文献类型：期刊文献

中文题名：基于多源遥感的森林地上生物量KNN-FIFS估测

英文题名：Forest Above-Ground Biomass Estimation Using KNN-FIFS Method Based on Multi-Source Remote Sensing Data

作者：韩宗涛[1,2] 江洪[1,4] 王威[3] 李增元[2] 陈尔学[2] 闫敏[2] 田昕[2]

第一作者：韩宗涛

通信作者：Tian, Xin

机构：[1]福州大学地理空间信息技术国家地方联合工程研究中心,空间数据挖掘与信息共享教育部重点实验室,福州350002;[2]中国林业科学研究院资源信息研究所,北京100091;[3]国家林业和草原局调查规划设计院,北京100714;[4]海西政务大数据应用协同创新中心,福州350003

年份：2018

卷号：54

期号：9

起止页码：70-79

中文期刊名：林业科学

外文期刊名：Scientia Silvae Sinicae

收录：CSTPCD;;EI(收录号：20184706117177);Scopus;北大核心:【北大核心2017】；CSCD:【CSCD2017_2018】；

基金：中央级公益性科研院所基本科研业务费专项资金“森林资源动态变化时空连续监测方法研究”(CAFYBB2017QC005)。

语种：中文

中文关键词：KNN-FIFS;特征选择;地上生物量

外文关键词：KNN-FIFS;feature selection;above-ground biomass(AGB)

分类号：S757

摘要：【目的】针对多源遥感数据及其派生特征因子数据维度高、信息冗余、易造成估测模型过拟合等问题,从高维度遥感特征因子中高效优化特征组合,优化区域森林地上生物量(AGB)的k最近邻(k-NN)估测模型。【方法】提出基于快速迭代特征选择的k最近邻法(KNN-FIFS),以森林资源样地调查数据计算的森林AGB为参考,以留一法交叉验证(LOO)相应的k-NN模型反演的森林AGB均方根误差(RMSE)最小为原则,依次迭代选取遥感特征,优化区域森林AGB的k-NN估测模型。以大兴安岭根河森林保护区为研究区,结合Landsat-8 OLI各波段光谱信息、植被指数、纹理、地形因子、机载合成孔径雷达(SAR)P-波段HV极化后向散射强度信息(PHV)以及森林资源样地调查数据,利用KNN-FIFS方法估测研究区森林AGB,并与多元线性逐步回归法(SMLR)进行对比分析。【结果】利用KNN-FIFS方法,得到当k为3,特征组合为PHV、短波红外波段一均一性(H6)、短波红外波段一二阶矩(S6)、短波红外波段二二阶矩(S7)、海蓝波段相关性(Cr1)、近红外波段相关性(Cr5)、海蓝波段相异性(D1)、增强型植被指数(EVI)时,研究区森林AGB估测结果最优,其精度(R^2=0.77,RMSE=22.74 t·hm^(-2))显著优于SMLR估测精度(R^2=0.53,RMSE=32.37 t·hm^(-2))。【结论】KNN-FIFS方法相比SMLR更适用于森林AGB多源遥感估测;KNN-FIFS方法可以从高维度遥感特征因子中高效选取相关特征进行森林AGB估测。
【Objective】Aiming at the over-fitting problem caused by information redundancy from multi-source remote sensing data and their derived high-dimensional features,this study is to effectively pre-select the optimal feature combination to optimize the k-nearest neighbor(k-NN)for regional forest above-ground biomass(AGB)estimation.【Method】This study proposed a fast iterative features selection method for k-NN method(KNN-FIFS).This method iteratively pre-select the optimal features which determined by the minimum root mean square error(RMSE)between the measured forest AGB values and the k-NN estimates based on the leave-one-out(LOO)cross-validation.Based on KNN-FIFS,multi-source data,including Landsat-8 OLI and its vegetation indices,texture metrics,topographic factors,HV polarization of P-band synthetic aperture radar(SAR)data,and forest inventory data(P HV),were used to estimate forest AGB over Daxing’an Mountain Genhe forest reserve located in Inner Mongolia.Afterwards,the model behaviors between KNN-FIFS and stepwise multiple linear regression(SMLR)method were compared.【Result】For KNN-FIFS method,the best configuration was that one with k of 3,the remotely sensed features using P HV,second moment of 1 st and 2 nd short-wave infrared bands(S6,S7),homogeneity of 1 st short-wave infrared band(H6),correlation of coastal aerosol(Cr1),correlation of the near infrared(Cr5),dissimilarity of coastal aerosol(D1)and the enhanced vegetation index(EVI).This configuration generated the most accurate estimates with R 2=0.77 and RMSE=22.74 t·hm-2,which performed much better than SMLR with R 2=0.53 and RMSE=32.37 t·hm-2.【Conclusion】KNN-FIFS is a more suitable method for forest AGB estimation than SMLR.KNN-FIFS can efficiently select the optimal feature combination to estimate regional forest AGB by use of multi-source remote sensing data with high-dimensional information.