Fusion of hyperspectral images and LiDAR data for civil engineering structure monitoring

Abstract

Investigation of civil engineering materials includes a wide range of applications that requires three-dimensional (3D) information. Complex structures shapes and formations within heterogeneous artificial/natural land covers under varying environmental conditions requires knowledge on the 3D status of the urban materials for better (visual) interpretation of polluted sources. Obtaining 3D information and merge them with aerial photography is not a trivial task. It is thus, strongly needed to develop new approaches for near real time analysis of the urban environment with natural 3D visualization of extensive coverage. The hyperspectral remote sensing (HRS) technology is a promising and powerful tool to assess degradation of urban materials in artificial structures by exploring possible chemical physical changes using spectral information across the VIS-NIR-SWIR spectral region (400–2500nm). This technique provides the ability for easy, rapid and accurate in situ assessment of many materials on a spatial domain within near real time condition and high temporal resolution. LiDAR technology, on the other hand, offers precise information about the geometrical properties of the surfaces within the study areas and can reflect different shapes and formations of the complex urban environment. Generating a monitoring system that is based on the integrative fusion between HRS and LiDAR data may enlarge the application envelop of each technology separately and contribute valuable information on urban runoff and planning. The aim of the presented research is to implement this direction and define set of rules for practical integration between the two datasets. A fusion process defined by integrative decision tree analysis includes spectral/spatial and 3D information is developed and presented.