An IO-efficient parallel implementation of an R2 viewshed algorithm for large terrain maps on a CUDA GPU

Abstract

A rapid and flexible parallel approach for viewshed computation on large digital elevation models is presented. Our work is focused on the implementation of a derivate of the R2 viewshed algorithm. Emphasis has been placed on input/output (IO) efficiency that can be achieved by memory segmentation and coalesced memory access. An implementation of the parallel viewshed algorithm on the Compute Unified Device Architecture (CUDA), which exploits the high parallelism of the graphics processing unit, is presented. This version is referred to as r.cuda.visibility. The accuracy of our algorithm is compared to the r.los R3 algorithm (integrated into the open-source Geographic Resources Analysis Support System geographic information system environment) and other IO-efficient algorithms. Our results demonstrate that the proposed implementation of the R2 algorithm is faster and more IO efficient than previously presented IO-efficient algorithms, and that it achieves moderate calculation precision compared to the R3 algorithm. Thus, to the best of our knowledge, the algorithm presented here is the most efficient viewshed approach, in terms of computational speed, for large data sets.

Comparison of Communication Viewsheds Derived from High-Resolution Digital Surface Models Using Line-of-Sight, 2D Fresnel Zone, and 3D Fresnel Zone Analysis

“…To improve the computational efficiency of 3D Fresnel zone analysis, the following enhancements are required: (1) a solution to the time-consumption problem through the application of the latest viewshed techniques, such as GPU parallel processing [22][23][24]27]; (2) the efficient application of a large high-resolution DSM processing algorithm [29]; (3) the application of a visibility analysis technique which considers vegetation in the DSM [64,65]; and (4) the consideration of the sampling order of cells included in 3D Fresnel zones [66][67][68]. …”

HiXDraw: An Improved XDraw Algorithm Free of Chunk Distortion

“…Although various approximate algorithms run in O (r 2 ) time, it can be unbearably time-consuming to compute a viewshed when handling DEM of high accuracy or of massive scale that is as large as tens of gigabytes [6]. Later researches mainly focus on accelerating the calculation by implementing these algorithms I/O efficiently [6,[9][10][11][12][13][14][15] or redesigning them with various parallel technologies [15][16][17][18][19][20].…”

3D WebGIS: From Visualization to Analysis. An Efficient Browser-Based 3D Line-of-Sight Analysis

“…A special case of ray shooting can be used to compute point-to-point visibility in 2D and 3D [14]. Many visibility analyses in GIS are available for 2.5D data based on raster [18][19][20][21] or TIN structure [22][23][24]. In these cases, for the line-of-sight computation, several restrictions can be observed: (1) observer and target must be on or above the surface; and (2) all participating data layers must be preprocessed into a single digital surface model.…”