批量遥感影像湖泊提取后的矢量拼接策略问题  

Vector Data Union Policy of Lakes Extracted from Multi Remote Sensing Images

作　　者：沈占锋[1] 李均力[2] 夏列钢[1] 吴炜[3] 骆剑承[1] 

机构地区：[1]中国科学院遥感与数字地球研究所,北京100101 [2]中国科学院新疆生态与地理研究所,新疆乌鲁木齐830001 [3]浙江工业大学计算机科学与技术学院,浙江杭州310023

出　　处：《武汉大学学报（信息科学版）》2015年第4期444-451,共8页Geomatics and Information Science of Wuhan University

基　　金：国家863计划资助项目(2013AA12A401);国家科技支撑计划资助项目(2012BAH33B01;2011BAH06B02);NSFC-新疆联合基金资助项目(U1178302)~~

摘　　要：在遥感影像湖泊提取结果进行矢量化之后,需要对多景影像的湖泊提取矢量数据进行拼接并形成湖泊信息专题图,不同的拼接策略直接决定区域湖泊数据拼接的效率。结合湖泊提取结果的矢量数据拼接与专题图制作问题,在分析矢量数据拼接原理与过程的基础上,首先分析了两景矢量数据的拼接方法与最佳策略,重点研究了面向大区域多景矢量数据的拼接策略,提出了两种实用的高效矢量数据拼接方法,即基于任务队列与基于间隔选择的策略,分析了不同策略的特点与适用情况。实际应用中,可以根据实际情况进行选择,较大程度地提高矢量数据拼接的效率。Efficiency is an important issue that must be considered in the union process of the regional vector data.After the vectorization of the remote sensing images of lakes extraction result,the extracted lakes from multi-scene images should be merged to a lakes thematic map,and different union policies will directly determine the efficiency of the regional lake data union.Focused on the key technique of vector data union and lake thematic map production,this paper firstly analyzes the premium union policy for two scene vector data,then two different union policies for multi scene vector data are proposed,that is,the policy based on task queue and the policy based on interval selection.Furthermore,the characteristics of both policies are then analyzed and we can select different policies for the different situations so as to gain more efficiency.

关 键 词：矢量拼接 湖泊 策略 任务队列 间隔选择 

分 类 号：P237.3[天文地球—摄影测量与遥感] P208[天文地球—测绘科学与技术]