基于SVR的GF1号遥感影像森林蓄积量估测
|
摘要
国产高分辨率卫星遥感影像已成为森林资源调查和监测的重要数据源。基于国产卫星遥感的森林蓄积量估算成为重要的研究方向之一。本研究以福建省将乐县为研究区,选择国产高分辨率高分一号卫星2 m分辨率遥感影像为主要数据源,加以辅助野外实地调查数据,分别采用多元线性回归和SVM(support vector machine)回归方法开展亚热带针叶林蓄积量估算效果评价研究。首先,从融合影像中提取遥感因子,包括11个光谱因子和10个纹理因子等;其次,对21个遥感因子进行相关性分析,选取皮尔森相关系数较大的6个遥感因子;第三,应用多元线性回归和支持向量机回归(support vector regression,SVR)对所选遥感因子建立模型,选取最优模型反演将乐县蓄积量分布图。结果表明:支持向量机回归(SVR)估测蓄积量的模型预估精度达到98. 22%。
High resolution satellite remote sensing image data has become the main source of forest resource survey and monitoring. The domestic satellite remote sensing forest volume estimation has become one of the important research direction. The paper takes Jiangle County,Fujian Province as the research area,selects the High resolution GF-1 satellite with 2 m resolution remote sensing image made in China as the main data source,and assists the field investigation data. Multiple linear regression and support vector machine( SVM) regression were used to evaluate the effect of subtropical coniferous forest volume estimation. Firstly,a series of remote sensing factors such as 11 spectral factors and 10 texture factors were extracted from the fusion image. Secondly,the correlation of 21 remote sensing factors was analyzed and 6 remote sensing factors with large Pearson correlation coefficient were selected. The multiple linear regression and support vector regression( SVR) were used to establish the model of the selected remote sensing factors,and the optimal model was selected to retrieve the distribution map of the volume in Jiangle County. The results show that the prediction accuracy of SVR model is 98.22%.
High resolution satellite remote sensing image data has become the main source of forest resource survey and monitoring. The domestic satellite remote sensing forest volume estimation has become one of the important research direction. The paper takes Jiangle County,Fujian Province as the research area,selects the High resolution GF-1 satellite with 2 m resolution remote sensing image made in China as the main data source,and assists the field investigation data. Multiple linear regression and support vector machine( SVM) regression were used to evaluate the effect of subtropical coniferous forest volume estimation. Firstly,a series of remote sensing factors such as 11 spectral factors and 10 texture factors were extracted from the fusion image. Secondly,the correlation of 21 remote sensing factors was analyzed and 6 remote sensing factors with large Pearson correlation coefficient were selected. The multiple linear regression and support vector regression( SVR) were used to establish the model of the selected remote sensing factors,and the optimal model was selected to retrieve the distribution map of the volume in Jiangle County. The results show that the prediction accuracy of SVR model is 98.22%.
引文
[1]TOMPPO E,NILSSON M,ROSENGREN M,et al.Simultaneous use of Landsat-TM and IRS-1CWiFS data in estimating large area tree stem volume and aboveground biomass[J].Remote Sensing of Environment,2002,82(1):156-171.
[2]李亦秋,冯仲科,邓欧,等.基于3S技术的山东省森林蓄积量估测[J].林业科学,2009,45(9):85-93.
[3]王臣立,牛铮,郭治兴,等.基于植被指数和神经网络的热带人工林地上蓄积量遥感估测[J].生态环境学报,2009,18(5):1830-1834.
[4]黄平,杨燕琼,侯长谋.基于RS、GIS的杉木林分蓄积量判读模型研究[J].中南林业调查规划,2003,22(1):25-27.
[5]WIDLOWSKI J L,PINTY B,GOBRON N,et al.Detection and characterization of boreal coniferous forests fromremotesensing data[J].Journal of Geo Physical Research-Atmospheres,2001,106:33405-33419.
[6]DONG J R,KAUFMANN R K,MYNENI R B,et al.Remote sensing estimates of boreal and temperate forest woody,biomass:carbon,pools,sources,and sinks[J].Remote Sensing of Environment,2003,84:393-410.
[7]邰晓雷,潘晓露,李一民.高分辨率遥感影像城区地物信息提取技术研究[J].贵州大学学报(自然科学版),2012,29(1):88-91.
[8]刘茜,杨乐,柳钦火,等.森林地上生物量遥感反演方法综述[J].遥感学报,2015,19(1):62-74.
[9]BAAH-ACHEAMFOUR M,CHANG S X,CARLYLE C N,et al.Carbon pool size and stability are affected by trees and grassland cover types within agroforestry systems of western Canada[J].Agriculture,Ecosystems and Environment,2015,213:105-113.
[10]申鑫,曹林,佘光辉.高光谱与高空间分辨率遥感数据的亚热带森林生物量反演[J].遥感学报,2016,20(6):1446-1460.
[11]张雅,尹小君,王伟强.基于Landsat 8 OLI遥感影像的天山北坡草地地上生物量估算[J].遥感技术与应用,2017,32(6):1012-1021.
[12]ZHENG S,CAO C,DANG Y,et al.Retrieval of forest growing stock volume by two different methods using Landsat TM images[J].International Journal of Remote Sensing,2014,35(1):29-43.
[13]刘明艳,王秀兰,冯仲科,等.基于主成分分析法的老秃顶子自然保护区森林蓄积量遥感估测[J].中南林业科技大学学报,2017,37(10):80-83.
[14]陈红艳,赵庚星,陈敬春,等.基于改进植被指数的黄河口区盐渍土盐分遥感反演[J].农业工程学报,2015,31(5):107-112+114+113.
[15]杨柳,冯仲科,岳德鹏,等.结合纹理因子和地形因子的森林蓄积量多光谱估测模型[J].光谱学与光谱分析,2017,37(7):2140-2145.
[16]向安民,刘凤伶,于宝义,等.基于k-NN方法和GF遥感影像的森林蓄积量估测[J].浙江农林大学学报,2017,34(3):406-412.
[17]PAL M,MATHER P M.MATHER.Support vector machines for classification in remote sensing[J].International Journal of Remote Sensing,2015,26(5):1007-1011.
[18]BROWN M,LEWIS H G,GUNN S R.Linear spectral mixture models and support vector machines for remote sensing[J].IEEETransactions on Geoscience&Remote Sensing,2000,38(5):2346-2360.
[19]平跃鹏,臧淑英.基于MODIS时间序列及物候特征的农作物分类[J].自然资源学报,2016,31(3):503-513.
[20]付虎艳,徐云栋,李圣娇,等.基于SVM方法的高山松林蓄积量遥感估测研究[J].西部林业科学,2014(4):83-88.
[21]蒙诗栎,庞勇,张钟军,等.WorldView-2纹理的森林地上生物量反演[J].遥感学报,2017(5):812-824.
[22]王佳,宋珊芸,刘霞,等.结合影像光谱与地形因子的森林蓄积量估测模型[J].农业机械学报,2014,45(5):216-220.
[23]傅银贞,汪小钦.基于北京一号CCD数据的植被指数特性分析[J].测绘科学,2010,35(6):35-38.
[24]DUTTA S,DATTA A,CHAKLADAR N D,et al.Detection of tool condition from the turned surface images using an accurate grey level co-occurrence technique[J].Precision Engineering,2012,36(3):458-466.
[25]SULOCHANA S,VIDHYA R.Texture Based Image Retrieval U-sing Framelet Transform-Gray Level Co-occurrence Matrix(GL-CM)[J].International Journal of Advanced Research in Artificial Intelligence,2013,2(2).
[26]任国贞,江涛.基于灰度共生矩阵的纹理提取方法研究[J].计算机应用与软件,2014(11):190-192.
[27]AYDIN I,KARAKOSE M,AKIN E.A multi-objective artificial immune algorithm for parameter optimization in support vector machine.[J].Applied Soft Computing,2011,11(1):120-129.
[28]曾伟生,唐守正.立木生物量方程的优度评价和精度分析[J].林业科学,2011,47(11):106-113.
[29]张新乐,徐梦园,刘焕军,等.引入地形因子的黑土区大豆干生物量遥感反演模型及验证[J].农业工程学报,2017,33(16):168-173.
[30]AREL I,ROSE D C,KARNOWSKI T P.Deep machine learninga new frontier in artificial intelligence research[Research Frontier][J].Computational Intelligence Magazine IEEE,2010,5(4):13-18.
[31]涂云燕,彭道黎.基于神经网络的森林蓄积量估测[J].中南林业科技大学学报,2012,32(3):49-52.
[2]李亦秋,冯仲科,邓欧,等.基于3S技术的山东省森林蓄积量估测[J].林业科学,2009,45(9):85-93.
[3]王臣立,牛铮,郭治兴,等.基于植被指数和神经网络的热带人工林地上蓄积量遥感估测[J].生态环境学报,2009,18(5):1830-1834.
[4]黄平,杨燕琼,侯长谋.基于RS、GIS的杉木林分蓄积量判读模型研究[J].中南林业调查规划,2003,22(1):25-27.
[5]WIDLOWSKI J L,PINTY B,GOBRON N,et al.Detection and characterization of boreal coniferous forests fromremotesensing data[J].Journal of Geo Physical Research-Atmospheres,2001,106:33405-33419.
[6]DONG J R,KAUFMANN R K,MYNENI R B,et al.Remote sensing estimates of boreal and temperate forest woody,biomass:carbon,pools,sources,and sinks[J].Remote Sensing of Environment,2003,84:393-410.
[7]邰晓雷,潘晓露,李一民.高分辨率遥感影像城区地物信息提取技术研究[J].贵州大学学报(自然科学版),2012,29(1):88-91.
[8]刘茜,杨乐,柳钦火,等.森林地上生物量遥感反演方法综述[J].遥感学报,2015,19(1):62-74.
[9]BAAH-ACHEAMFOUR M,CHANG S X,CARLYLE C N,et al.Carbon pool size and stability are affected by trees and grassland cover types within agroforestry systems of western Canada[J].Agriculture,Ecosystems and Environment,2015,213:105-113.
[10]申鑫,曹林,佘光辉.高光谱与高空间分辨率遥感数据的亚热带森林生物量反演[J].遥感学报,2016,20(6):1446-1460.
[11]张雅,尹小君,王伟强.基于Landsat 8 OLI遥感影像的天山北坡草地地上生物量估算[J].遥感技术与应用,2017,32(6):1012-1021.
[12]ZHENG S,CAO C,DANG Y,et al.Retrieval of forest growing stock volume by two different methods using Landsat TM images[J].International Journal of Remote Sensing,2014,35(1):29-43.
[13]刘明艳,王秀兰,冯仲科,等.基于主成分分析法的老秃顶子自然保护区森林蓄积量遥感估测[J].中南林业科技大学学报,2017,37(10):80-83.
[14]陈红艳,赵庚星,陈敬春,等.基于改进植被指数的黄河口区盐渍土盐分遥感反演[J].农业工程学报,2015,31(5):107-112+114+113.
[15]杨柳,冯仲科,岳德鹏,等.结合纹理因子和地形因子的森林蓄积量多光谱估测模型[J].光谱学与光谱分析,2017,37(7):2140-2145.
[16]向安民,刘凤伶,于宝义,等.基于k-NN方法和GF遥感影像的森林蓄积量估测[J].浙江农林大学学报,2017,34(3):406-412.
[17]PAL M,MATHER P M.MATHER.Support vector machines for classification in remote sensing[J].International Journal of Remote Sensing,2015,26(5):1007-1011.
[18]BROWN M,LEWIS H G,GUNN S R.Linear spectral mixture models and support vector machines for remote sensing[J].IEEETransactions on Geoscience&Remote Sensing,2000,38(5):2346-2360.
[19]平跃鹏,臧淑英.基于MODIS时间序列及物候特征的农作物分类[J].自然资源学报,2016,31(3):503-513.
[20]付虎艳,徐云栋,李圣娇,等.基于SVM方法的高山松林蓄积量遥感估测研究[J].西部林业科学,2014(4):83-88.
[21]蒙诗栎,庞勇,张钟军,等.WorldView-2纹理的森林地上生物量反演[J].遥感学报,2017(5):812-824.
[22]王佳,宋珊芸,刘霞,等.结合影像光谱与地形因子的森林蓄积量估测模型[J].农业机械学报,2014,45(5):216-220.
[23]傅银贞,汪小钦.基于北京一号CCD数据的植被指数特性分析[J].测绘科学,2010,35(6):35-38.
[24]DUTTA S,DATTA A,CHAKLADAR N D,et al.Detection of tool condition from the turned surface images using an accurate grey level co-occurrence technique[J].Precision Engineering,2012,36(3):458-466.
[25]SULOCHANA S,VIDHYA R.Texture Based Image Retrieval U-sing Framelet Transform-Gray Level Co-occurrence Matrix(GL-CM)[J].International Journal of Advanced Research in Artificial Intelligence,2013,2(2).
[26]任国贞,江涛.基于灰度共生矩阵的纹理提取方法研究[J].计算机应用与软件,2014(11):190-192.
[27]AYDIN I,KARAKOSE M,AKIN E.A multi-objective artificial immune algorithm for parameter optimization in support vector machine.[J].Applied Soft Computing,2011,11(1):120-129.
[28]曾伟生,唐守正.立木生物量方程的优度评价和精度分析[J].林业科学,2011,47(11):106-113.
[29]张新乐,徐梦园,刘焕军,等.引入地形因子的黑土区大豆干生物量遥感反演模型及验证[J].农业工程学报,2017,33(16):168-173.
[30]AREL I,ROSE D C,KARNOWSKI T P.Deep machine learninga new frontier in artificial intelligence research[Research Frontier][J].Computational Intelligence Magazine IEEE,2010,5(4):13-18.
[31]涂云燕,彭道黎.基于神经网络的森林蓄积量估测[J].中南林业科技大学学报,2012,32(3):49-52.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |