资源简介

摘要： 决策树算法是一种非参数化、非线性的监督分类法。以2010年8月1日Landsat TM影像为基础遥感信息源,以内蒙古自治区赤峰市中部巴林右旗、林西县、克什克腾旗、翁牛特旗交汇处的区域为研究区,通过多次修改完善训练样本数据集,然后把6个原始波段和NDVI、主成分分析后的前3个主分量、常用8个纹理特征以及3个地形特征等共21个特征变量组合成5个不同特征变量组合,采用典型决策树算法C5.0进行了遥感影像分类实验,与最大似然分类结果进行对比。结果表明:C5.0决策树的分类结果优于最大似然结果,尤其是特征变量组合恰当的时候,能够有效利用相关辅助信息,因而最终的分类结果更能满足用户需求。 关键词： 遥感影像;  决策树;  分类;  C5.0     Abstract: The decision tree algorithm is a non\|parametric,nonlinear supervised classification method.This study takes Landsat TM images of August 1,2010 as the basic remote sensing information source,choosing the intersection of Bairin Right County,Linxi County,Hexigten County and Wengniute County,the center region of Chifeng City,Inner Mongolia,China as the study area and by repeatedly modified to improve the training data set,then select twenty\|one characteristic variables combinations whicht consist of five different characteristic variables combinations,and they include six original bands and the NDVI based on them,principal components (PC1,PC2 and PC3),eight texture features (Mean,Variance,Homogeneity,Contrast,Dissimilarity,Entropy and the Second Moment and the Correlation) and three topographical features (DEM,Slope and Aspect),then this study utilizes the typical decision tree algorithm C5.0 to classify them,and the final results were compared with that of maximum likelihood classification results.The result shows that the decision tree classification result is better than maximum likelihood classification result ,especially,when the combinations of characteristic variables are appropriate,which is able to effectively use the related auxiliary information,then their final classification results are more satisfy the users demand.

相关资源

1. 水质遥感监测的关键要素叶绿素a的反演算法研究进展
2. 基于无人机遥感技术的森林参数获取研究进展
3. 基于无线传感器网络的森林生态系统观测试验平台构建
4. 基于无人机激光雷达的森林冠层高度分析
5. 多源时序国产卫星影像的森林火灾动态监测
6. 基于无人机遥感技术的台风灾害倒伏绿化树木检测

下载说明

·如果您发现该资源不能下载，请通知管理员.gissky@gmail.com

·为确保下载的资源能正常使用，请使用[WinRAR v3.8]或以上版本解压本站资源,缺省解压密码www.gissky.net ，如果是压缩文件为分卷多文件，请依次下载每一个文件，并按照顺序命名为1.rar,2.rar,3.rar...，然后鼠标右击1.rar解压.

·为了保证您快速的下载速度，我们推荐您使用[网际快车]等专业工具下载.

·站内提供的资源纯属学习交流之用,如侵犯您的版权请与我们联系.