遥感图像数据库是小城镇信息化建设中的重要组成部分，用于存储、管理和检索遥感图像相关的各种数据。近年来，随着遥感技术和计算机科学的快速发展，作为一种重要的信息资源，遥感图像数据量迅速增加，基于文本的图像检索技术已经无法适应海量数据检索的需要。特别是如今，Internet技术已开始进入Web2.0时代，如何准确地、高效地检索这些遥感图像，并以网络的方式实现它们的发布与共享，成为小城镇信息化建设中的一个重要问题。为了解决这个问题，作者尝试结合网络信息发布与共享技术，实现基于内容检索的遥感图像数据库系统。本文对基于内容的图像检索的发展、关键技术及其效果评价进行了概述，其中，图像特征的提取是基于内容图像检索的基础，本文分别从颜色特征、纹理特征和形状特征等几方面研究了图像内容的描述方法。遥感图像较适合提取其纹理特征，因此，本文着重讨论了基小波为二维Gabor函数的小波变换来提取图像纹理特征的方法，并对整个J2EE架构中的关键技术进行了全面的阐述，提出了适合遥感图像数据的基于内容检索的系统框架。最终，在Windows XP操作系统平台下，采用J2EE架构和Oracle数据库设计并实现了一个B/S结构的基于图像内容检索的遥感图像数据库系统，并对系统的性能进行了评价，证明了方法的可行性。

Remote sensing image database, which is used for storing, managing and searching various data relevant to remote sensing image, constitutes the important part of the small town informatization construction. In the recent years, as an important kind of information resource, the amount of remote sensing image data is increasing rapidly along with the development of remote sensing technology and computer science, so the Text-based Image Retrieval (TBIR) skill can not meet the need of searching the large amount of data. Especially nowadays, the internet technology has begun to be the Web2.0 times. How to exactly and efficiently retrieval these images and how to through network release and share them become an important issue of the small town informatization construction.To solve this problem, we try to implement a remote sensing image database system of content based retrieval, combining with the technology of network information release and share. This paper presents the development of Content-based Image Retrieval (CBIR), the key technology of CBIR and evaluation of CBIR. Thereinto, image feature extraction is the basis of CBIR. This paper researches the description methods of image content form the aspects of color feature, shape feature, texture feature and so on.Remote sensing image is fit to extract its texture feature relatively, so this paper mainly discuss the method of image texture feature extraction of two-dimension Gabor wavelets transform coefficients. Totally illustrate the key technology in the whole J2EE and take out the content-based retrieval system framework which is proper to remote sensing image data. Ultimately, under the Windows XP platform, design and implement a CBIR remote sensing image database system of B/S structure, using J2EE framework and Oracle database. And at last evaluate the system’s performance to prove the feasibility of this method.