摘要
快速、准确的获取洪水淹没范围是洪灾损失评估和防洪决策的核心环节.常规种子蔓延算法由于存在大量的递归运算,效率不高且无法处理大区域海量地形数据.针对上述问题,提出了一种大区域洪水淹没范围快速提取的分块种子蔓延算法,将研究区域分成预先设定大小的块,种子蔓延以块为单位进行组织处理,块的数量远小于原始DEM格网单元的数量.该算法有效解决了常规种子蔓延算法递归运算层次过深的问题,可实现大区域海量数据在给定淹没洪水水位条件下的淹没区快速提取.
Fast and accurate data extraction of flood submerged area is the key factor of flood disaster assessment and flood control policy. Because of its great amount of recursive processing, conventional seed spread algorithm has low efficiency and high deficiency in processing large area terrain data. This paper provides a new blocking seed spread algorithm which partitions the original terrain into small blocks with predesigned size. The recursive processing of algorithm is based on the predesigned blocks which are much fewer than the original terrain cells in amount; therefore this new algorithm improves data processing efficiency significantly. It is demonstrated that the proposed approach is able to be employed to extract flood submerged area of large terrain data with given flood water level fast and accurately.
出处
《华中师范大学学报（自然科学版）》
CAS
北大核心
2015年第4期603-607,共5页
Journal of Central China Normal University：Natural Sciences
基金
水利部科技推广计划项目(TG1418)
长江勘测规划设计研究有限责任公司自主科研项目(CX2013Z09)
关键词
数字高程模型
种子蔓延算法
数据分块
淹没分析
Digital Elevation Model （DEM）
seed spread algorithm
data partition
flood analysis
