Abstract: The problem statement of disaster risk assessment, based on heterogeneous information (from satellites and in-situ data, and modelling data) is proposed, the problem solving method is grounded and considers its practical use for risk assessment of flooding in Namibia. The basis of the method is the ensemble approach to the heterogeneous data analysis with the use of the data fusion techniques and evaluation the probability density function of a natural disaster using this method.

Keywords: risk assessment, natural disasters, geospatial data, remote sensing, data fusion, ensemble data processing, probability density function, parametric statistics, maximum likelihood classifier, neural network classifier

ACM Classification Keywords: I.5 PATTERN RECOGNITION - I.5.1 Models – Neural nets; G.1 NUMERICAL ANALYSIS - G.1.8 Partial Differential Equations - Inverse problems; F. Theory of Computation - F.1.1 Models of Computation - Probabilistic computation; G.4 MATHEMATICAL SOFTWARE - Parallel and vector implementations; H. Information Systems - H.3 INFORMATION STORAGE AND RETRIEVAL - H.3.5 Online Information Services; I.4 IMAGE PROCESSING AND COMPUTER VISION - I.4.6 Segmentation - Pixel classification; I.4.8 Scene Analysis - Sensor fusion; J. Computer Applications - J.2 PHYSICAL SCIENCES AND ENGINEERING - Earth and atmospheric sciences

Link:

FLOOD RISK ASSESSMENT BASED ON GEOSPATIAL DATA

Nataliia Kussul, Sergii Skakun, Andrii Shelestov, Yarema Zyelyk

http://foibg.com/ibs_isc/ibs-18/ibs-18-p11.pdf