Sima - A Software System For Motion Analysis In Satellite . . .

• Main Authors: Andrea Miene, Rene Bartholomä, Markus Geschke, David Mikota, Roland Neumann, Andreas Winter
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Subjects: Greenland; Weddell; Weddell Sea
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.20.5346; http://earth.esa.int/pub/ESA_DOC/gothenburg/074miene.pdf

The aim of the Sima (Satellite Image Analysis) project is to examine digital image analysis approaches for object recognition and tracking in SAR satellite image sequences. The focus was put on the development of a new software system for the recognition and tracking of ice floes and icebergs. In order to extract the drift information from the SAR image sequence various image processing steps have to be carried out, i.e. preprocessing, contour analysis and motion tracking. The system provides miscellaneous operators for image preprocessing, edge detection methods and various approaches to image motion analysis. By building up an image processing graph, the different operators may be configured and composed together by the user to enhance the quality of the single images and to detect objects and analyse object motion. The results of the motion analysis are motion vector fields, which allow an interpretation of the ice drift. For visual representation of the motion vector fields deformation grids and needle images are available. The system was tested to analyse ice drift on SAR images taken from the the Weddell Sea and at the east coast from Greenland.