基于最小距离法的遥感图像分类  

Remote Sensing Image Classification Based on Minimum Distance Method

作　　者：冯登超[1] 陈刚[1] 肖楷乐[1] 杜文雅[2] 吴新颖[1] 

机构地区：[1]北华航天工业学院电子工程系,河北廊坊065000 [2]太原科技大学电子信息学院,山西太原030024

出　　处：《北华航天工业学院学报》2012年第3期1-2,5,共3页Journal of North China Institute of Aerospace Engineering

基　　金：河北省自然科学基金项目(F2011409008);河北省教育厅青年基金项目(2010208);北华航天工业学院博士基金项目(KY-2008-02-B)

摘　　要：本文在ENVI遥感图像分析软件环境下,利用最小距离法对遥感图像进行了有监督分类。仿真实验通过对遥感图像中地物目标的光谱信息和空间信息进行分析及特征提取,利用判别函数计算到中心的距离,解读遥感图像,并对分类结果进行定量评估。实验结果表明用最小距离法来分析遥感图像可以获得较高精确度。In this paper, the supervised classification technology based on minimum distance method was realized to classify the remote sensing image with the support of ENVI development environment. Feature analysis and extraction of the spectrum and spatial infor- mation were completed in the simulation experiment. The discriminant function was adopted to calculate the distance of center in or- der to explain the remote sensing image. The higher accuracy was proved by the quantitative evaluation of the classification result. The experiment results show the validity of the method.

关 键 词：遥感 ENVI 有监督分类 最小距离法 

分 类 号：P237[天文地球—摄影测量与遥感]