摘要就目前的情形而言，遥感技术在近几十年来不断的提升发展，光谱分辨率的上升也渐渐形成了一种趋势。在已经过去的几十年中，因为人们的不懈努力，在对地观测方面所取得了很大进展和不少的技术突破，而高光谱遥感技术就是这些突破之一。由于高光谱分辨率的性质，人们对其暗藏的可应用性有所关注。分类就是取得高光谱信息的有效手段之一。80674

本文对高光谱遥感技术的概念进行了一定的描述，不仅介绍了高光谱图像的分类原则和分类流程，还阐述了高光谱遥感图像的统计分类方法。其主要内容包括：高光谱遥感图像监督分类下基于距离判决函数（分类器）的统计分类方法研究；比较分析了枣园乡图像分类方法的效果。同时也对图像分类的研究现状进行了必要的说明。

毕业论文关键词：高光谱遥感 图像分类 分类方法 

 Abstract  On the current situation, the remote sensing technology in recent decades, constantly enhance the development, increased spectral resolution has gradually formed a trend。 In the last few decades, because people's unremitting efforts in the area of Earth observation has made great progress and a lot of technological breakthroughs, and hyperspectral remote sensing technology is one of those breakthroughs。 Due to the nature of high spectral resolution, it is generally hidden to focus on its applicability。 Classification is one of the effective means to achieve high spectral information。

    In this paper, the concept of hyperspectral remote sensing technology for a certain description, not only describes the principles of hyperspectral image classification and classification processes, statistical classification of hyperspectral remote sensing image。 The main contents include: high-spectral remote sensing image supervised classification based on distance discriminant function (classification) of the statistical classification method based on IDL language studies and computer-implemented; compare hyperspectral remote sensing image classification method effects。 Also on the research status of image classification was necessary instructions。

Key words： Hyperspectral Remote Sensing，Image Classification，Classification

目 录

摘要 I

Abstract II

1。绪论 1

   1。1研究的目的和意义 1

   1。2遥感图像研究现状 1

2。遥感图像分类原则流程 2

   2。1高光谱遥感基本概念 2

   2。2高光谱图像分类的原则 2

   2。3高光谱图像分类的流程 2

3。图像分类 5

   3。1监督分类 7

       3。1。1监督分类方法 7

   3。2非监督分类 7

       3。2。1非监督分类方法 7

   3。3面对对象分类 7

       3。3。1面对对象分类方法 7

4。高光谱遥感图像分类算法研究 8

4。1分类方法实现 8

4。2分类效果评价 14

5。结论