高级检索
苏颖, 李梅, 智宁, 毛善君. 煤矿一张图数据均衡非均匀矢量瓦片构建研究[J]. 煤炭科学技术, 2019, (10).
引用本文: 苏颖, 李梅, 智宁, 毛善君. 煤矿一张图数据均衡非均匀矢量瓦片构建研究[J]. 煤炭科学技术, 2019, (10).
SU Ying, LI Mei, ZHI Ning, MAO Shanjun. Research on construction of “one map” data equalization non-uniform vector tile in coal mine[J]. COAL SCIENCE AND TECHNOLOGY, 2019, (10).
Citation: SU Ying, LI Mei, ZHI Ning, MAO Shanjun. Research on construction of “one map” data equalization non-uniform vector tile in coal mine[J]. COAL SCIENCE AND TECHNOLOGY, 2019, (10).

煤矿一张图数据均衡非均匀矢量瓦片构建研究

Research on construction of “one map” data equalization non-uniform vector tile in coal mine

  • 摘要: 针对传统地图瓦片传输煤矿一张图空间数据存在定制化显示困难、数据加载不均衡、时间过长等问题,结合煤矿一张图空间数据特点,在分析矢量地图瓦片划分和渲染过程的基础上,发现数据不均衡现象导致瓦片加载效率低,提出了基于数据均衡的非均匀矢量瓦片构建方法。将瓦片划分分解为初划分和再划分2个部分:初划分时,对数据量小于阈值的区域不再产生瓦片,在地图渲染时将直接调用该区域现存缩放层级最大瓦片;再划分时,对初划分所得瓦片集中不符合要求的瓦片进行四等分,使瓦片数据量更均衡。以此保证地物密集区域瓦片划分更细腻,地物稀疏区瓦片格网划分更粗糙。对比了均匀矢量瓦片、要素简化后的均匀矢量瓦片及基于数据均衡的非均匀瓦片的加载效率。原始数据为356 MB,137个图层。试验结果表明,煤矿一张图数据均衡多尺度瓦片与要素简化后的均匀矢量瓦片对比能够加载速度提升2倍以上,可大幅优化用户体验。

     

    Abstract: Traditional map tiles are widely used for spatial data transmission in coal mine “one map”. However, there have been some problems,such as difficulties in customizing display, unbalanced data loading, and excessive time. Combined with the characteristics of coal mine “one map” spatial data, the vector map tile is divided and analyzed. Based on the rendering process, it is found that the data imbalance is low and the tile loading efficiency is low. In order to solve this problem, a non-uniform vector tile construction method based on data equalization is proposed in this paper. Specifically, the division of tiles is implemented in two steps: initial division and subdivision. When the initial partition is used, the tile is no longer generated for the region whose data amount is smaller than the threshold, and the existing zoom scale maximum tile in the region is directly called when the map is rendered;In terms of the subdivision step, the tile with the largest amount of the set obtained by the initial division is equally divided, so the amount of tiles can be more balanced. In this way, it is ensured that the tile level of the areas of dense features is finer and the tile level of the areas of sparse features is rougher. Furthermore, using the original vector data size with 356 MB and 137 layers, the loading efficiency of uniform vector tile, simplified uniform vector tile and non-uniform tile based on data balance are compared with each other. Experiments have shown that the method proposed in this paper can increase the loading speed by more than two times compared with the method of simplified uniform vector tiles, which can greatly optimize the user experience.

     

/

返回文章
返回