Extracting Ground-Level DEM From SRTM DEM in Forest Environments Based on Mathematical Morphology

Abstract

The demands for bare-earth or ground-level digital elevation models (DEMs) are significant for various environmental and ecological studies. As one of the most widely used global DEMs, the Shuttle Radar Topography Mission (SRTM) DEM is available to the public. However, the SRTM DEM is not a bare-earth DEM because it includes man-made structures and vegetation. The objective of this paper is to develop a mathematical morphology-based approach to generate ground-level DEM (GLD) from the SRTM DEM in forest environments. The proposed algorithm is implemented as follows. First, an initial GLD is derived from the SRTM DEM by using morphological operations with a single structuring element. Second, homogeneous forest patches are generated by applying watershed segmentation to pseudocanopy height (PCH) that is obtained by subtracting the initial GLD from the SRTM DEM. Based on segmented patches, a refined GLD is derived by using morphological operations with adaptive structuring elements of different sizes. Third, PCH is updated with the refined GLD and then resegmented into forest patches, from which a final GLD is obtained by subtracting the updated mean PCH from the SRTM DEM. Finally, bare-earth DEMs from the National Elevation Data Set (NED) are used to validate the extracted GLD. The results show that the root mean square error of the final GLD compared with the NED is significantly reduced for two study sites. This type of GLD would be applicable to large-scale environmental studies where accurate topographical information is not available. Index Terms-Ground-level digital elevation model (DEM), mathematical morphology, National Elevation Data Set (NED), Shuttle Radar Topography Mission (SRTM) DEM.

TanDEM-X for Large-Area Modeling of Urban Vegetation Height: Evidence from Berlin, Germany

“…Similar to urban-like vegetation, the ability to compare both DTMs to the outcome quality of other studies is limited because few studies deal with DTM filtering in dense forested areas using mathematical morphology and low-resolution raster data. Liu et al [53] used a sequence of morphological operators combined with segmentation to derive a DTM using SRTM data for two forest areas. The authors were able to reduce the RMSE from 23.58 m to 12.59 m and 14.32 m to 7.55 m, respectively, using the outcome compared to a national elevation dataset (NED).…”

Forest structure and degradation drive canopy gap sizes across the Brazilian Amazon

“…The elevation was computed based on the third version of the Shuttle Radar Topography Mission (SRTM in m) provided by NASA JPL with a spatial resolution of 30 m (Farr et al 2007; Liu et al 2014). The SRTM mission occurred in February 2000 and collected data during ten days of operations.…”

Resource availability and disturbance shape maximum tree height across the Amazon

“…The elevation was based on the third version of the Shuttle Radar Topography Mission (SRTM) provided by the National Aeronautics and Space Administration Jet Propulsion Laboratory (NASA JPL) (Farr et al., 2007; Liu et al., 2014). The SRTM mission collected data during 10 days of operations, using two synthetic aperture radars: NASA’s C‐band system (5.6 cm wavelength) and an X band system supplied by DLR (3.1 cm).…”