A key methodology used in the EO4HumEn project is object based image analysis (OBIA), which aims at automating image analysis and developing transferable classification routines.

Over the last decade, new EO techniques and concepts from GIScience have led to the emerging field of OBIA. The main purpose of OBIA in the context of remote sensing applications is to provide adequate and automated methods for the analysis of HR and VHR imagery by describing the imaged reality using spectral, textural, spatial and topological characteristics. OBIA offers a methodological framework for machine-based interpretation of complex classes taking into account multiple properties of image objects. In many application domains, OBIA was pushed by the advent of fine resolution image data providing a high resolution situation, where the pixel size is significantly smaller than the average size of the object of interest. In this case, segmentation as a means of regionalization is an efficient means of aggregating the high level of detail and producing usable objects. Spatial properties like size, form, neighbourhood, context, scale and hierarchy are utilized for better exploiting imagery and other image-like continuous data. Advances in sensor technology and new processing methods (e.g. grid computing) have strongly supported the maturing of OBIA. Today, driven by international programmes and initiatives like GEO (Group on Earth Observation) or GMES/Copernicus, the provision of imagery should no longer be a bottle-neck per se (leaving aside specific requirements such as real time provision, particular detail or atmospheric conditions). The bigger challenge, instead, is information extraction and the provision of added-value products with constant quality, reliability, high transferability and re-usability of algorithms and rule sets as well as a clear validation concept.   

Within the EO4HumEn project OBIA methods are being developed tested and applied for automated extraction of dwelling structures, land use/land cover (LULC) classifications as well as for change analysis. Potential locations for groundwater extraction will be identified based on Earth observation data and auxiliary data. In this context the EO4HumEn project aims at integrating different data sources and developing (semi-automated) methods for identifying relevant geological structures (i.e. lineaments).

The accuracy of the results will be assessed by the producers taking into account ground reference data provided by the user.

The developed products are being provided as maps, online web services, reports and will be fully validated by the user in terms of their relevance and usability.