中国农业科学院农业信息研究所 中国农学会
Agricultural Information Institute of CAAS China Association of Agricultural Science Societies
En
Ch
中国农业科学院农业信息研究所 中国农学会 主办

AI检索

订阅

返回顶部
当前位置: 首页 > 文章 > 基于k-NN方法和GF遥感影像的森林蓄积量估测 浙江农林大学学报 2017,34 (3) 406-412
Position: Home > Articles > Forest stock volume estimation based on the k-NN method and GF remote sensing data Journal of Zhejiang A&F University 2017,34 (3) 406-412

基于k-NN方法和GF遥感影像的森林蓄积量估测

作  者：
向安民;刘凤伶;于宝义;李崇贵
单  位：
国家林业局;西安科技大学
关键词：
森林经理学;k-NN方法;蓄积量估测;最小二乘估计;稳健估计
摘  要：
综合利用黑龙江省某林业局的一类样地调查资料、GF-1号卫星影像、数字高程(DEM)模型以及土地利用类型图,采用k-近邻(k-nearest neighbor,k-NN)法进行森林蓄积量估测研究,分析k-NN方法及GF-1卫星数据在森林资源调查与监测中的应用效果。为对比k-NN方法的估测精度,对相同试验数据也进行了最小二乘估计和稳健估计建模。采用GF-1号16 m分辨率的多光谱数据,在林业局级尺度上分别应用这3种方法进行森林蓄积量建模估测,生成了监测区域森林蓄积量分布图并统计得到监测区域总的蓄积量值。将3种方法估测结果与二类调查实测结果进行比较,k-NN方法估测精度达到97.3%,略优于传统的最小二乘估计和稳健估计建模估测精度。因k-NN方法不受Gauss-Markov假设限制,且能有效克服建模变量间的复共线性问题,研究成果可用于县/林业局级尺度的森林蓄积量估测,且国产GF-1卫星影像能有效应用于森林资源监测。
译  名：
Forest stock volume estimation based on the k-NN method and GF remote sensing data
作  者：
XIANG Anmin;LIU Fengling;YU Baoyi;LI Chonggui;Northwest Institute of Forest Inventory,Planning and Design, State Forestry Administration;College of Geomatics, Xi'an University of Science and Technology;
关键词：
forest management;;k-NN method;;stock volume estimation;;least squares regression;;robust regression
摘  要：
To analyze application results of the k-nearest neighbor(k-NN) metho d and Gao Fen-1(GF-1)satellite data in forest resources investigation and monitoring, sample plot data of national continuous forest inventory, GF-1 satellite images, digital elevation model(DEM), and land utilization type pictures from a Forestry Bureau in Heilongjiang Province were used. To compare estimation accuracy of the k-NN method, least squares regression and robust regression were used based on the same test data. By using GF-1 satellite images of 16 m resolution wide field view(WFV) Multi-spectral data, models based on k-NN, least squares regression,and robust regression, models were built with a map showing stock volume distribution. Results comparing the estimated stock volume and the survey value showed that the overall accuracy for the forestry bureau scale was over 90% with performance of the k-NN method being 0.4% higher than the least squares regression and 0.2%higher than robust regression methods. Because the k-NN method was neither limited by the Gauss-Markov hypothesis nor the effects of the Multi-collinearity between the modeling variables, the research results could be used for county or Forestry Bureau scale forest stock volume estimations, and domestic GF-1 satellite images could be effectively applied to forest resource monitoring.

相似文章

导出引用
计量
文章访问数: 12
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊

友情链接Friendly link
农业信息研究所Agricultural information institute 农业智能知识服务平台Agricultural intelligent knowledge service platform 农业学术服务平台Agricultural Knowledge Service System 中国知网CNKI 维普网VIP 万方WANFANG DATA
联系我们Contact us

电话Tell：010-82109652-808

邮箱E-mail：agrijournal@caas.cn

服务Service
关于我们About us 意见建议Opinions and suggest
版权说明Copyright
单位：中国农业科学院农业信息研究所 中国农学会 © 2017中国农业科学院农业信息研究所 ICP备案号：京ICP备09089781号-46