In this thesis, an approach to automatically derive information about land cover from the remotely sensed data is presented. The data interpretation was done with classification process and performed in software eCognition Developer. The Object-based image analysis, which assignes the classes - for example land cover types, to clusters of pixels (=objects), was used. For the classification, products of two different data sources were combined - the orthophotos generated from aerial imagery and Normalized Digital surface model derived from LiDAR data. Five types of landscape elements were identified and classified.

This master thesis deals with a classification technique for an automatic detection of different land cover types from combination of high resolution imagery and LiDAR data sets. The main aim is to introduce additional post-processing method to commonly accessible quality data sets which can replace traditional mapping techniques for certain type of applications. Classification is the process of dividing the image into land cover categories which helps with continuous and up-to-date monitoring management. Nowadays, with all the technologies and software available, it is possible to replace traditional monitoring methods with more automated processes to generate accurate and cost-effective results. This project uses object-oriented image analysis (OBIA) to classify available data sets into five main land cover classes. The automate classification rule set providing overall accuracy of 88% of correctly classified land cover types was developed and evaluated in this research. Further, the transferability of developed approach was tested upon the same type of data sets within different study area with similar success – overall accuracy was 87%. Also the limitations found during the investigation procedure are discussed and brief further approach in this field is outlined.

Interpreting a Synthetic Aperture Radar (SAR) image and detecting buildings can be a difficult task visually. In order acquire an overview of an area that has been affected by a disaster, such as an earthquake, SAR is useful due to its independence of weather conditions and the time of the day. GeoRaySAR, a simulator that has been developed by German Aerospace Center (DLR) and the Technical University of Munich (TUM), uses prior knowledge about the geometry extracted, from e.g. a Digital Surface Model (DSM), in order to identify buildings in high resolution SAR data. The simulator has previously utilized DSMs generated from Light Detection And Ranging (LiDAR) data with a vertical and horizontal resolution of 0.1 meters and 1 meter respectively without vegetation. However, DSMs of such high quality is not available everywhere. The objective of this thesis is to evaluate DSMs generated from high-resolution optical data for identifying building in high resolution SAR data in GeoRaySAR. Specifically, images from the spaceborne sensor WorldView-2 have been utilized in this thesis for the extraction of the geometry. The DSMs have been preprocessed in terms of removal of vegetation and reduction of the noise level. The SAR images, acquired from TerraSAR-X, were utilized in GeoRaySAR in order to detect buildings with the assistance of the DSM. An image size limitation that existed in GeoRaySAR has been addressed by adding tiling, which is based on the size of the study scene. Normalized DSM (nDSM) can be determined by calculating the difference between a DSM and a DTM. A nDSM, that received some adjustments, was used as input to GeoRaySAR and compared with the results from the normal DSM. Study areas in three cities, Munich, London and Istanbul, have been used to determine the advantages and limitations of GeoRaySAR and the impact the quality of the DSM has on the building extraction results. The results indicate that building extents can be detected with DSMs generated from optical data with various success, dependent on the quality of the DSM and on which incidence angle the SAR image was acquired in. The ability to interpret a scene increases with the usage of DSMs of higher quality and with SAR images taken in less steep incidence angles. The building DSM depends heavily on the quality of the DTM, but indicates good results and little data loss in study scenes where the DTM successfully removed all objects above ground.