Thematic mapping at regional scale using SIASGE Radar data at X and L band and optical images

This work aims to assess the potential of Synthetic Aperture Radar (SAR) data combined with optical data to support local administrations in the knowledge of the land use and land cover at regional scale. In particular, the contribution of data available in the future through the SIASGE project, combining L-band and X-band radar imagery, is assessed in order to produce thematic maps. Moreover, the further contribution brought by C-band has been evaluated. The classification, focused on two regions in the north side of Italy, is driven by the legend of already existing maps tackling the real needs of the land managing authorities. As the combination of data from optical imagery is fundamental to achieve good thematic accuracy, the work has exploited the Support Vector Machine learning technique, which is more suitable than standard statistical parametric approaches in this respect. Concerning the classification step, some algorithmic issues has been faced to improve the results, such as training set selection strategy and data fusion techniques. The work has proved as the multi source data set (SAR and optical) is fairly suitable to produce thematic maps comparable to what already in use at local administrative level, allowing to obtain reliable maps with a classification accuracy in the order of 90%.