A second ALOS project has been approved by ESA. It's about human settlement mapping in Africa, in collaboration with JRC and with the interest of Columbia University, CIESIN center.
Title: Human settlements mapping in Africa by SAR
Executive Summary
The main activity of the project consists in developing automated extraction routines for the classification of fine spatial resolution SAR images of structured and unstructured human settlements in various parts of Africa. The aim is to define settlment boundaries and refine the characterization of the settlement typology in a (semi)automatic way. Of course, the existing and already developed algorithms need to be suited for rapid mapping, made cost-effective and based on standard or very simple techniques, possibly already available in commercial off-the- shelf programs. In fact, this is the second part of the work. Expected results are effective urban extent and possibly land cover maps of the human settlement test site(s) using PALSAR data. Maps will be compared where possible with existing or recorded ground truth, like those available trhough th GRUMP project by CIESIN and the AFRICOVER project. Moreover, comparison and combination with optical data from AVNIR-2, as well as information available through X-band or C-band SAR sensors (SRTM, RADARSAT.2, TerraSAR-X) will be considered, where available. A detailed description of the methodology used for fine resolution data analysis and accuracy issues will be provided, to be released to non-expert operators of undeveloped country to improve the usefulness of the data and the availability of tools for their exploitation. Moreover, a detailed comparison of the satellite-based and ground data will be used to define the potential and limits of the proposed approach and to what extent fine resolution and dual polarization SAR data may be used for the proposed task. Finally, the methodology will include an algorithm to provide, as output, a GIS layer in a standard (e.g. ArcView) format, containing all the relevant information extracted, as well as a map for immediate exploitation by the operators and managers on the field. This work will be carried out by team members using internal research funds, but external funds from major grant assigning organization will actively seeked before and during the project lifetime, also in order to assure contunuing the data processing chain started during the project.