In: Journal of Selected Topics in Applied Earth Observation and Remote Sensing, 6, p. 22. Okujeni A., S. van der Linden, L. Tits, B. Somers & P. Hostert (2013): Support vector regression and synthetically mixed training data for quantifying urban land cover. – In: Remote Sensing of Environment, 137, pp. 184-197. Roscher R., B. Waske & W. Forstner (2012): Incremental Import Vector Machines for Classifying Hyperspectral Data. – In: IEEE Transactions on Geoscience and Remote Sensing, 50, 3463-3473. Suess S., S. van der Linden, P.J. Leitao, A. Okujeni, B. Waske & P.Hostert (2013): Import vector machines for quantitative analysis ofhyperspectral data. – In: IEEE Geoscience and Remote Sensing Letters, in press. Waske, B., Benediktsson, J.A. (2007): Fusion of SupportVector Machines for Classification of Mulitsensor Data. IEEE Transaction on Geoscience and Remote Sensing, vol. 45, no. 12, pp. 3858-3866, doi:10.1109/TGRS.2007.898446. Waske, B., Braun, M. (2009): Classifier ensembles for land cover mapping using multitem-poral SAR imagery.ISPRS Journal of Photogrammetry and Remote Sensing, vol. 64, no.5, pp.450-457, doi: 10.1016/j.isprsjprs.2009.01.003. Waske B., M. Chi, J.A. Benediktsson, S. van der Linden & B. Koetz (2009): Algorithms andapplications for land cover classification – A review. – In: Li D., Gong J., Shan J. (Eds.), Geospatial Technology for Earth Observation, pp.203-233. Waske, B & S. van der Linden(2008): Classifyingmultilevel imagery from SAR and optical sensors by decision fusion. IEEE Transaction on Geoscience and Remote Sensing, vol. 46, no. 5, pp.1457-1466, doi: 10.1109/TGRS.2008.916089. Waske B., S. van derLinden, J.A. Benediktsson, A. Rabe & P. Hostert (2010): Sensitivityof Support Vector Machines to Random Feature Selection in Classification of Hyperspectral Data. – In: IEEE Transaction on Geoscience and RemoteSensing, 48 (7), pp. 2880-2889.

Back to list of proposals

© DLR 2004-2016