D6.4 Novel methodologies for damage detection and assessment along the CH assets and the surrounding disaster affected area

The deliverable 6.4 focuses on the description of the novel methodologies for damage detection and assessment along the CH assets and the surrounding disaster affected area. The developed methodologies were evaluated on areas where natural disasters have occurred. The methodologies are applied according to a flowchart that provides the step-by-step approach for damage assessment in the broader area and the CH asset as well. Three flowcharts have been defined for three different disaster scenarios: flood, landslide and earthquake (chapter 2). In chapter 3, the developed methodologies are described. The flood detection methodology is based on the detection of changes due to flood events exploiting the backscattering change between the pre-flood images and the flood image. The method consists of four main steps: a) data procurement b) pre-processing c) time series analysis and d) classification. The landslide detection methodology is based on ESA SNAP polarimetric functionalities for the delineation of a landslide. Entropy (H), Alpha (a) and Anisotropy (A) decompositions are calculated. An RGB composite based on the multitemporal ratio (before and after the event) is calculated. The delineation of the landslide is based on visual interpretation of the RGB composite. The adopted technique for earthquake induced deformation is based on ESA SNAP functionalities for the detection of millimeter co-seismic surface displacements. Upon estimating the co-seismic ground deformation from ascending and descending geometries, a clear signal is identified and ground deformation is locally estimated. Finally in chapter 4, the sensor strategies for damage assessment along the CH assets after disasters are discussed. Damage assessment includes the estimation of structural deformation as well as the detection of acute material deterioration. Τhe consequences of limited sensor payload of currently available drones suitable for mobile deployment, and ways to balance information needs against abilities of different drone types and regulatory limitations are discussed in this chapter.