Soil and Water Research (Jun 2008)
Creating of digital surface model and orthophoto from ASTER satellite data and their application in land/water research
Abstract
Satellite data has become a commonly used information source. Landscapes components such as water, inorganic substances, vegetation, and the atmosphere may be distinguished making use of their spectral characteristics. The above mentioned components may be further divided. For example, inorganic substances may be subdivided into soil, minerals, build up areas etc. The spectral characteristics of soils are determined by moisture, humus contents, mineral composition, surface structure, and the stage of eroding processes. The development in remote sensing tends either to the data acquisition in more spectral bands or the improvement of the resolution of remote sensing data. The terra satellite ranks among new generation satellites; its orbital parameters are similar to the parameters of the Landsat system. ASTER (Advanced Spaceborn Thermal Emission and Reflection Radiometer) is one of the onboard instruments on Terra satellite and captures data in 14 spectral bands. The VNIR (Visible Near Infrared) subsystem provides 15 m spatial resolution data. Two of the VNIR subsystem telescopes enable stereoscopic data evaluation. A stereo-pair consists of 3N (nadir) and 3B (backward) images. A couple of 3N and 3B images can be used for the creation of a digital surface model (DSM) and orthophoto. This article describes the creation of DSM and orthophoto of an area located in the north-west part of the Czech Republic. Images of the area were made in years 2002 and 2005. In this work, level 1B images were used, i.e. images with radiometric and geometric corrections already applied. The model was created through the use of 21 control points selected in each scene. The standard error of co-ordinates of the control points is up to 15 m, the elevation standard error is approx. 30 m. The accuracy of the final DSM and orthophoto was tested on a set of 13 check points. The position standard error in DSM and orthophoto is approx. 15 m, i.e. just about the size of one pixel of the original data. The elevation standard error of the checkpoints is up to 40 m. The output can be used as a basis for small-scale maps. Using one scene acquired by ASTER instruments, a DSM and orthophoto covering an area of 60 × 60 km can be created. Keywords: remote sensing; ASTER; digital surface model; orthophoto
Keywords
