Efficient viewshed computation on terrain in external memory

Abstract

The recent availability of detailed geographic data permits terrain applications to process large areas at high resolution. However the required massive data processing presents significant challenges, demanding algorithms optimized for both data movement and computation. One such application is viewshed computation, that is, to determine all the points visible from a given point p. In this paper, we present an efficient algorithm to compute viewsheds on terrain stored in external memory. In the usual case where the observer's radius of interest is smaller than the terrain size, the algorithm complexity is θ(scan(n 2 )) where n 2 is the number of points in an n × n DEM and scan(n 2 ) is the minimum number of I/O operations required to read n 2 contiguous items from external memory. This is much faster than existing published algorithms.

More efficient terrain viewshed computation on massive datasets using external memory

“…We also compared our new algorithm TiledVS against our previous one EMViewshed [2]. We used different datasets generated from two distinct USA regions sampled at different resolutions using 2 bytes per elevation value.…”

“…Among the methods for DEM terrains, we can point out the one proposed by Van Kreveld [9], and the one by Franklin et al, named RFVS [6]. These two methods are very efficient and are particularly important in this context because they were used as the base for some very recent and efficient methods for the viewshed computation in external memory: Fishman et al [4] adapted Van Kreveld's method, and Andrade et al [2] adapted the RFVS method. This work also presents an IO-efficient adaptation of the RFVS method.…”

“…The large number of disk accesses is optimized using a new library to manage the data swap between the external and internal memories. This new algorithm was compared against the most recent and most efficient published methods: EMViewshed [2] and io radial2, io radial3 and io centrifugal [4]. Our new method is much simpler and, also, the tests showed that it is more than four times faster than all of them.…”

Terrain Visibility With Multiple Viewpoints

“…I/O-efficient versions of this problem for grid terrains have also been studied recently (e.g. [1,16,19]), as well as other terrain visibility structures like horizons and offsets [17].…”

An efficient GPU multiple-observer siting method based on sparse-matrix multiplication

“…Many applications concern visibility, that is, determining the set of points that are visible from a particular point, called the observer, which can be located at some height above the terrain. These applications include telecommunications, environmental planning, autonomous vehicle navigation and military monitoring [1,6,7,12]. Among these applications, we can point out the siting problem, where the goal is to select a set of observers in order to "optimally cover the terrain".…”