11:30 Automated feature extraction by combining polarimetric SAR and object-based image analysis for monitoring of natural resource exploitation
Plank, Simon; Mager, Alexander; Schoepfer, Elisabeth German Aerospace Center (DLR), Germany
Show abstract
An automated feature extraction procedure based on the combination of a pixel-based unsupervised classification of polarimetric synthetic aperture radar data (PolSAR) and an object-based post-classification is presented. High resolution SpotLight dual-polarimetric (HH/VV) TerraSAR-X imagery acquired over the Doba basin, Chad, is used for method development and validation. In an iterative training procedure the best suited polarimetric speckle filter, processing parameters for the following entropy/anisotropy/alpha (H/A/α) decomposition and the heron based unsupervised Wishart classification are determined. By considering feature properties such as shape and area, the subsequent object-based post-classification increases the user’s and producer’s accuracy of the feature extraction procedure by an order of a magnitude to finally 59% to 71% in each case (valid for an area based accuracy assessment), or to even 74% to 89%, taking only the numbers of correctly/falsely detected features into account. In addition, the high transferability of the methodology is verified by an application to a second TerraSAR-X acquisition. The feature extraction procedure is developed for monitoring oil field infrastructure. For developing countries, several studies reported a high correlation between the dependence of oil exports and violent conflicts. Moreover, land use and environmental issues also occur in countries characterized by a peaceful development of their oil industry. Consequently, to support problem solving, an independent monitoring of the oil field infrastructure by Earth observation is proposed, enabling monitoring of large areas within a short time to compare the real amount of land used by the oil exploitation and the companies’ contractual obligations. The developed method focuses on the monitoring of the oil well pads, characterized by rectangular, approximately 50 m x 100 m large patches of bare land.
11:10 Polarimetry-based land cover classification with Sentinel-1 data
Banqué, Xavier (1); Lopez-Sanchez, Juan M (2); Duro, Javier (1); Ballester, David (2); Koudogbo, Fifame (1) 1: Altamira Information, Barcelona Spain; 2: Universidad de Alicante, Spain
Show abstract
With the arrival of Sentinel-1, SAR based Earth Observation applications will benefit from a new open source of dual polarization coherent data with a short revisit time that will boost the spectrum of remote sensing applications. Among them, land cover classification is a baseline technology upon which a myriad of higher level applications such as emergency management or urban planning can be built. The availability of coherent dual polarization products, comprising complex values and inter-channel phase information, will involve a major advantage in land cover classification with respect to former SAR missions. The presented research focuses on the assessment of the exploitation of the Sentinel-1 dual polarization data for land cover classification. In order to take advantage of massive data availability produced by Sentinel-1, data used in this research work will be acquired in Interferometric Wide Swath mode. Test site is located in Munich area, Germany, with an interest zone big enough to contain the different main land cover classes. Other sites might be added shortly according to data acquisition plan as soon as they will be available. The developed preliminary classifier is based on the interpretation of S.Cloude’s Dual Polarization Entropy/Alpha Decomposition as well as on the use of other polarimetric figures. Specifically, the following polarimetric indicators will be assessed: the inter-channel phase difference, the channels cross-correlation, the cross and copolar channels ratio and both cross and copolar normalized backscattering coefficients. The work carried out concentrates on the joint interpretation of the backscattering response of the cross and copolar channels channels for four or five different distributed targets that set the basis for an unsupervised simple land cover classifier. Data calibration in both channels is accomplished based on annotation data, prior to the speckle filtering and polarimetric analysis so that interpretation of the polarimetric properties of the terrain is robust. The availability of copolar and crosspolar channels provides two of the three components of the lexicographic feature vector, theoretically enabling, for example, the detection of volume scattering produced by the forest canopy. The developed research targets a preliminary classifier able to differentiate between four or five terrain classes, including water, urban, forest and bare soil. The decision principle will be trained using one area of interest and validated using the rest of the test site. In conclusion, this paper presents the potential of using dual polarization Sentinel-1 data for land cover classification, highlighting the major step forward with respect to Sentinel-1 predecessors with no polarimetric capabilities. The preliminary classifier presented is able to differentiate between several basic ground cover classes providing a baseline tool to be used in higher level Earth Observation applications.
In May 2013, the Biomass mission concept was selected to become the next in the series of satellites developed to further our understanding of Earth.
REPLAY EARTH EXPLORER 7 USER CONSULTATION MEETING (PART 1)
POLinSAR 2013 Workshop