A Study on Automated Lineament Extraction with Respect to Spatial Resolution of Digital Elevation Model

The lineament is a linear or curved terrain element to discriminate adjacent geological structures in each other. It has been widely used for analysis of geology, mineral exploration, natural disasters, and earthquake, etc. In the past, the lineament has been extracted using cartographic map or field survey. However, it is possible to extract more efficiently the lineament for a very wide area thanks to development of remote sensing technique. Remotely sensed observation by aircraft, satellite, or digital elevation model (DEM) has been used for visual recognition for manual lineament extraction. Automatic approaches using computer science have been proposed to extract lineament more objectively. In this study, we evaluate the characteristics of lineament which is automatically extracted with respect to difference of spatial resolution of DEM. We utilized two types of DEM: one is Shuttle Radar Topography Mission (SRTM) with spatial resolution of about 90 m (3 arc sec), and the other is the latest world DEM of TerraSAR-X add-on for Global DEM with 12 m spatial resolution. In addition, a global DEM was resampled to produce a DEM with a spatial resolution of 30 m (1 arc sec). The shaded relief map was constructed considering various sun elevation and solar azimuth angle. In order to extract lineament automatically, we used the LINE module in PCI Geomatica software. We found that predominant direction of the extracted lineament is about N15-25°E (NNE), regardless of spatial resolution of DEM. However, more fine and detailed lineament were extracted using higher spatial resolution of DEM. The result shows that the lineament density is proportional to the spatial resolution of DEM. Thus, the DEM with appropriate spatial resolution should be selected according to the purpose of the study.