摘要
以昆明中心城区遥感影像为研究数据,采用基于面向对象分类的研究方法,按照监督分类的思想,将地物分为城市绿地、林地、裸地、水体、建筑和道路6种类型,运用支持向量机的分类器对绿地信息进行提取,并根据总体分类精度和kappa系数的大小对分类结果进行验证,结果表明:城市绿地信息提取精度良好,符合城市绿地调查的精度标准,认为此种绿地信息提取方法值得推广运用。
The object-oriented classification method was used to study the data of remote sensing image in Kunming central urban area based on the thought of supervised classification. The surface features were divided into six types of urban green space, forest land, bare land, water, building and road. The information of green space was extracted by the support vector machine classifier and the overall classification accuracy and the kappa coefficient were used to verify the classification results. The results showed that the accuracy of information extraction of urban green space by the object - oriented classification method met the standard of the urban green space investigation, which should be applied and promoted.
作者
张夏梦
刘敏
魏开云
ZHANG Xiameng;LIU Min;WEI Kaiyun(School of Landscape Architecture,Southwest Forestry University,Kunming 650224,China)
出处
《林业调查规划》
2019年第2期18-22,共5页
Forest Inventory and Planning
关键词
面向对象
监督分类
遥感影像
绿地信息
分类精度
kappa系数
object-oriented
supervised classification
remote sensingimage
information of green space
classification accuracy
kappacoefficient
