基于优化k-NN模型的高山松地上生物量遥感估测
|
摘要
针对传统k-最近邻法(k-nearest neighbor,k-NN)在搜索最近邻单元时赋予特征变量相等的权重,缺少对特征变量加权优化等不足问题,在云南省香格里拉市,以高山松Pinus densata为研究对象,基于49块实测标准地,116株高山松样木和Landsat 8/OLI影像,在前期进行基于遗传算法(genetic algorithm,GA)优化的k-NN模型实现的基础上,对k-NN的3个参数(k,t和d)进行反复测试优化组合,在像元尺度上对研究区高山松地上生物量进行遥感估算。结果表明:基于遗传算法优化的k-NN模型精度优于传统的k-NN模型,优化前均方根误差为30.0 t·hm~(-2),偏差为-0.418 t·hm~(-2),相对标准误差百分比(R_(MSE))为54.8%;优化后均方根误差为24.0 t·hm~(-2),偏差为-0.123 t·hm~(-2),R_(MSE)为43.7%。基于优化k-NN模型的研究区高山松地上生物量总储量估测结果为0.89×10~7t。图7表6参20
For the traditional k-nearest neighbor(k-NN),there are insufficient problems that give the weight of the feature variables equally when searching the nearest neighbor population units and a lack of weight vectors for the feature variables.In this study,Shangri-la City,Yunnan Province,was selected as the research area,and Pinus densata was taken as the research object.Based on 49 field data plots,116 P.densata data samples,and Landsat 8/Operational Land Imager(OLI)imaging,a genetic algorithm was used to optimize the k-nearest neighbor model in the early stages,and the aboveground biomass of P.densata in the study area was estimated at the pixel scale after the k-NN three parameters(k,t,and d)were repeatedly tested and optimized.Results showed that accuracy of the k-NN model optimized by a genetic algorithm was better than the traditional k-NN model.Before optimization,the root mean square error was 30.0 t·hm~(-2),deviation was-0.418 t·hm~(-2),and R_(MSE )was 54.8%;after optimization,the root mean square error was 24.0 t·hm~(-2),deviation was-0.123 t·hm~(-2),and R_(MSE)was 43.7%.Finally,the estimated total aboveground biomass of P.densata in the study area was 0.89×10~7t based on the optimized k-NN model.[Ch,7 fig.6 tab.20 ref.]
For the traditional k-nearest neighbor(k-NN),there are insufficient problems that give the weight of the feature variables equally when searching the nearest neighbor population units and a lack of weight vectors for the feature variables.In this study,Shangri-la City,Yunnan Province,was selected as the research area,and Pinus densata was taken as the research object.Based on 49 field data plots,116 P.densata data samples,and Landsat 8/Operational Land Imager(OLI)imaging,a genetic algorithm was used to optimize the k-nearest neighbor model in the early stages,and the aboveground biomass of P.densata in the study area was estimated at the pixel scale after the k-NN three parameters(k,t,and d)were repeatedly tested and optimized.Results showed that accuracy of the k-NN model optimized by a genetic algorithm was better than the traditional k-NN model.Before optimization,the root mean square error was 30.0 t·hm~(-2),deviation was-0.418 t·hm~(-2),and R_(MSE )was 54.8%;after optimization,the root mean square error was 24.0 t·hm~(-2),deviation was-0.123 t·hm~(-2),and R_(MSE)was 43.7%.Finally,the estimated total aboveground biomass of P.densata in the study area was 0.89×10~7t based on the optimized k-NN model.[Ch,7 fig.6 tab.20 ref.]
引文
[1]王效科,冯宗炜.中国森林生态系统中植物固定大气碳的潜力[J].生态学杂志, 2000, 19(4):72-74.WANG Xiaoke, FENG Zongwei. The potential to sequester atmospheric carbon through forest ecosystems in China[J].Chin J Ecol, 2000, 19(4):72-74.
[2]胡会峰,刘国华.中国天然林保护工程的固碳能力估算[J].生态学报, 2006, 26(1):291-296.HU Huifeng, LIU Guohua. Carbon sequestration of China’s National Natural Forest Protection Project[J]. Acta Ecol Sin, 26(1):291-296.
[3]胡会峰,刘国华.森林管理在全球CO2减排中的作用[J].应用生态学报, 2006, 17(4):709-714.HU Huifeng, LIU Guohua. Roles of forest management in global carbon dioxide mitigation[J]. Chin J Appl Ecol,2006, 17(4):709-714.
[4]汤旭光,刘殿伟,王宗明,等.森林地上生物量遥感估算研究进展[J].生态学杂志, 2012, 31(5):1311-1318.TANG Xuguang, LIU Dianwei, WANG Zongming, et al. Estimation of forest aboveground biomass based on re mote sensing data:a review[J]. Chin J Ecol, 2012, 31(5):1311-1318.
[5]TOMPPO E. Satellite imagery-based national inventory of Finland[J]. Int Arch Photogramm Remote Sensing, 1991, 28(7/1):419-424.
[6]MCROBERTS R E. Estimating forest attribute parameters for small areas using nearest neighbors techniques[J]. For Ecol Manage, 2012, 272(3):3-12.
[7]MCROBERTS R E, N覷SSET E, GOBAKKEN T. Optimizing the k-Nearest Neighbors technique for estimating forest aboveground biomass using airborne laser scanning data[J]. Remote Sensing Environ, 2015, 163:13-22.
[8]MURA M, MCROBERTS R E, CHIRICI G, et al. Statistical inference for forest structural diversity indices using airborne laser scanning data and the k-Nearest Neighbors technique[J]. Remote Sensing Environ, 2016, 186:678-686.
[9]MCROBERTS R E, DOMKE G M, CHEN Q, et al. Using genetic algorithms to optimize k-Nearest Neighbors configurations for use with airborne laser scanning data[J]. Remote Sensing Environ, 2016, 184:387-395.
[10]MCROBERTS R E, CHEN Q, WALTERS B F. Multivariate inference for forest inventories using auxiliary airborne laser scanning data[J]. For Ecol Manage, 2017, 401:295-303.
[11]KATILA M, TOMPPO E. Stratification by ancillary data in multisource forest inventories employing k-nearest neighbor estimation[J]. Can J For Res, 2002, 32(9):1548-1561.
[12]TOMPPO E, HALME M. Using coarse scale forest variables as ancillary information and weighting of variables in kNN estimation:a genetic algorithm approach[J]. Remote Sensing Environ, 2004, 92(1):1-20.
[13]TOMPPO E, GAGLIANO C, NATALE F D, et al. Predicting categorical forest variables using an improved k-Nearest Neighbour estimator and Landsat imagery[J]. Remote Sensing Environ, 2009, 113(3):500-517.
[14]陈尔学,李增元,武红敢,等.基于k-NN和Landsat数据的小面积统计单元森林蓄积量估测方法[J].林业科学研究, 2008, 21(6):745-750.CHEN Erxue, LI Zengyuan, WU Honggan, et al. Forest volume estimation method for small areas based on k-NN and Landsat data[J]. For Res, 2008, 21(6):745-750.
[15]郭颖.森林地上生物量的非参数遥感估测方法优化[D].北京:中国林业科学研究院, 2011.GUO Ying. Optimum Non-Parametric Method for Forest Aboveground Biomass Estimation based on Remote Sensing Data[D]. Beijing:Chinese Academy of Forestry, 2011.
[16]胥辉,张会儒.林木生物量模型研究[M].昆明:云南科技出版社, 2002.
[17]CHIRICI G, MURA M, MCINEMEY D, et al. A meta-analysis and review of the literature on the k-Nearest Neighbors technique for forestry applications that use remotely sensed data[J]. Remote Sensing Environ, 2016, 176(2):282-294.
[18]谢福明,舒清态,字李,等.基于k-NN非参数模型的高山松生物量遥感估测研究[J].江西农业大学学报,2018, 40(4):743-750.XIE Fuming, SHU Qingtai, ZI Li, et al. Remote sensing estimation of Pinus densata aboveground biomass based on k-NN nonparametric model[J]. Acta Agric Univ Jiangxi, 2018, 40(4):743-750.
[19]BEAUDOIN A, BERNIER P Y, GUINDON L, et al. Mapping attributes of Canada’s forests at moderate resolution through k-NN and MODIS imagery[J]. Can J For Res, 2014, 44(5):521-532.
[20]MCROBERTS R E. Estimating forest attribute parameters for small areas using nearest neighbors techniques[J]. For Ecol Manage, 2012, 272(3):3-12.
[2]胡会峰,刘国华.中国天然林保护工程的固碳能力估算[J].生态学报, 2006, 26(1):291-296.HU Huifeng, LIU Guohua. Carbon sequestration of China’s National Natural Forest Protection Project[J]. Acta Ecol Sin, 26(1):291-296.
[3]胡会峰,刘国华.森林管理在全球CO2减排中的作用[J].应用生态学报, 2006, 17(4):709-714.HU Huifeng, LIU Guohua. Roles of forest management in global carbon dioxide mitigation[J]. Chin J Appl Ecol,2006, 17(4):709-714.
[4]汤旭光,刘殿伟,王宗明,等.森林地上生物量遥感估算研究进展[J].生态学杂志, 2012, 31(5):1311-1318.TANG Xuguang, LIU Dianwei, WANG Zongming, et al. Estimation of forest aboveground biomass based on re mote sensing data:a review[J]. Chin J Ecol, 2012, 31(5):1311-1318.
[5]TOMPPO E. Satellite imagery-based national inventory of Finland[J]. Int Arch Photogramm Remote Sensing, 1991, 28(7/1):419-424.
[6]MCROBERTS R E. Estimating forest attribute parameters for small areas using nearest neighbors techniques[J]. For Ecol Manage, 2012, 272(3):3-12.
[7]MCROBERTS R E, N覷SSET E, GOBAKKEN T. Optimizing the k-Nearest Neighbors technique for estimating forest aboveground biomass using airborne laser scanning data[J]. Remote Sensing Environ, 2015, 163:13-22.
[8]MURA M, MCROBERTS R E, CHIRICI G, et al. Statistical inference for forest structural diversity indices using airborne laser scanning data and the k-Nearest Neighbors technique[J]. Remote Sensing Environ, 2016, 186:678-686.
[9]MCROBERTS R E, DOMKE G M, CHEN Q, et al. Using genetic algorithms to optimize k-Nearest Neighbors configurations for use with airborne laser scanning data[J]. Remote Sensing Environ, 2016, 184:387-395.
[10]MCROBERTS R E, CHEN Q, WALTERS B F. Multivariate inference for forest inventories using auxiliary airborne laser scanning data[J]. For Ecol Manage, 2017, 401:295-303.
[11]KATILA M, TOMPPO E. Stratification by ancillary data in multisource forest inventories employing k-nearest neighbor estimation[J]. Can J For Res, 2002, 32(9):1548-1561.
[12]TOMPPO E, HALME M. Using coarse scale forest variables as ancillary information and weighting of variables in kNN estimation:a genetic algorithm approach[J]. Remote Sensing Environ, 2004, 92(1):1-20.
[13]TOMPPO E, GAGLIANO C, NATALE F D, et al. Predicting categorical forest variables using an improved k-Nearest Neighbour estimator and Landsat imagery[J]. Remote Sensing Environ, 2009, 113(3):500-517.
[14]陈尔学,李增元,武红敢,等.基于k-NN和Landsat数据的小面积统计单元森林蓄积量估测方法[J].林业科学研究, 2008, 21(6):745-750.CHEN Erxue, LI Zengyuan, WU Honggan, et al. Forest volume estimation method for small areas based on k-NN and Landsat data[J]. For Res, 2008, 21(6):745-750.
[15]郭颖.森林地上生物量的非参数遥感估测方法优化[D].北京:中国林业科学研究院, 2011.GUO Ying. Optimum Non-Parametric Method for Forest Aboveground Biomass Estimation based on Remote Sensing Data[D]. Beijing:Chinese Academy of Forestry, 2011.
[16]胥辉,张会儒.林木生物量模型研究[M].昆明:云南科技出版社, 2002.
[17]CHIRICI G, MURA M, MCINEMEY D, et al. A meta-analysis and review of the literature on the k-Nearest Neighbors technique for forestry applications that use remotely sensed data[J]. Remote Sensing Environ, 2016, 176(2):282-294.
[18]谢福明,舒清态,字李,等.基于k-NN非参数模型的高山松生物量遥感估测研究[J].江西农业大学学报,2018, 40(4):743-750.XIE Fuming, SHU Qingtai, ZI Li, et al. Remote sensing estimation of Pinus densata aboveground biomass based on k-NN nonparametric model[J]. Acta Agric Univ Jiangxi, 2018, 40(4):743-750.
[19]BEAUDOIN A, BERNIER P Y, GUINDON L, et al. Mapping attributes of Canada’s forests at moderate resolution through k-NN and MODIS imagery[J]. Can J For Res, 2014, 44(5):521-532.
[20]MCROBERTS R E. Estimating forest attribute parameters for small areas using nearest neighbors techniques[J]. For Ecol Manage, 2012, 272(3):3-12.