粒子群优化神经网络的土壤有机质高光谱估测
|
摘要
针对提高土壤有机质高光谱估测精度的问题,该文对山东省泰安市的92个棕壤样本进行光谱去噪,剔除异常样本处理后,对光谱反射率进行11种变换,发现一阶微分变换最佳;然后计算土壤有机质含量与变换后光谱反射率的相关系数,选取5个特征波段,分别利用多元线性回归、BP神经网络、支持向量机、粒子群优化神经网络4种方法建立土壤有机质含量高光谱估测模型并进行精度比较。实验结果表明,多元线性回归、BP神经网络、支持向量机和粒子群优化神经网络模型的决定系数R2分别为0.520 3、0.665 4、0.735 0和0.853 0,均方根误差分别为2.12、1.99、1.45和1.08。研究结果表明,粒子群优化神经网络的反演精度高、稳定性强,可有效提高土壤有机质的光谱估测能力。
In order to improve the accuracy of hyperspectral estimation of soil organic matter,92 brown soil samples of Tai'an city in Shandong province were selected to be processed by spectral denoising,after removing the abnormal samples and spectrum transforming by 11 kind methods,the first order differential transformation was found to be the best;then,the correlation coefficient between soil organic matter content and spectral reflectance was calculated,and five characteristic bands were selected,the hyperspectral estimation models of soil organic matter content were established by using multiple linear regression,BP neural network,support vector machine regression and particle swarm optimization neural network,then their accuracies were compared.The experimental results showed that the determination coefficients of the multiple linear regression,BP neural network,support vector machine regression and particle swarm optimization neural network models are 0.520 3,0.665 4,0.735 0 and 0.853 0,respectively,the root mean square error are 2.12,1.99,1.45 and 1.08,respectively.The results showed that particle swarm optimization neural network had high estimation accuracy and strong stability,and could effectively improve the hyperspectral estimation ability of soil organic matter.
In order to improve the accuracy of hyperspectral estimation of soil organic matter,92 brown soil samples of Tai'an city in Shandong province were selected to be processed by spectral denoising,after removing the abnormal samples and spectrum transforming by 11 kind methods,the first order differential transformation was found to be the best;then,the correlation coefficient between soil organic matter content and spectral reflectance was calculated,and five characteristic bands were selected,the hyperspectral estimation models of soil organic matter content were established by using multiple linear regression,BP neural network,support vector machine regression and particle swarm optimization neural network,then their accuracies were compared.The experimental results showed that the determination coefficients of the multiple linear regression,BP neural network,support vector machine regression and particle swarm optimization neural network models are 0.520 3,0.665 4,0.735 0 and 0.853 0,respectively,the root mean square error are 2.12,1.99,1.45 and 1.08,respectively.The results showed that particle swarm optimization neural network had high estimation accuracy and strong stability,and could effectively improve the hyperspectral estimation ability of soil organic matter.
引文
[1]叶勤,姜雪芹,李西灿,等.基于高光谱数据的土壤有机质含量反演模型比较[J].农业机械学报,2017,48(3):164-172.(YE Qin,JIANG Xueqin,LI Xican,et al.Comparison on inversion model of soil organic matter content based on hyperspectral data[J].Transactions of the Chinese Society for Agricultural Machinery,2017,48(3):164-172.)
[2]李西灿,赵庚星,陈红艳,等.土壤有机质含量区间值高光谱估测[J].测绘科学技术学报,2014,31(6):593-597.(LI Xican,ZHAO Gengxing,CHEN Hongyan,et al.The interval estimation of soil organic matter content based on hyper-spectral data[J].Journal of Geomatics Science and Technology,2014,31(6):593-597.)
[3]陈红艳,赵庚星,张晓辉,等.去除水分影响提高土壤有机质含量高光谱估测精度[J].农业工程学报,2014,30(8):91-100.(CHEN Hongyan,ZHAO Gengxing,ZHANG Xiaohui,et al.Improving estimation precision of soil organic matter content by removing effect of soil moisture from hyperspectra[J].Transactions of the Chinese Society of Agricultural Engineering,2014,30(8):91-100.)
[4]LIU F,ROSSITER D G,SONG X D,et al.A similaritybased method for three-dimensional prediction of soil organic matter concentration[J].Geoderma,2015,263(1):254-263.
[5]刘效栋.黄土台塬区土壤有机质高光谱特征及反演研究[D].西安:西北农林科技大学,2015.(LIU Xiaodong.The study on hyper-spectral characteristics and inversion of soil on loess plateau[D].Xi’an:Northwest Agriculture and Forestry University,2015.)
[6]栾福明,张小雷,熊黑钢,等.基于不同模型的土壤有机质含量高光谱反演比较分析[J].光谱学与光谱分析,2013,33(1):196-200.(LUAN Fuming,ZHANG Xiaolei,XIONG Heigang,et al.Comparative analysis of soil organic matter content based on different hyperspectral inversion models[J].Spectroscopy and Spectral Analysis,2013,33(1):196-200.)
[7]沈润平,丁国香,魏国栓,等.基于人工神经网络的土壤有机质含量高光谱反演[J].土壤学报,2009,46(3):391-397.(SHEN Runping,DING Guoxiang,WEI Guoshuan,et al.Retrieval of soil organic matter content from hyper-spectrum based on ann[J].Acta Pedologica sinica,2009,46(3):391-397.)
[8]王茵茵,齐雁冰,陈洋,等.基于多分辨率遥感数据与随机森林算法的土壤有机质预测研究[J].土壤学报,2016,53(2):342-354.(WANG Yinyin,QI Yanbing,CHEN Yang,et al.Prediction of soil organic matter based on multi-resolution remote sensing data and random forest algorithm[J].Acta Pedologica Sinica,2016,53(2):342-354.)
[9]袁征,李西灿,于涛,等.高光谱土壤有机质估测模型对比研究[J].测绘科学,2014,39(5):117-120.(YUAN Zheng,LI Xican,YU Tao,et al.Construction and innovation on comprehensive practice course of landuse remote sensing change detection[J].Science of Surveying and Mapping,2014,39(5):117-120.)
[10]卢延年,刘艳芳,陈奕云,等.江汉平原土壤有机碳含量高光谱预测模型优选[J].中国农学通报,2014,30(26):127-133.(LU Yannian,LIU Yanfang,CHEN Yiyun,et al.Optimization of the hyperspectral prediction model of soil organic carbon contents of Jianghan Plain[J].Chinese Agricultural Science Bulletin,2014,30(26):127-133.)
[11]CONFORTI M,BUTTAFUOCO G,LEONE A P,et al.Studying the relationship between water-induced soil erosion and soil organic matter using Vis-NIR spectroscopy and geomorphological analysis:a case study in Southern Italy[J].Catena,2013,110(2):44-58.
[12]侯艳军,塔西甫拉提·特依拜,买买提·沙吾提,等.荒漠土壤有机质含量高光谱估算模型[J].农业工程学报,2014,30(16):113-120.(HOU Yanjun,TAXIPULATI·Teyibai,MAIMAITI·Shawuti,et al.Estimation model of desert soil organic matter content using hyper-spectral data[J].Transactions of the Chinese Society of Agricultural Engineering,2014,30(16):113-120.)
[13]李胜男,曹萌萌,李盛楠,等.黑土典型区有机质高光谱预测模型[J].国土与自然资源研究,2016(4):73-76.(LI Shengnan,CAO Mengmeng,LI Shengnan,et al.Hyper-spectral prediction model of organic matter in black soil region[J].Territory and Natural Resources Study,2016(4):73-76.)
[14]于雷,洪永胜,耿雷,等.基于偏最小二乘回归的土壤有机质含量高光谱估算[J].农业工程学报,2015,31(14):103-109.(YU Lei,HONG Yongsheng,GENG Lei,et al.Hyperspectral estimation of soil organic matter content based on partial least squares regression[J].Transactions of the Chinese Society of Agricultural Engineering,2015,31(14):103-109.)
[15]韩兆迎,朱西存,刘庆,等.黄河三角洲土壤有机质含量的高光谱反演[J].植物营养与肥料学报,2014,20(6):1545-1552.(HAN Zhaoying,ZHU Xicun,LU Qing,et al.Hyperspectral inversion models for soil organic matter content in the Yellow River Delta[J].Journal of Plant Nutrition and Fertilizer,2014,20(6):1545-1552.)
[16]谭琨,张倩倩,曹茜,等.基于粒子群优化支持向量机的矿区土壤有机质含量高光谱反演[J].地球科学:中国地质大学学报,2015,40(8):1339-1344.(TAN Kun,ZHANG Qianqian,CAO Qian,et al.Hyperspectral retrieval model of soil organic matter content based on particle swarm optimization-support vector machines[J].Earth Science-Journal of China University of Geosciences,2015,40(8):1339-1344.)
[17]LI M L,LI X C,YE T,et al.Grey relation estimating pattern of soil organic matter with residual modification based on hyper-spectral data[J].The Journal of Grey System,2016,28(4):27-39.
[18]高海兵,高亮,周驰,等.基于粒子群优化的神经网络训练算法研究[J].电子学报,2004,32(9):1572-1574.(GAO Haibing,GAO Liang,ZHOU Chi,et al.Particle swarm optimization based algorithm for neural network learning[J].Acta Electronica sinica,2004,32(9):1572-1574.)
[19]尚璇,李西灿,徐邮邮,等.土壤水与有机质对高光谱的作用及交互作用规律[J].中国农业科学,2017,50(8):1465-1475.(SHANG Xuan,LI Xican,XU Youyou,et al.The role and interaction of soil water and organic matter on hyper-spectral reflectance[J].Scientia Agricultura Sinica,2017,50(8):1465-1475.)
[20]李明亮,李西灿,张爽.土壤含水量高光谱灰色关联度估测模式[J].测绘科学技术学报,2016,33(2):163-168.(LI Mingliang,LI Xican,ZHANG Shuang.Grey relation estimating pattern of soil water content based on hyper-spectral data[J].Journal of Geomatics Science and Technology,2016,33(2):163-168.)
[21]王彤彤,翟军海,何欢,等.BP神经网络和SVM模型对施加生物炭土壤水分预测的适用性[J].水土保持研究,2017,24(3):86-91.(WANG Tongtong,ZHAI Junhai,HE Huan,et al.Applicability of BP neural network model and SVM model to predicting soil moisture under incorporation of biochar into soils[J].Research of Soil and Water Conservation,2017,24(3):86-91.)
[22]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(1):2-10.(DING Shifei,QI Bingjuan,TAN Hongyan.An overview on theory and algorithm of support vector machines[J].Journal of University of Electronic Science and Technology of China,2011,40(1):2-10.)
[23]卓金武.MATLAB在数学建模中的应用[M].北京:北京航空航天大学出版社,2015:151-153.(ZHUO Jinwu.Application of MATLAB in mathematical modeling[M].Beijing:Beijing University of Aeronautics and Astronautics Press,2015:151-153.)
[2]李西灿,赵庚星,陈红艳,等.土壤有机质含量区间值高光谱估测[J].测绘科学技术学报,2014,31(6):593-597.(LI Xican,ZHAO Gengxing,CHEN Hongyan,et al.The interval estimation of soil organic matter content based on hyper-spectral data[J].Journal of Geomatics Science and Technology,2014,31(6):593-597.)
[3]陈红艳,赵庚星,张晓辉,等.去除水分影响提高土壤有机质含量高光谱估测精度[J].农业工程学报,2014,30(8):91-100.(CHEN Hongyan,ZHAO Gengxing,ZHANG Xiaohui,et al.Improving estimation precision of soil organic matter content by removing effect of soil moisture from hyperspectra[J].Transactions of the Chinese Society of Agricultural Engineering,2014,30(8):91-100.)
[4]LIU F,ROSSITER D G,SONG X D,et al.A similaritybased method for three-dimensional prediction of soil organic matter concentration[J].Geoderma,2015,263(1):254-263.
[5]刘效栋.黄土台塬区土壤有机质高光谱特征及反演研究[D].西安:西北农林科技大学,2015.(LIU Xiaodong.The study on hyper-spectral characteristics and inversion of soil on loess plateau[D].Xi’an:Northwest Agriculture and Forestry University,2015.)
[6]栾福明,张小雷,熊黑钢,等.基于不同模型的土壤有机质含量高光谱反演比较分析[J].光谱学与光谱分析,2013,33(1):196-200.(LUAN Fuming,ZHANG Xiaolei,XIONG Heigang,et al.Comparative analysis of soil organic matter content based on different hyperspectral inversion models[J].Spectroscopy and Spectral Analysis,2013,33(1):196-200.)
[7]沈润平,丁国香,魏国栓,等.基于人工神经网络的土壤有机质含量高光谱反演[J].土壤学报,2009,46(3):391-397.(SHEN Runping,DING Guoxiang,WEI Guoshuan,et al.Retrieval of soil organic matter content from hyper-spectrum based on ann[J].Acta Pedologica sinica,2009,46(3):391-397.)
[8]王茵茵,齐雁冰,陈洋,等.基于多分辨率遥感数据与随机森林算法的土壤有机质预测研究[J].土壤学报,2016,53(2):342-354.(WANG Yinyin,QI Yanbing,CHEN Yang,et al.Prediction of soil organic matter based on multi-resolution remote sensing data and random forest algorithm[J].Acta Pedologica Sinica,2016,53(2):342-354.)
[9]袁征,李西灿,于涛,等.高光谱土壤有机质估测模型对比研究[J].测绘科学,2014,39(5):117-120.(YUAN Zheng,LI Xican,YU Tao,et al.Construction and innovation on comprehensive practice course of landuse remote sensing change detection[J].Science of Surveying and Mapping,2014,39(5):117-120.)
[10]卢延年,刘艳芳,陈奕云,等.江汉平原土壤有机碳含量高光谱预测模型优选[J].中国农学通报,2014,30(26):127-133.(LU Yannian,LIU Yanfang,CHEN Yiyun,et al.Optimization of the hyperspectral prediction model of soil organic carbon contents of Jianghan Plain[J].Chinese Agricultural Science Bulletin,2014,30(26):127-133.)
[11]CONFORTI M,BUTTAFUOCO G,LEONE A P,et al.Studying the relationship between water-induced soil erosion and soil organic matter using Vis-NIR spectroscopy and geomorphological analysis:a case study in Southern Italy[J].Catena,2013,110(2):44-58.
[12]侯艳军,塔西甫拉提·特依拜,买买提·沙吾提,等.荒漠土壤有机质含量高光谱估算模型[J].农业工程学报,2014,30(16):113-120.(HOU Yanjun,TAXIPULATI·Teyibai,MAIMAITI·Shawuti,et al.Estimation model of desert soil organic matter content using hyper-spectral data[J].Transactions of the Chinese Society of Agricultural Engineering,2014,30(16):113-120.)
[13]李胜男,曹萌萌,李盛楠,等.黑土典型区有机质高光谱预测模型[J].国土与自然资源研究,2016(4):73-76.(LI Shengnan,CAO Mengmeng,LI Shengnan,et al.Hyper-spectral prediction model of organic matter in black soil region[J].Territory and Natural Resources Study,2016(4):73-76.)
[14]于雷,洪永胜,耿雷,等.基于偏最小二乘回归的土壤有机质含量高光谱估算[J].农业工程学报,2015,31(14):103-109.(YU Lei,HONG Yongsheng,GENG Lei,et al.Hyperspectral estimation of soil organic matter content based on partial least squares regression[J].Transactions of the Chinese Society of Agricultural Engineering,2015,31(14):103-109.)
[15]韩兆迎,朱西存,刘庆,等.黄河三角洲土壤有机质含量的高光谱反演[J].植物营养与肥料学报,2014,20(6):1545-1552.(HAN Zhaoying,ZHU Xicun,LU Qing,et al.Hyperspectral inversion models for soil organic matter content in the Yellow River Delta[J].Journal of Plant Nutrition and Fertilizer,2014,20(6):1545-1552.)
[16]谭琨,张倩倩,曹茜,等.基于粒子群优化支持向量机的矿区土壤有机质含量高光谱反演[J].地球科学:中国地质大学学报,2015,40(8):1339-1344.(TAN Kun,ZHANG Qianqian,CAO Qian,et al.Hyperspectral retrieval model of soil organic matter content based on particle swarm optimization-support vector machines[J].Earth Science-Journal of China University of Geosciences,2015,40(8):1339-1344.)
[17]LI M L,LI X C,YE T,et al.Grey relation estimating pattern of soil organic matter with residual modification based on hyper-spectral data[J].The Journal of Grey System,2016,28(4):27-39.
[18]高海兵,高亮,周驰,等.基于粒子群优化的神经网络训练算法研究[J].电子学报,2004,32(9):1572-1574.(GAO Haibing,GAO Liang,ZHOU Chi,et al.Particle swarm optimization based algorithm for neural network learning[J].Acta Electronica sinica,2004,32(9):1572-1574.)
[19]尚璇,李西灿,徐邮邮,等.土壤水与有机质对高光谱的作用及交互作用规律[J].中国农业科学,2017,50(8):1465-1475.(SHANG Xuan,LI Xican,XU Youyou,et al.The role and interaction of soil water and organic matter on hyper-spectral reflectance[J].Scientia Agricultura Sinica,2017,50(8):1465-1475.)
[20]李明亮,李西灿,张爽.土壤含水量高光谱灰色关联度估测模式[J].测绘科学技术学报,2016,33(2):163-168.(LI Mingliang,LI Xican,ZHANG Shuang.Grey relation estimating pattern of soil water content based on hyper-spectral data[J].Journal of Geomatics Science and Technology,2016,33(2):163-168.)
[21]王彤彤,翟军海,何欢,等.BP神经网络和SVM模型对施加生物炭土壤水分预测的适用性[J].水土保持研究,2017,24(3):86-91.(WANG Tongtong,ZHAI Junhai,HE Huan,et al.Applicability of BP neural network model and SVM model to predicting soil moisture under incorporation of biochar into soils[J].Research of Soil and Water Conservation,2017,24(3):86-91.)
[22]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(1):2-10.(DING Shifei,QI Bingjuan,TAN Hongyan.An overview on theory and algorithm of support vector machines[J].Journal of University of Electronic Science and Technology of China,2011,40(1):2-10.)
[23]卓金武.MATLAB在数学建模中的应用[M].北京:北京航空航天大学出版社,2015:151-153.(ZHUO Jinwu.Application of MATLAB in mathematical modeling[M].Beijing:Beijing University of Aeronautics and Astronautics Press,2015:151-153.)
© 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 |