Optimization for viewshed analysis on GPU

Gao, Yong; Yu, Hao; Liu, Yu; Liu, Yuehu; Liu, Mingchao; Zhao, Yong

doi:10.1109/geoinformatics.2011.5980830

Cited by 15 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…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]. …”

Section: Discussionmentioning

confidence: 99%

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

Baek

Choi

2018

IJGI

View full text Add to dashboard Cite

Abstract:We compared three methods for deriving communication viewsheds, which indicate the coverage areas for transmitter points from high-resolution digital surface models. Communication viewsheds were analyzed with a novel 3D Fresnel zone method, as well as line-of-sight (LOS) analysis and 2D Fresnel zone analysis, using high-resolution digital surface models (DSM) from a topographical survey. A LOS analysis calculates a visibility index by comparing the profile elevations of landforms between the transmitter and the receiver, using LOS elevations. A 2D Fresnel zone analysis calculates a 2D Fresnel index by comparing the profile elevations of landforms with the transverse plane elevations of the Fresnel zone. A 3D Fresnel zone analysis quantitatively analyzes communication stability by calculating a 3D Fresnel index, obtained by comparing the elevations of every terrain cell in a Fresnel zone with the total altitude of the Fresnel zone. The latter produced the most accurate results. Indexes derived by applying different transmitter offset heights, signal frequencies, and DSM resolutions for each of the three methods were then quantitatively analyzed. As both the offset height of the transmitter and the signal frequency decreased, the differences between the results derived from each method increased significantly. Moreover, larger DSM cells generated less accurate results.

show abstract

Section: Discussionmentioning

confidence: 99%

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

Baek

Choi

2018

IJGI

View full text Add to dashboard Cite

show abstract

“…This volume of threads fits well with the GPGPU paradigm, which argues for a very high number of independent threaded operations being executed simultaneously over a sustained [10,14]. There has been a concerted effort to discover the potential performance benefits of using the GPU as a viewshed processor [15,12,3], which aims to either modify existing CPU algorithms, or design new algorithms specifically for CUDA hardware; [6] presents a novel algorithm for 'combing' the DEM via thread directions. Whilst the algorithm gained notable performance increases, the speedup relied on a CUDA specific design, particularly optimizing for CUDA warps.…”

Section: Introductionmentioning

confidence: 96%

GPU enabled XDraw viewshed analysis

Cauchi-Saunders

Lewis

2015

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

“…The partitioned data blocks are distributed to these processors to execute currently by means of the existing sequential algorithms, but the performance improvement of these methods is still very limited. 8,17,18 The parallel computing technique based on the GPU is utilized to perform viewshed analysis more efficiently in some cases. 14,16 In recent years, with the development of GPU (graphics processing unit) programming technology, the parallel algorithm based on GPU is widely studied.…”

mentioning

confidence: 99%

“…19 An efficient modified R3 algorithm is designed, which executed on the GPU by a two-level spatial domain decomposition strategy. 17 Although these algorithms have gained a notable increase in computation performance, the speedup ratio of them depends on a specific design and a platform-specific optimization to yield significant gains. 12 The basic interpolation method is mapped into GPU application to achieve high speed by assigning the data into different memory and regularizing the computing instructions.…”

mentioning

confidence: 99%

“…14,16 In recent years, with the development of GPU (graphics processing unit) programming technology, the parallel algorithm based on GPU is widely studied. 8,17,18 The parallel computing technique based on the GPU is utilized to perform viewshed analysis more efficiently in some cases. 19 An efficient modified R3 algorithm is designed, which executed on the GPU by a two-level spatial domain decomposition strategy.…”

mentioning

confidence: 99%

See 1 more Smart Citation

An equal‐area triangulated partition method for parallel Xdraw viewshed analysis

Dou

Wang

2019

Concurrency and Computation

View full text Add to dashboard Cite

This paper presents an effective equal-area triangulated partition method (EATP) for parallel viewshed computation based on Xdraw algorithm that is a viewshed analysis algorithm with lower overhead. Our approach is to first partition the target domain into eight 45 • octants along a coordinate-axis direction and diagonal line direction around a given observer. Each octant remains mutually independent on data dependency based on the principle of Xdraw algorithm.Then, each octant is repartitioned into equal-area sub-regions with the shape of the triangle according to the requirements of system platform such as the number of computing nodes or processors and the size of memory of each node. Our method is beneficial to group these triangle sub-regions for determination of their dependent relationships during the computation of visibility and distribution of each triangle sub-region to a suitable node. The unified expression of the sub-regions is simple for implementation of parallel Xdraw viewshed algorithm. The experimental results are evaluated to demonstrate a distinct improvement in computation performance compared with other partition methods such as equal-angle and simple equal-area methods. KEYWORDS equal-area triangulated partition, parallel computing, terrain visibility analysis, Xdraw viewshed algorithm INTRODUCTIONTerrain visibility between two cells with an elevation value, respectively, denotes whether they are visible each other on a terrain, which is also called inter-visibility analysis. A viewshed in terrain is a particular region of the terrain that is visible to a specified observer on the terrain. The viewshed analysis denotes the process whether the part of the terrain is determined to be visible to a specified observer at a given location on the terrain, which is widely applied spatial analysis to determine the visible regions of the terrain. 1,2 There are many applications in GISs (geographic information systems) including the optimal placement of observers according to visibility requirements, line-of-sight (LOS) communication problems, and terrain path planning with certain visibility properties. 3 However, the viewshed computation for a large-scale terrain viewshed analysis will be very time-consuming because a mass of interpolation operations had to be done in many viewshed analysis algorithms such as R2, R3, Xdraw, 4 reference plane algorithms, and so on. It has been a significant challenge of reducing a large amount of numeric calculation in the existing viewshed analysis algorithms. 5 In particular, the current DEM (digital elevation model) permits the applications to process large terrain with high resolution. 6 Viewshed calculation is usually implemented by emanating a LOS ray from the viewpoint or observer to a target point given an elevation value, respectively. 4,7 Considering the conventional sequential computation approaches are unable to provide effective support, the viewshed computation for large-scale terrain had to entail a massive number of LOS-based calculations. Hence, in ...

show abstract

Optimization for viewshed analysis on GPU

Cited by 15 publications

References 8 publications

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

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

GPU enabled XDraw viewshed analysis

An equal‐area triangulated partition method for parallel Xdraw viewshed analysis

Contact Info

Product

Resources

About