The viewshed analysis functionality allows the user to perform a viewshed analysis directly from one or more Scalable Terrain Model (i.e : STM) elements. The full resolution data contains in the STM element(s) overlapping the source area (i.e. : terrain area consider during the analysis) is used during the viewshed analysis, which limits the size of the source area (see Extract A TM From STM functionality). Note that the viewshed analysis is done on the TIN representation of the terrain which results in increased precision of the analysis compared to doing the analysis on a DEM raster representation, at the expanse of longer computation time.
The result of the viewshed analysis is a raster grid whose pixel value represents the number of times the region encompassed by a given pixel was visible from the viewer position during the analysis. Currently there is only one output file format supported, which is the iTiff. Note that the raster grid created is geo-referenced to ensure that the raster grid is located over the source area.
There are two types of viewshed analysis that can be performed with Bentley Descartes SELECTSeries 4 : from a static position and along a path.
The static position is useful when the viewshed analysis is performed only for one position. In that case the pixel value of the resulting raster grid can only take two values with this type of analysis, 0 (i.e. : not visible) or 1 (i.e. : visible).
With this type of analysis the viewshed is performed for one or more positions along the path, the number of positions depending on the length of the path and the increment specified.
Here is a video showing a viewshed analysis along a path and the resulting raster grid.
Performance Tips : Limiting the source area to only the region for which the analysis is required will increase the speed of the analysis. Since the source area can be any kind of closed polygon it might be worth taking a bit more time to define the region precisely instead of using a simple block.
Performance Tips : Using a very small pixel size can tremendously diminish the speed of the analysis, especially when performing an analysis along a path.