Earth Observation based information about the land cover / land use change is currently mostly supported by monitoring programmes working with the period of several years. 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, e.g. Sentinel-2 or Landsat 8, start to provide EO datasets with high density temporal coverage. More, the industry access to the Sentinel and Landsat data is open and free, which creates new opportunities in the use of the full capacity of the satellites without financial constraints. This allows paradigm shift from EO based periodical mapping to continuous monitoring providing richer trend analysis, more detailed insight into the analysed processes and great potential for Near Real Time processing and forecasting. To efficiently use the upcoming databases of satellite images huge amount of data shall be processed and new generation of EO processing tools needs to be developed.
The specific goal of the project is to explore the applicability of selected advanced image processing technologies for the use within continuous urban expansion monitoring. Up to now, change detection approaches in EO are limited to typically bi-temporal analysis for two points in time. Another approach, the temporal trajectory concept is highly relevant for vegetated land covers as it assumes continuous seasonal development. Urban expansion monitoring needs different algorithms. Continuous image acquisition opens potential for reduction of the false change detection as well as potential of high automation in operational use. It is a challenge to separate variation between consequent images that represents changes of interest, which are typically fragments of the total area. The selected approaches keep high potential of solving the currently existing constraints that are objective reasons for low automation of the change detection procedure (besides the temporal limits). The proposed advanced technologies of image processing proved to be highly successful in other application domains (such as bioinformatics, medicine, real-time tracking, document classification, gaming, etc.).
The basic concept of the continuous urban expansion monitoring reflects the current status of operational application enhanced by the new satellite systems. The crucial aspect of the “Urban Monitor processor” is to provide robust change detection in continuous image acquisition environment. The robustness of the processor can be measured by the accuracy assessment, while the challenging aspect is full transferability without the need of modification. The project builds on three different approaches to modeling and detecting temporal changes of urban land cover.
The combination of these approaches will allow to (i) account for priors of possible changes between the types of land-covers, (ii) include spatial coherence of land-cover classes, and (iii) locally model the variation of appearance due to external conditions (atmospheric, sun illumination, seasonal changes and others).
GISAT has been selected by the World Bank as the consultant providing services for the development of web based geospatial software Platform for Urban Management and Analysis (PUMA). The first generation of the PUMA platform consists mainly of the analytical tools based on the pre-processed bi-temporal change layers. It is intended to enhance the capability of the platform by automated continuous change detection using the time series of Sentinel-2 / Landsat 8 images. The developed “Urban Monitor processor” will be ingested into PUMA platform, which enhances the capability of the whole concept by rapid operational urban expansion monitoring. Successful implementation of the processor would significantly contribute to the demonstration of effectiveness of EO data inclusion into regular World Bank urban related internal work-flow by providing flexible, on-demand and cost-effective city monitoring capacity with the potential of large number of cities to be mapped around the world.
Urban Dynamic Monitor project is funded by the ESA Innovation Triangle Initiative (ITI) Programme. The project is based on collaboration of GISAT s.r.o. (Developer) and Czech Technical University in Prague (Inventor).
Gisat provides wide range of geoinformation services based on Earth Observation technology. It focuses on operational application of satellite mapping to monitor various aspects of our environment and development of dedicated web based platforms for geoinformation analysis and assessment
Web // E-mail // Tel:+420 271741935 // Fax: +420 271741936