面向关系数据库扩展的自适应影像金字塔模型

Relational Database Extension Oriented,Self-adaptive Imagery Pyramid Model

随着遥感技术的发展,特别是传感器分辨率的提高,影像数据量日益增长,这对高效、智能地管理海量数据提出了更高的要求,如何有效、智能地存取海量遥感数据成为当前越来越热门的话题。本文针对空间数据管理系统的发展现状,结合数据库存储、网络传输和客户端的硬件情况,从影像数据切片和影像数据的细节层次(level of detail,LOD)模型两个层面来解决遥感数据的存储问题,提出了自适应的遥感影像切片机制,以及与之相适应的LOD模型构建方法。通过试验证实,该影像管理机制能够在多种不同的数据库、网络和客户端条件下实现影像数据智能化高效存储与访问。本文的研究为高效的影像数据管理提供了可行的思路和方法,有助于在C/S架构的网络环境下实现基于数据库技术的遥感影像数据的高效存取和管理。

With the development of remote sensing technology,especially the improvement of sensor resolution,the amount of image data is increasing.This puts forward higher requirements to manage huge amount of data efficiently and intelligently.And how to access massive remote sensing data with efficiency and smartness becomes an increasingly popular topic.In this paper,against current development status of Spatial Data Management System,we proposed a self-adaptive strategy for image blocking and a method for LoD(level of detail)model construction that adapts,with the combination of database storage,network transmission and the hardware of the client.Confirmed by experiments,this imagery management mechanism can achieve intelligent and efficient storage and access in a variety of different conditions of database,network and client.This study provides a feasible idea and method for efficient image data management,contributing to the efficient access and management for remote sensing image data which are based on database technology under network environment of C/S architecture.