存档日期： 存档编号：
本科生毕业设计（论文）
论 文 题 目：基于ENVI的多源遥感影像数据融合
姓 名： 董 建 系 别： 环境与测绘系 专 业： 测绘工程 年 级 、 学 号： 11 测绘 118324107 指 导 教 师： 林 卉
江苏师范大学科文学院印制
摘 要
在世界遥感技术领域极速发展的情况下，对于用卫星传感器来观测和获取某一地域遥感影像数据的方法越来越多，此时多时相、多平台、高光谱和高分辨率卫星等影像数据大量涌现，并且在有关地学的多个领域都用到了这些影像数据。多源遥感影像融合技术在处理怎么使各种有差别的的遥感影像既保存着各自重要的使用特点和对象，同时也会将其局限性缩至最小这方面是最有效的途径之一。并且在不同的学科范畴都涉及到多源遥感影像数据融合技术，因此该技术被不断地完善与推行。而当前在世界上对该技术还尚未形成一套完善的理论与方法，于是目前在遥感领域研究的一个重要的研究技术就是周密处理与剖析遥感信息。
本文基于ENVI对多源遥感影像数据融合的研究。第一章介绍了遥感影像数据融合的基本理论知识和本论文内容研究的背景和意义，以及国内外目前对遥感图像融合技术的发展现状。第二章是对数据融合三个层次（像素级融合、决策级融合和特征级融合）的对比介绍，像素级融合、决策级融合和特征级融合，另外还展示了每个层次数据处理流程图。第三章对像素级融合中加权融合法变换、Brovery 变换、IHS变换和PCA变换等融合方法进行了分析比较。第四章介绍了融合的评价指标（主观评价、客观评价）。第五章通过对Quickbird多光谱影像的4、3、2波段和Quickbird全色影像的1波段数据的进行融合实验，对实验结果进行分析，总结出这四种融合方法的特点和适用范围，得到更准确，更可靠、更安全的估计和判断，为相关工作提供帮助。
关键词：数据融合；ENVI；IHS变换；PCA变换；评价指标；多源遥感影像；
I
Abstract
Fast development in the field of remote sensing technology in the world, with satellite sensors to measure and obtain a regional method of remote sensing image data is becoming more and more this time multi temporal, multi platform, hyperspectral and high-resolution satellite , and in the study of the image data used in these areas. Multi-source remote sensing image fusion technology in dealing with how to make a variety of discriminating the use of remote sensing image is kept their important characteristics and objects, and at the same time its limitations will be shrunk to a minimum is one of the most effective way. And in different disciplines category involves multi-source remote sensing image data fusion technology, thus being constantly perfected and the implementation of the technology. And the current in the world and also with the technology has not yet formed a perfect theory and method, so the current research in the field of remote sensing is an important research techniques careful processing and analysis of remote sensing information.
This article is based on ENVI of multi-source remote sensing image data fusion research. The first chapter introduces the basic theoretical knowledge of remote sensing image data fusion and the content of this thesis research background and significance, as well as the present situation on the development of remote sensing image fusion technology at home and abroad. Second chapter on three levels of data fusion (pixel level fusion and decision level fusion and feature level fusion) contrast, pixel level fusion and decision level fusion and feature level fusion, it also shows the data processing flow chart of each level. The third chapter of weighted fusion method in pixel level fusion transformation, Brovery transformation, IHS transform and PCA transform fusion method are analyzed and compared. The fourth chapter of the fusion evaluation (subjective evaluation and objective evaluation). Fifth chapter through to the Quickbird multispectral image of 4, 3, 2 band and Quickbird panchromatic image 1 band data fusion experiments, analysis of the experimental results, summarized the characteristics of the four fusion method and the applicable scope, get more accurate, more reliable and safer estimates and judgment, to offer help for related work.
Key words： Data Fusion；ENVI；IHS transform；PCA transform；Evaluation;Multi-source remote sensing image
1 II
目 录
摘要…………………………………………………………………………………………………Ⅰ Abstract………………………………………………………………………………Ⅱ
1 绪论…………………………………………………………………………………1 1.1 概述………………………………………………………………………………1 1.2 选题研究的背景与意义…………………………………………………………………1 1.3 国内外研究现状 …………………………………………………………………2
1.3.1 国外研究现状……………………………………………………………2
1.3.2国内研究现状…………………………………………………………3
1.4 论文结构…………………………………………………………………………4 2 多源遥感影像数据融合……………………………………………………5 2.1 层次分类 …………………………………………………………………………5 2.2 像素级融合 ………………………………………………………………………5 2.3特征级融合 ………………………………………………………………………5 2.4 决策级融合 ………………………………………………………………………6 3 影像融合的常用方法………………………………………………………………10 3.1 融合常用方法分类 ……………………………………………………………10 3.2 加权融合法………………………………………………………………………11 3.3 Brovery 变换法 …………………………………………………………………11 3.4 IHS 变换 ………………………………………………………………………11 3.5 PCA变换 …………………………………………………………………………13 3.6方法比较………………………………………………………………………14 4 融合影像质量评价…………………………………………………………………15 4.1 主观评价………………………………………………………………………16 4.2 客观评价…………………………………………………………………………17
4.2.1 均值………………………………………………………………………18 4.2.2 标准差…………………………………………………………………18 4.2.3 信息熵…………………………………………………………………18 4.2.4 平均梯度………………………………………………………………18
1 III
4.2.5 相关系数………………………………………………………………18 5 融合实验数据分析…………………………………………………………………20 6 总结与展望……………………………………………………………………25 6.1 总结…………………………………………………………………25 6.2 展望…………………………………………………………………25 7 致谢 ………………………………………………………………………27
1 IV