基于HBase与静态多级格网索引的地表覆盖数据高效检索方法

An efficient retrieval method of land cover data based on HBase and static multi-level grid index

地表覆盖是地理国情监测的重要对象,为地理国情分析评价模型提供了可靠的数据源.高效的地表覆盖数据检索方法是挖掘地表覆盖数据潜在价值的前提.由于地表覆盖数据体量庞大、更新频繁,要素分布密集且不均匀,传统的空间检索方法出现了扩展困难、检索能力不足等问题.提出了一种基于HBase与静态多级格网索引的地表覆盖数据空间检索方法,针对地表覆盖数据特征设计了基于HBase的静态多级格网索引,利用MapReduce实现索引并行构建,通过多级过滤的方式,提高了地表覆盖空间范围的查询效率.实验表明,该方法能快速完成大规模、密集分布的地表覆盖数据的空间索引构建,提升空间检索性能,并具有良好的扩展性,可为其他海量空间矢量数据的检索提供借鉴.

As an important object of National Geographic State Monitoring,land cover provides national geographic analysis and evaluation model with reliable data source.Efficient retrieval methods are essential for mining the potential value of land cover data.Due to the huge amount of the data as well as the frequent updating and the dense uneven distribution of land cover elements,there have been problems on scalability and retrieval cap ability for the traditional spatial retrieval methods.This paper puts forward an improved spatial retrieval method of land cover data based on HBase and static multi-level grid index.The new approach allows parallel index construction with MapReduce and improves the query efficiency of land cover spatial range through multi-level filtering.Experiments show that this method can quickly complete the spatial index construction of the large-scale and densely distributed land cover data,improve the performance of spatial retrieval and have good expansibility,which can provide reference for the other mass spatial vector data retrieval.