基于云计算平台的遥感图像特征选择与优化研究  

Remote Sensing Image Feature Selection and Optimization Based on Cloud Computing Platform

作　　者：黄孝斌[1] 高雪 钱利军[1] HUANG Xiaobin;GAO Xue;QIAN Lijun(The Engineering&Technical College,Chengdu University of Technology,Leshan 614000,China)

机构地区：[1]成都理工大学工程技术学院,四川乐山614000

出　　处：《微型电脑应用》2021年第8期28-31,共4页Microcomputer Applications

基　　金：乐山市重点科技计划项目(19GZD025);湖北省自然资源厅地级城市地质调查试点示范项目(DKC-2018-7-1);成都理工大学工程技术学院青年科学基金(C122018032)。

摘　　要：遥感图像分类一直是人们关注的焦点,针对当前遥感图像分类方法中的特征选择和优化问题,为了提高遥感图像分类效率,提出基于云计算平台的遥感图像特征选择与优化方法。首先对当前遥感图像特征选择与优化的研究进展进行分析,找到不同方法存在的局限性,然后采集遥感图像集,通过云计算平台对遥感图像集进行细分,并采用并行方式进行遥感图像特征选择与优化操作,最后建立遥感图像特征分类器,并与当前经典遥感图像特征选择与优化方法进行了对比实验,结果表明,这种方法加快了遥感图像特征选择与优化速度,并可以提升遥感图像分类效率,具有一定的实际应用价值。Remote sensing image classification has always been the focus of attention.Aiming at the problem of feature selection and optimization in current remote sensing image methods,in order to improve the efficiency of remote sensing image classification,a remote sensing image feature selection and optimization method based on cloud computing platform is proposed.Firstly,the current research progress of remote sensing image feature selection and optimization is analyzed to find out the limitations of different methods.Then,the remote sensing image set is collected,and the remote sensing image set is subdivided through cloud computing platform,and the remote sensing image feature selection and optimization operation are carried out in parallel.Finally,the remote sensing image feature classifier is established.Comparing with the current classical remote sensing image feature selection and optimization methods,the experimental results show that this method can accelerate the speed of remote sensing image feature selection and optimization,and improve the efficiency of remote sensing image classification.

关 键 词：遥感图像分类器 原始特征 特征选择 对比测试 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置] TP391[自动化与计算机技术—控制科学与工程]