Earth Observation (EO) based Information about land cover / land use change is currently mostly supported by programmes with 3 to 6 years period. Such update cycle may be considered sufficient for the back-casting application, but has apparent limitations supporting now-casting or even forecasting ones. At the same time new satellite systems as for instance ESA Sentinel-2 or NASA LDCM (Landsat Data Continuity Mission / Landsat 8) will provide high density temporal coverage in very near future (starting 2013) due to short revisit frequency of both constellations (~2 days in future). More, the industry access to the Sentinel and Landsat data will be open and free, which creates new opportunities in the use of the full capacity of the satellites without financial constraints and potentially bring new generation of monitoring services. Such EO constellations allow paradigm shift from periodical mapping to continuous monitoring providing richer trend analysis, more detailed insight into the analyzed processes and great potential for Near Real Time processing and forecasting. Beside that there will be need of huge amount of data processing in order to efficiently use the upcoming databases of EO images.
Nevertheless, this shift generates also new requirements and challenges for the EO data processing and the EO service providers. In order to utilize the full advantage of the new satellites constellation potential, it requires development of new generation of EO processing tools.
The main objective of the project is to evaluate the applicability of selected advanced image processing technologies for continuous urban change monitoring; as for instance continuous adaptive mixture modelling, Markov Random Fields for contextual modelling and Dynamic Bayesian Networks for representing cover type development; driven by high temporal density images as Sentinel-2.