基于多光谱变换和主成分分析的土壤全铁含量随机森林模型反演

doi:10.3969/j.issn.1004-1524.20240733

浙江农业学报 ›› 2025, Vol. 37 ›› Issue (7): 1521-1532.DOI: 10.3969/j.issn.1004-1524.20240733

基于多光谱变换和主成分分析的土壤全铁含量随机森林模型反演

江振蓝¹(), 陈付勋¹, 罗双飞¹, 罗烨琴¹, 沙晋明²^,^*()

1.闽江学院地理与海洋学院,福建福州 350108
2.福建师范大学地理科学学院,福建福州 350108

收稿日期:2024-08-12 出版日期:2025-07-25 发布日期:2025-08-20
作者简介:江振蓝(1977—),女,福建政和人,博士,教授,主要从事生态环境遥感与信息技术方面的研究。E-mail: jessie33cn@163.com
通讯作者: ^*沙晋明,E-mail: jmsha@fjnu.edu.cn
基金资助:
福建省自然科学基金面上项目(2021J011020);福建省自然科学基金面上项目(2021J011022)

Inversion of soil total iron content using random forest model based on multi-spectral transformation and principle compoment analysis

JIANG Zhenlan¹(), CHEN Fuxun¹, LUO Shuangfei¹, LUO Yeqin¹, SHA Jinming²^,^*()

1. College of Geography and Oceanography, Minjiang University, Fuzhou 350108, China
2. School of Geographical Science, Fujian Normal University, Fuzhou 350108, China

Received:2024-08-12 Online:2025-07-25 Published:2025-08-20

摘要/Abstract

摘要：

目前,土壤全铁含量的高光谱反演研究多采用单一光谱变量作为输入,忽视了光谱变量间的互补性。同时,光谱波段间的冗余信息也影响了模型的预测精度和泛化能力。为解决以上问题,以福州市土壤全铁含量为研究对象,提出了一种基于组合光谱和主成分分析(PCA)优化的随机森林(RF)模型。通过整合原始反射率及其13种数学变换,构建组合光谱变量集,并结合PCA与多元线性回归(MLR)、竞争性自适应重加权采样(CARS)、遗传算法(GA)、连续投影算法(SPA)、无信息变量去除(UVE)等变量选择方法进行变量优化。基于优化后的变量集,建立RF模型,用于土壤全铁含量的预测。结果表明,所构建的模型在验证集上的决定系数(R²)和相对分析误差(RPD)分别大于0.8和2.8,显示出良好的预测能力。其中,CARS-PCA-RF、GA-PCA-RF和MLR-PCA-RF模型在验证集上的RPD均大于3,预测能力突出,特别是CARS-PCA-RF模型的表现尤为出色,在验证集上的RPD值为3.292,显示了PCA结合CARS的变量选择方法在土壤全铁含量高光谱预测中的优势和潜力。该研究提出了一种基于多种光谱变换和PCA优化输入变量的土壤全铁含量预测方法,显著提升了土壤全铁含量预测的精度和稳定性,为区域土壤全铁含量的高光谱预测提供了新的解决方案。

关键词: 土壤全铁含量, 光谱变换, 随机森林, 主成分分析, 高光谱预测

Abstract:

Typical hyperspectral inversion models for soil total iron content use single spectral variables as input, which neglect the complementarity among spectral variables. Additionally, the redundancy of spectral bands affects the prediction accuracy and generalization ability of models. In the present study, a random forest (RF) model optimized by integrating spectral variables and principal component analysis (PCA) was proposed, and the soil total iron content in Fuzhou City of China was selected as the study object. By incorporating the original reflectance and its 13 mathematical transformations, a combined spectral variable set was constructed. For variable optimization, PCA was employed in conjunction with various variable selection methods, including multiple linear regression (MLR), competitive adaptive reweighted sampling (CARS), genetic algorithm (GA), successive projections algorithm (SPA), and uninformative variable elimination (UVE). Based on the optimized variable set, RF inversion models were established to predict soil total iron content. The results indicated that all the constructed models exhibited excellent predictive capability in the validation set, with determination coefficient (R²) values higher than 0.8 and relative percent difference (RPD) values exceeding 2.8. Among these model, the CARS-PCA-RF, GA-PCA-RF and MLR-PCA-RF models demonstrated strong predictive abilities, with RPD values in the validation set exceeding 3. Notably, the CARS-PCA-RF model performed the best, with an RPD value of 3.292 in the validation set, highlighting the advantages and potential of the variable selection method which combines PCA and CARS in the hyperspectral prediction of soil total iron content. This study proposed a method for predicting soil total iron content based on multiple spectral transformations and PCA-optimized input variables. This approach improved the accuracy and stability of soil total iron content prediction, providing a new solution for the hyperspectral prediction of regional soil total iron content.

Key words: soil total iron content, spectral transformation, random forest, principal component analysis, hyperspectral prediction

中图分类号:

S127
S153.6

江振蓝, 陈付勋, 罗双飞, 罗烨琴, 沙晋明. 基于多光谱变换和主成分分析的土壤全铁含量随机森林模型反演[J]. 浙江农业学报, 2025, 37(7): 1521-1532.

JIANG Zhenlan, CHEN Fuxun, LUO Shuangfei, LUO Yeqin, SHA Jinming. Inversion of soil total iron content using random forest model based on multi-spectral transformation and principle compoment analysis[J]. Acta Agriculturae Zhejiangensis, 2025, 37(7): 1521-1532.

图/表 8

表1 土壤全铁含量的统计特征

Table 1 Statistical characteristics of soil total iron content

样本集 Data set	样本数 Sample size	最小值 Minimum/ (g·kg^-1)	最大值 Maximum/ (g·kg^-1)	平均值 Mean/ (g·kg^-1)	标准差 Standard deviation/ (g·kg^-1)	变异系数 Coefficient of variation/%
总集Whole set	132	5.54	106.35	24.14	14.62	60.56
建模集Modeling set	88	7.89	106.35	24.53	15.45	62.98
验证集Validation set	44	5.54	66.36	23.35	12.91	55.29

图1 土壤光谱与全铁含量的皮尔逊(Pearson)相关系数 SR,原始光谱(土壤反射率);FD,FD(一阶微分)变换光谱;SD,SD(二阶微分)变换光谱;RT,RT(倒数变换)光谱;RTFD,RTFD(倒数的一阶微分)变换光谱;RTSD,RTSD(倒数的二阶微分)变换光谱;AT,AT(倒数的对数)变换光谱;ATFD,ATFD(倒数对数的一阶微分)变换光谱;ATSD,ATSD(倒数对数的二阶微分)变换光谱;CR,CR(连续统去除)变换光谱;DT,DT(去趋势校正)变换光谱;MSC,MSC(多元散射校正)变换光谱;SNV,SNV(标准正态变换)变换光谱;WT,WT(小波变换)变换光谱。下同。r,相关系数。

Fig.1 Pearson correlation coefficient between soil spectra and total iron content SR, Raw spectra (soil reflectance); FD, Spectra after FD (first derivative) transformation; SD, Spectra after SD (second derivative) transformation; RT, Spectra after RT (reciprocal transformation) transformation; RTFD, Spectra after RTFD (first derivative of reciprocal) transformation; RTSD, Spectra after RTSD (second derivative of reciprocal) transformation; AT, Spectra after AT (absorbance transformation) transformation; ATFD, Spectra after ATFD (first derivative of absorbance) transformation; ATSD, Spectra after ATSD (second derivative of absorbance) transformation; CR, Spectra after CR (continuum removal) transformation; DT, Spectra after DT (detrending) transformation; MSC, Spectra after MSC (multiplicative scatter correction) transformation; SNV, Spectra after SNV (standard normal variate) transformation; WT, Spectra after WT (wavelet transformation) transformation. The same as below.r,Correlation coefficient.

表2 土壤光谱与全铁含量相关性最强的特征波段

Table 2 Characteristic bands with the strongest correlation between soil spectra and total iron content

光谱 Spectra	特征波长 Characteristic wavelengths/nm	相关系数 Correlation coefficient
SR	350~359	-0.523^**
RT	380~389	0.651^**
RTFD	530~539	-0.716^**
RTSD	450~459	0.703^**
AT	350~359	0.589^**
ATFD	540~549	-0.596^**
ATSD	740~749	0.592^**
CR	490~499	-0.606^**
FD	1 150~1 159	0.453^**
SD	860~869	0.486^**
SNV	1 440~1 449	0.477^**
MSC	1 440~1 449	0.480^**
DT	1 430~1 439	0.438^**
	2 150~2 159	-0.438^**
WT	350~359	-0.523^**

图2 变量间的皮尔逊(Pearson)相关系数 ATSD_740~749表示ATSD变换光谱740~749 nm波段数据,其他以此类推。“*”和“**”分别表示在P<0.05和P<0.01水平上显著相关。

Fig.2 Pearson correlation coefficients among variables ATSD_740~749 denotes the 740-749 nm band data of the spectra after ATSD (second derivative of absorbance) transformation. The rest may be deduced from this example. “*” and “**” indicate significant correlation at P<0.05 and P<0.01, respectively.

表3 主成分分析(PCA)优化前土壤全铁含量反演RF模型的参数

Table 3 Parameters of RF model for soil total iron content inversion before principle component analysis (PCA) optimization

变量选择方法 Variable selection method	变量数量 Number of variables	输入变量 Input variables	在建模集上的效果 Performance on the modeling set		在验证集上的效果 Performance on the validation set
变量选择方法 Variable selection method	变量数量 Number of variables	输入变量 Input variables	R²	RMSE/ (g·kg^-1)	R²	RMSE/ (g·kg^-1)	RPD
CSV	15	SR_350~359, FD_1 150~1 159, SD_860~869, AT_350~359, ATFD_540~549, ATSD_740~749, RT_380~389, RTFD_530~539, RTSD_450~459, CR_490~499, SNV_1 440~1 449, MSC_1 440~1 449, DT_1 430~1 439, DT_2 150~2 159, WT_350~359	0.918	4.463	0.444	9.744	1.325
MLR	6	SR_350~359, ATFD_540~549, ATSD_740~749, RTSD_450~459, MSC_1 440~1 449, DT_1 430~1 439	0.915	4.531	0.526	8.998	1.435
CARS	12	SR_350~359, SD_860~869, AT_350~359, ATFD_540~549, ATSD_740~749, RT_380~389, CR_490~499, SNV_1 440~1 449, MSC_1 440~1 449, DT_1 430~1 439, DT_2 150~2 159, WT_350~359	0.917	4.468	0.443	9.750	1.324
GA	7	SR_350~359, ATFD_540~549, ATSD_740~749, RTSD_450~459, MSC_1 440~1 449, DT_1 430~1 439, WT_350~359	0.919	4.419	0.539	8.871	1.455
SPA	5	FD_1150~1159, ATSD_740~749, RTSD_450~459, SNV_1 440~1 449, DT_2 150~2 159	0.908	4.719	0.369	10.377	1.244
UVE	10	SR_350~359, ATFD_540~549, ATSD_740~749, RT_370~389, RTSD_450~459, CR_490~499, SNV_1 440~1 449, MSC_1 440~1 449, DT_1 430~1 439, DT_2 150~2 159	0.919	4.425	0.440	9.780	1.320

图3 主成分分析(PCA)优化前的RF模型对土壤全铁含量的预测散点图 CSV,组合光谱变量集;MLR,多元线性回归;CARS,竞争性自适应重加权采样法;GA,遗传算法;SPA,连续投影法;UVE,无信息变量去除算法。R2,决定系数。下同。

Fig.3 Scatter plot of the predicted and measured soil total iron content by using RF model before principle component analysis (PCA) optimization CSV, Combination of spectral variables; MLR, Multiple linear regression; CARS, Competitive adaptive reweighted sampling; GA, Genetic algorithm; SPA, Successive projections algorithm; UVE, Uninformative variable elimination. R2, Coefficient of determination. The same as below.

表4 主成分分析(PCA)优化后土壤全铁含量反演RF模型的参数

Table 4 Parameters of RF model for soil total iron content inversion after principle component analysis (PCA) optimization

变量选择方法 Variable selection method	主成分个数 Number of principal components	累积贡献率 Cumulative contribution rate/%	在建模集上的效果 Performance on the modeling set		在验证集上的效果 Performance on the validation set
变量选择方法 Variable selection method	主成分个数 Number of principal components	累积贡献率 Cumulative contribution rate/%	R²	RMSE/(g·kg^-1)	R²	RMSE/(g·kg^-1)	RPD
CSV-PCA	6	95.83	0.908	4.713	0.888	4.375	2.951
MLR-PCA	4	95.92	0.906	4.778	0.893	4.277	3.019
CARS-PCA	5	96.14	0.936	3.925	0.910	3.922	3.292
GA-PCA	4	96.36	0.912	4.614	0.896	4.219	3.060
SPA-PCA	4	95.77	0.929	4.141	0.888	4.382	2.946
UVE-PCA	5	96.16	0.907	4.742	0.876	4.609	2.801

图4 主成分分析(PCA)优化后的RF模型对土壤全铁含量的预测散点图

Fig.4 Scatter plot of the predicted and measured soil total iron content by using RF model after principle component analysis (PCA) optimization

参考文献 34

[1]	PENG Y P, ZHAO L, HU Y M, et al. Prediction of soil nutrient contents using visible and near-infrared reflectance spectroscopy[J]. ISPRS International Journal of Geo-Information, 2019, 8(10): 437.
[2]	JIA X Y, O’CONNOR D, SHI Z, et al. VIRS based detection in combination with machine learning for mapping soil pollution[J]. Environmental Pollution, 2021, 268: 115845.
[3]	BEN-DOR E, BANIN A. Near-infrared analysis as a rapid method to simultaneously evaluate several soil properties[J]. Soil Science Society of America Journal, 1995, 59(2): 364-372.
[4]	何挺, 王静, 程烨, 等. 土壤氧化铁光谱特征研究[J]. 地理与地理信息科学, 2006, 22(2): 30-34.
	HE T, WANG J, CHENG Y, et al. Study on spectral features of soil Fe₂O₃[J]. Geography and Geo-Information Science, 2006, 22(2): 30-34. (in Chinese with English abstract)
[5]	彭杰, 向红英, 周清, 等. 土壤氧化铁的高光谱响应研究[J]. 光谱学与光谱分析, 2013, 33(2): 502-506.
	PENG J, XIANG H Y, ZHOU Q, et al. Influence of soil iron oxide on VNIR diffuse reflectance spectroscopy[J]. Spectroscopy and Spectral Analysis, 2013, 33(2): 502-506. (in Chinese with English abstract)
[6]	熊俊峰, 郑光辉, 林晨. 基于反射光谱的土壤铁元素含量估算[J]. 光谱学与光谱分析, 2016, 36(11): 3615-3619.
	XIONG J F, ZHENG G H, LIN C. Estimating soil iron content based on reflectance spectra[J]. Spectroscopy and Spectral Analysis, 2016, 36(11): 3615-3619. (in Chinese with English abstract)
[7]	谢文, 赵小敏, 郭熙, 等. 基于组合模型的庐山森林土壤有效铁光谱间接反演研究[J]. 土壤学报, 2017, 54(3): 601-612.
	XIE W, ZHAO X M, GUO X, et al. Composite-model-based indirect reversion of soil available iron spectrum of forest soil in Lushan[J]. Acta Pedologica Sinica, 2017, 54(3): 601-612. (in Chinese with English abstract)
[8]	丁海宁, 陈玉, 陈芸芝. 黄土高原土壤铁元素含量遥感反演方法[J]. 遥感技术与应用, 2019, 34(2): 275-283.
	DING H N, CHEN Y, CHEN Y Z. Remote sensing inversion method of soil iron content in the Loess Plateau[J]. Remote Sensing Technology and Application, 2019, 34(2): 275-283. (in Chinese with English abstract)
[9]	马驰. HJ-1A高光谱影像的表层土壤游离氧化铁含量反演[J]. 农业工程学报, 2020, 36(20): 164-170.
	MA C. Inversion of free ferric oxide content in surface soil based on HJ-1A hyperspectral images[J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(20): 164-170. (in Chinese with English abstract)
[10]	魏昌龙, 赵玉国, 邬登巍, 等. 基于光谱分析的土壤游离铁预测研究[J]. 土壤, 2014, 46(4): 678-683.
	WEI C L, ZHAO Y G, WU D W, et al. Prediction of soil free iron oxide content based on spectral analysis[J]. Soils, 2014, 46(4): 678-683. (in Chinese with English abstract)
[11]	贺军亮, 崔军丽, 张淑媛, 等. 基于偏最小二乘的土壤重金属铜含量高光谱估算[J]. 遥感技术与应用, 2019, 34(5): 998-1004.
	HE J L, CUI J L, ZHANG S Y, et al. Hyperspectral estimation of heavy metal Cu content in soil based on partial least square method[J]. Remote Sensing Technology and Application, 2019, 34(5): 998-1004. (in Chinese with English abstract)
[12]	任红艳, 庄大方, 邱冬生, 等. 矿区农田土壤砷污染的可见-近红外反射光谱分析研究[J]. 光谱学与光谱分析, 2009, 29(1): 114-118.
	REN H Y, ZHUANG D F, QIU D S, et al. Analysis of visible and near-infrared spectra of As-contaminated soil in croplands beside mines[J]. Spectroscopy and Spectral Analysis, 2009, 29(1): 114-118. (in Chinese with English abstract)
[13]	张笑寒, 孟祥添, 唐海涛, 等. 优化光谱输入量的土壤有机质随机森林预测模型[J]. 农业工程学报, 2023, 39(2): 90-99.
	ZHANG X H, MENG X T, TANG H T, et al. Random forest prediction model for the soil organic matter with optimized spectral inputs[J]. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(2): 90-99. (in Chinese with English abstract)
[14]	郭颖, 郭治兴, 刘佳, 等. 亚热带典型区域水稻土氧化铁高光谱反演: 以珠江三角洲为例[J]. 应用生态学报, 2017, 28(11): 3675-3683.
	GUO Y, GUO Z X, LIU J, et al. Hyperspectral inversion of paddy soil iron oxide in typical subtropical area with Pearl River Delta, China as illustration[J]. Chinese Journal of Applied Ecology, 2017, 28(11): 3675-3683. (in Chinese with English abstract)
[15]	高伟, 杨可明, 李孟倩, 等. 铁矿粉中全铁含量的SFIM-RFR高光谱预测模型[J]. 光谱学与光谱分析, 2020, 40(8): 2546-2551.
	GAO W, YANG K M, LI M Q, et al. Hyperspectral SFIM-RFR model on predicting the total iron contents of iron ore powders[J]. Spectroscopy and Spectral Analysis, 2020, 40(8): 2546-2551. (in Chinese with English abstract)
[16]	白宗璠, 韩玲, 姜旭海, 等. 微分光谱变换方法对土壤重金属含量反演精度的影响研究[J]. 光谱学与光谱分析, 2024, 44(5): 1449-1456.
	BAI Z F, HAN L, JIANG X H, et al. Effect of differential spectral transformation on soil heavy metal content inversion accuracy[J]. Spectroscopy and Spectral Analysis, 2024, 44(5): 1449-1456. (in Chinese with English abstract)
[17]	包青岭, 丁建丽, 王敬哲, 等. 基于随机森林算法的土壤有机质含量高光谱检测[J]. 干旱区地理, 2019, 42(6): 1404-1414.
	BAO Q L, DING J L, WANG J Z, et al. Hyperspectral detection of soil organic matter content based on random forest algorithm[J]. Arid Land Geography, 2019, 42(6): 1404-1414. (in Chinese with English abstract)
[18]	郭飞, 许镇, 马宏宏, 等. 基于PCA的土壤Cd含量高光谱反演模型对比研究[J]. 光谱学与光谱分析, 2021, 41(5): 1625-1630.
	GUO F, XU Z, MA H H, et al. A comparative study of the hyperspectral inversion models based on the PCA for retrieving the Cd content in the soil[J]. Spectroscopy and Spectral Analysis, 2021, 41(5): 1625-1630. (in Chinese with English abstract)
[19]	秦倩如, 齐雁冰, 吴娟, 等. 基于高光谱的土壤游离铁随机森林模型估算研究[J]. 土壤通报, 2018, 49(6): 1286-1293.
	QIN Q R, QI Y B, WU J, et al. Estimation of random forest model of soil free iron based on hyperspectral data[J]. Chinese Journal of Soil Science, 2018, 49(6): 1286-1293. (in Chinese with English abstract)
[20]	江振蓝, 杨玉盛, 沙晋明. 福州市土壤铬含量高光谱预测的GWR模型研究[J]. 生态学报, 2017, 37(23): 8117-8127.
	JIANG Z L, YANG Y S, SHA J M. Study on GWR model applied for hyperspectral prediction of soil chromium in Fuzhou City[J]. Acta Ecologica Sinica, 2017, 37(23): 8117-8127. (in Chinese with English abstract)
[21]	江振蓝, 杨玉盛, 沙晋明. GWR模型在土壤重金属高光谱预测中的应用[J]. 地理学报, 2017, 72(3): 533-544.
	JIANG Z L, YANG Y S, SHA J M. Application of GWR model in hyperspectral prediction of soil heavy metals[J]. Acta Geographica Sinica, 2017, 72(3): 533-544. (in Chinese with English abstract)
[22]	贺军亮, 韩超山, 韦锐, 等. 基于偏最小二乘的土壤重金属镉间接反演模型[J]. 国土资源遥感, 2019, 31(4): 96-103.
	HE J L, HAN C S, WEI R, et al. Research on indirect hyperspectral estimating model of heavy metal Cd based on partial least squares regression[J]. Remote Sensing for Land & Resources, 2019, 31(4): 96-103. (in Chinese with English abstract)
[23]	王文才, 赵刘, 李绍稳, 等. 基于特征波长选择和建模的高光谱土壤总氮含量估测方法研究[J]. 浙江农业学报, 2018, 30(9): 1576-1584.
	WANG W C, ZHAO L, LI S W, et al. Prediction of soil total nitrogen content from hyperspectral data based on charateristic wave-length selection and modelling[J]. Acta Agriculturae Zhejiangensis, 2018, 30(9): 1576-1584. (in Chinese with English abstract)
[24]	LEARDI R, LUPIÁÑEZ GONZÁLEZ A. Genetic algorithms applied to feature selection in PLS regression: how and when to use them[J]. Chemometrics and Intelligent Laboratory Systems, 1998, 41(2): 195-207.
[25]	钟翔君, 杨丽, 张东兴, 等. 砂壤潮土有机质含量可见-近红外光谱预测[J]. 光谱学与光谱分析, 2022, 42(9): 2924-2930.
	ZHONG X J, YANG L, ZHANG D X, et al. Prediction of organic matter content in sandy fluvo-aquic soil by visible-near infrared spectroscopy[J]. Spectroscopy and Spectral Analysis, 2022, 42(9): 2924-2930. (in Chinese with English abstract)
[26]	于雷, 洪永胜, 周勇, 等. 高光谱估算土壤有机质含量的波长变量筛选方法[J]. 农业工程学报, 2016, 32(13): 95-102.
	YU L, HONG Y S, ZHOU Y, et al. Wavelength variable selection methods for estimation of soil organic matter content using hyperspectral technique[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(13): 95-102. (in Chinese with English abstract)
[27]	BREIMAN L. Random forests[J]. Machine Learning, 2001, 45: 5-32.
[28]	SHERMAN D M, WAITE T D. Electronic spectra of Fe³⁺ oxides and oxide hydroxides in the near IR to near UV[J]. American Mineralogist, 1985, 70(11/12): 1262-1269.
[29]	SHEN Q, XIA K, ZHANG S W, et al. Hyperspectral indirect inversion of heavy-metal copper in reclaimed soil of iron ore area[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2019, 222: 117191.
[30]	郭云开, 张思爱, 谢晓峰, 等. 基于GA-SVM的耕地土壤重金属含量高光谱反演方法的研究[J]. 土壤通报, 2021, 52(4): 968-974.
	GUO Y K, ZHANG S A, XIE X F, et al. The hyperspectral inversion method of heavy metal contents in cultivated soils based on GA-SVM[J]. Chinese Journal of Soil Science, 2021, 52(4): 968-974. (in Chinese with English abstract)
[31]	王丽爱, 马昌, 周旭东, 等. 基于随机森林回归算法的小麦叶片SPAD值遥感估算[J]. 农业机械学报, 2015, 46(1): 259-265.
	WANG L A, MA C, ZHOU X D, et al. Estimation of wheat leaf SPAD value using RF algorithmic model and remote sensing data[J]. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(1): 259-265. (in Chinese with English abstract)
[32]	冯海宽, 杨福芹, 杨贵军, 等. 基于特征光谱参数的苹果叶片叶绿素含量估算[J]. 农业工程学报, 2018, 34(6): 182-188.
	FENG H K, YANG F Q, YANG G J, et al. Estimation of chlorophyll content in apple leaves base(d) on spectral feature parameters[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(6): 182-188. (in Chinese with English abstract)
[33]	张文钧, 蒋良孝, 张欢, 等. 一种基于偏差-方差权衡的贝叶斯分类学习框架[J]. 中国科学: 信息科学, 2023, 53(6): 1078-1095.
	ZHANG W J, JIANG L X, ZHANG H, et al. Bayesian classification learning framework based on bias-variance trade-off[J]. Scientia Sinica(Informationis), 2023, 53(6): 1078-1095. (in Chinese with English abstract)
[34]	崔禹, 韩玲, 韩霁昌, 等. 基于PCA-PLS的土壤重金属高光谱反演研究[J]. 甘肃科学学报, 2022, 34(2): 15-22.
	CUI Y, HAN L, HAN J C, et al. Hyperspectral inversion study of soil heavy metals based on PCA-PLS[J]. Journal of Gansu Sciences, 2022, 34(2): 15-22. (in Chinese with English abstract)

基于多光谱变换和主成分分析的土壤全铁含量随机森林模型反演

Inversion of soil total iron content using random forest model based on multi-spectral transformation and principle compoment analysis

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 34

相关文章 15

编辑推荐

Metrics

本文评价

[1]	谈亚丽, 高梦祥, 李晓洁, 周英杰, 熊健, 曾子琦, 李啸, 杨华. 湖北省主栽核桃在不同采收期的品质[J]. 浙江农业学报, 2025, 37(7): 1459-1468.
[2]	岳丽, 庄红梅, 祖力皮牙·买买提, 王佳敏, 毛红艳, 张英仙, 尼格尔热依·亚迪卡尔, 于明. 基于主成分分析与聚类分析的芜菁肉质根质地品质综合评价[J]. 浙江农业学报, 2025, 37(5): 1057-1071.
[3]	刘思彤, 侯宇, 潘家荃, 周桦楠, 崔亮, 万博, 于涛. 甘薯对低温胁迫的生理响应及耐寒性评价[J]. 浙江农业学报, 2025, 37(4): 767-778.
[4]	谈静如, 胡齐赞, 岳智臣, 陶鹏, 雷娟利, 李必元, 赵彦婷, 臧运祥. 基于叶绿素荧光参数的苗用型大白菜耐热性综合评价体系[J]. 浙江农业学报, 2025, 37(2): 288-299.
[5]	潘志洪, 温雪婷, 杨华, 吕文涛, 张俊杰, 肖英平. 番鸭肌肉氨基酸谱发育性变化研究[J]. 浙江农业学报, 2024, 36(9): 2010-2019.
[6]	薛贤滨, 贾琼, 陈峥峰, 黎瑞源, 陈庆富, 石桃雄. 基于主成分分析的苦荞麦重组自交系农艺性状综合评价[J]. 浙江农业学报, 2024, 36(4): 748-759.
[7]	王宇, 汪泓, 肖玖军, 李可相, 邢丹, 张永亮, 陈阳, 张蓝月. 基于优化植被指数组合的多品种辣椒叶片叶绿素数值估测[J]. 浙江农业学报, 2023, 35(9): 2109-2120.
[8]	叶雷, 张波, 杨学圳, 李小林, 张小平, 谭伟. 竹屑替代木屑栽培毛木耳的可行性及其品质综合评价[J]. 浙江农业学报, 2023, 35(6): 1416-1426.
[9]	张红梅, 王保君, 沈亚强, 程旺大. 浙北地区不同粒形优质粳稻产量和品质对播期调控的响应[J]. 浙江农业学报, 2023, 35(12): 2751-2762.
[10]	王如月, 罗莎莎, 甄紫怡, 吴嘉龙, 徐业勇, 孙雅丽, 胡晓静, 虎海防. 风味皇后杏李果实不同成熟度特性研究[J]. 浙江农业学报, 2023, 35(12): 2865-2877.
[11]	翟艺兰, 张楚磊, 楚爱香, 高俊鸽, 夏晴情, 卢志昌. 二十七种槭属植物表型多样性分析[J]. 浙江农业学报, 2023, 35(11): 2621-2635.
[12]	彭丹丹, 陈大刚, 徐开未, 游浩宇, 杨然, 廖慧苹, 陈远学. 椰糠复合基质对猕猴桃砧木幼苗生长及根系特征的影响[J]. 浙江农业学报, 2023, 35(10): 2364-2377.
[13]	姜昊梁, 黄允, 梁绍芳, 谢梦晨, 徐天成, 宋芷婷, 向文文, 陈青春, 万小荣, 孙伟. 镉胁迫对不同甜玉米自交系幼苗生长的影响及其相关简单重复序列分子标记初筛[J]. 浙江农业学报, 2022, 34(8): 1582-1590.
[14]	麻仲花, 吴娜, 陈娟, 赵匆, 闫承宏, 刘吉利. 盐胁迫与供磷水平对柳枝稷苗期生理特性的影响[J]. 浙江农业学报, 2022, 34(6): 1205-1216.
[15]	赵宇洪, 何文, 李根, 王强, 谢锐, 王燕, 陈清, 王小蓉. 四川地区琯溪蜜柚及其芽变品种的果实品质[J]. 浙江农业学报, 2022, 34(5): 995-1004.