HiVision: Rapid visualization of large-scale spatial vector data

Ma, Mengyu; Wu, Ye; Ouyang, Xue; Luo, Chen; Zhang, Liangpei; Jing, Ning

doi:10.1016/j.cageo.2020.104665

Cited by 6 publications

(9 citation statements)

References 31 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It enables fast discrimination of the above spatial relationship by constructing an R-tree spatial index on the vector data. HiVision outperforms traditional methods (e.g., HadoopViz [16], GeoSparkViz [17] and Mapnik [18]) and provides real-time visualization of large-scale geographic vector data [15]. However, in the practical application, we found that with the increase of the scale of vector data, the method has the problems of long index construction time and large index memory occupation, which affect the popularization and application in practice.…”

Section: Introductionmentioning

confidence: 89%

“…Compared with the existing visualization method based on DisDC (HiVision) and traditional methods (HadoopViz, GeoSparkViz, and Mapnik), HiIndex significantly improves the efficiency of data pre-processing and visualization. To support real-time visualization of large-scale geographic data, parallel computing technologies are used to accelerate computation, and we extend the high-performance parallel processing architecture in HiVision [15].…”

Section: Methodsmentioning

confidence: 99%

“…Display-driven computing (DisDC) is a computing model that is especially suitable for data-intensive problems in GIS, which has a broad prospect of researches and applications in big data analysis [2]. In our previous works [15,19,20], we have successfully applied DisDC and solved some common spatial analysis problems in GIS. We have proposed HiVision, HiBuffer, and HiBO based on DisDC to realize the interactive visualization, spatial buffer analysis, and spatial overlap analysis of large-scale vector data.…”

Section: Display-driven Computingmentioning

confidence: 99%

“…A root node is created firstly, and then new child nodes are created based on whether the vector object intersects the recursively divided spatial range of the root node. In TQTG, vector objects are divided into small fragments [15]: for point objects, a point object is directly used as a fine-grained item; for line objects, the MBR of each line segment is taken as a fine-grained item; for polygon objects, take the MBR of each edge as a fine-grained item. The details of TQTG for point are shown in Algorithm 1, the details of TQTG for linestring or polygon edges are shown in Algorithm 2.…”

Section: Tqtg For Point Linestring and Polygon Edgesmentioning

confidence: 99%

“…We have designed a rapid visualization method called HiVision [15]. It enables fast discrimination of the above spatial relationship by constructing an R-tree spatial index on the vector data.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

HiIndex: An Efficient Spatial Index for Rapid Visualization of Large-Scale Geographic Vector Data

Liu

Chen

Yang

et al. 2021

IJGI

Self Cite

View full text Add to dashboard Cite

In the big data era, rapid visualization of large-scale vector data has become a serious challenge in Geographic Information Science (GIS). To fill the gap, we propose HiIndex, a spatial index that enables real-time and interactive visualization of large-scale vector data. HiIndex improves the state of the art with its low memory requirements, fast construction speed, and high visualization efficiency. In HiIndex, we present a tile-quadtree structure (TQ-tree) which divides the global geographic range based on the quadtree recursion method, and each node in the TQ-tree represents a specific and regular spatial range. In this paper, we propose a quick TQ-tree generation algorithm and an efficient visualization algorithm. Experiments show that the HiIndex is simple in structure, fast in construction, and less in memory occupation, and our approach can support interactive and real-time visualization of billion scale vector data with negligible pre-treatment time.

show abstract

Section: Introductionmentioning

confidence: 89%

Section: Methodsmentioning

confidence: 99%