TB级多源遥感影像高效建库方法研究  

Study of Efficient Method of TB-Level Multi-source Remote Sensing Image Database Construction

作　　者：张莉霞[1,2] 江南[1] 胡斌[1] 刘杰[1,3,4] 廖良[1] 杨泽东[1] 

机构地区：[1]南京师范大学虚拟地理环境教育部重点实验室,江苏南京210046 [2]镇江市规划科技信息中心,江苏镇江212004 [3]常州市公安局,江苏常州213022 [4]江苏省公安厅警用空间信息技术重点实验室,江苏常州213022

出　　处：《遥感技术与应用》2013年第3期496-504,共9页Remote Sensing Technology and Application

基　　金：江苏省遥感影像应用服务平台建设(K11110BY8C);地球系统科学数据共享平台--长江三角洲科学数据共享平台;江苏高校优势学科建设工程资助项目

摘　　要：随着各领域对多源遥感影像数据需求的不断增加,如何高效地管理多源海量遥感影像数据,更好地为各行各业服务是遥感数据库研究的热点。目前国内外很多学者在建设多源遥感影像数据库方面进行了相关研究,但对多源遥感影像数据组织、遥感影像数据预处理以及快速索引方面的研究尚存在不足。设计了一种多源遥感影像数据组织模型,实现了多源影像的集成化管理;实现了一种多源海量原始遥感影像(特别是航空影像)的半自动预处理机制,极大地缩短了原始影像数据的处理时间;设计了一种混合检索方法,达到了海量多源遥感影像数据快速高效检索的目的。结果表明:该方法建库效率显著提高,混合索引的查询速度比ArcSDE自动创建的空间索引检索速度有了快速提高,在数据量较小时查询速度至少可以提高4倍,而数据量较大时检索速度至少可以提高12.83倍。With the increasing needs of multi-source image in various fields,it becomes a hot spot that how to efficiently manage multi-source massive remote sensing data. At present, many scholars at home and abroad have researched into construction of multi-source remote sensing image database, but the study of multi-source remote sensing image data organization,the remote sensing image data pre-processing and fast index are still insufficient. This paper designed a multi source remote sensing data organization model, and it can achieve the aim of integrated management of multi-source image. This paper developed a semi auto- matic preprocessing mechanism of multi-source mass image, which greatly reduced the processing time. A hybrid retrieval method was proposed to search image data efficiently. The results showed that the efficiency of building a database significantly improved. According to experiments, the hybrid index greatly speeded retrieval efficiency. When the amount of image data is smaller, the hybrid index at least can improve query speed by 4 times. While the amount of image data is very larger,the hybrid index at least can improve query speed by 12.83 times.

关 键 词：多源遥感影像 数据组织 混合索引 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置] TP311.132[自动化与计算机技术—控制科学与工程]