摘要
遥感在森林资源调查中得到越来越广泛的应用,无疑是一种方便快捷地获取各类林业信息的新手段。结合新兴的数据挖掘技术对患病树木鉴别更是一种具有现实意义的检测方法。文章对原始数据进行数据预处理之后得到用于挖掘使用的数据集,通过支持向量机和随机森林两种数据挖掘方法建立患病树木检测模型,对比其分类效果,最终确立出针对森林健康情况判别的最优模型。本研究为相关部门对树木健康的检测监控提供参考,进而合理优化森林树木管理,提升森林虫害防治的研究效率。
Remote sensing has been applied more and more widely in forest resource survey.It is undoubtedly a new means to obtain all kinds of forestry information conveniently and quickly.Combined with the emerging data mining technology for the identification of diseased trees has become a practical detection method.In this paper,the data set used for mining is obtained after data preprocessing of the original data,and the detection model of diseased trees is established by two data mining methods,namely Support V ector Machine and Random Forest.After comparing the classification effect,the optimal model for forest health discrimination is finally established.This study provides reference for relevant departments to monitor and detect the health of trees,so as to rationally optimize the management of forest trees and improve the research efficiency of forest pest control.
作者
岳丽娅
邓洁莹
梁霄
YUE Liya;DENG Jieying;LIANG Xiao(School of Mathematics and Statistics,Hubci University of Arts and Science,Xiangyang,Hubei 441053,China)
出处
《计算机应用文摘》
2023年第1期120-123,共4页
Chinese Journal of Computer Application
关键词
数据挖掘
患病树木检测
随机森林
支持向量机
data mining
detection of diseased trees
Random Forest
Support Vector Machine
