二进制小波技术定量反演北方潮土土壤有机质含量

引用本文

王延仓, 杨秀峰, 赵起超, 顾晓鹤, 郭畅, 刘原萍. 二进制小波技术定量反演北方潮土土壤有机质含量[J]. 光谱学与光谱分析, 2019,39(9): 2855-2861.
WANG Yan-cang, YANG Xiu-feng, ZHAO Qi-chao, GU Xiao-he, GUO Chang, LIU Yuan-ping. Quantitative Inversion of Soil Organic Matter Content in Northern Alluvial Soil Based on Binary Wavelet Transform[J]. Spectroscopy and Spectral Analysis, 2019,39(9): 2855-2861.
Doi:10.3964/j.issn.1000-0593(2019)09-2855-07 复制到剪切板

Permissions

《光谱学与光谱分析》期刊社所有

二进制小波技术定量反演北方潮土土壤有机质含量

王延仓^1,³, 杨秀峰^1,³, 赵起超^1,³, 顾晓鹤^2,^4,^*, 郭畅^1,³, 刘原萍^1,³

1. 北华航天工业学院, 河北廊坊 065000

2. 北京农业信息技术研究中心, 北京 100097

3. 河北省航天遥感信息处理与应用协同创新中心, 河北廊坊 065000

4. 国家农业信息化工程技术研究中心, 北京 100097

*通讯联系人 e-mail: guxh@nercita.org.cn

作者简介: 王延仓, 1986年生, 北华航天工业学院讲师 e-mail: yancangwang@163.com

收稿日期: 2018-08-08 接受日期: 2018-12-25

基金: 国家重点研发计划(2016YFD0300609), 北京市农林科学院科技创新能力建设专项(KJCX20170705), 国家自然科学基金项目(41401419), 河北省青年基金项目(D2017409021)资助

摘要

为从土壤光谱中提取土壤有机质的光谱响应信息, 提升土壤有机质含量诊断精度与可靠性, 以潮土有机质含量为研究对象, 以北京市区域的96个耕层土壤参数与高光谱数据为数据源开展研究分析; 先采用二进制小波技术将土壤光谱数据分离为5个尺度的高频数据与低频数据, 再将低频数据、高频数据分别与土壤有机质实测数据进行相关性分析, 提取最佳波段组合, 构建有机质含量诊断模型。结果表明: (1)二进制小波技术可抑制噪声对高频信息的干扰, 能有效提升光谱对土壤有机质含量的敏感性, 进而提升有机质含量的诊断精度与可靠性; (2)在二进制小波技术下, 高频信息对有机质含量的诊断能力明显优于低频信息, 低频信息对土壤有机质含量的诊断能力随尺度增加而降低, 高频信息随尺度增加呈先提升而后降低的趋势; (3)与数学方法相比, 基于二进制小波变换算法构建的模型精度较高, 稳定性较好, 其最优模型的预测精度提高了31.5%, 可靠性增加了10.5%。

关键词: 土壤有机质; 二进制小波; 高光谱; 潮土

中图分类号:O657.3 文献标志码:A

Quantitative Inversion of Soil Organic Matter Content in Northern Alluvial Soil Based on Binary Wavelet Transform

WANG Yan-cang^1,³, YANG Xiu-feng^1,³, ZHAO Qi-chao^1,³, GU Xiao-he^2,^4,^*, GUO Chang^1,³, LIU Yuan-ping^1,³

1. Institute of Computer and Remote Sensing Information Technology, North China Institute of Aerospace Engineering, Langfang 065000, China

2. National Engineering Research Center for Information Technology in Agriculture, Beijing 100097, China

3. Aerospace Remote Sensing Information Processing and Application Collaborative Innovation Center of Hebei Province, Langfang 065000, China

4. Key Laboratory of Information Technology in Agriculture, Ministry of Agriculture, Beijing 100097, China

*Corresponding author

Abstract

In order to separate the information of the content of soil organic matter contained in soil spectra, to extract the spectral response information of the matter, to improve the diagnostic accuracy and reliability of soil organic matter content, this study takes the content of organic matter in tidal soil as the research object, and takes the soil parameters and hyperspectral data of 96 farmlands collected from Beijing area as the data source to research and analyze. First, the binary wavelet technique is used to separate the soil spectral data into 5 scales of high-frequency data and low-frequency data, and then these two kinds of data are respectively used for the correlation analysis with the measured soil organic matter data. Afterwards, the optimal band combination is extracted to build the diagnosis model of organic matter content. Finally, results of the study show that: (1) The binary wavelet technology can restrain the noise interference to high frequency information, and effectively enhance the spectral sensitivity to soil organic matter content so as to improve the diagnostic accuracy and reliability of organic matter content; (2) Under the binary wavelet technique, the diagnostic ability of high frequency information to organic matter content is obviously superior to that of low frequency information. The diagnostic ability of low frequency information to soil organic matter content decreases with the increase of scale, while high frequency information increases with the scale increasing and then decreases; (3) Compared with the mathematical method, the model based on the binary wavelet transform algorithm has higher accuracy and better stability. The prediction accuracy of the optimal model is improved by 31.5% and the reliability is increased by 10.5%.

Keyword: Soil organic matter; Binary wavelet transform; Hyperspectral; Alluvial soil

文章图片

引言

作为现代农业发展的重要方向, 精准农业是信息技术与传统农业相融合的结果, 其以现代农业科学知识为基础, 将农作过程进行精细设计、精准控制, 以实现节约资源、提升产出的目标^{[1, 2]}。因此, 实时、无损、准确获取田间耕层土壤水分、养分等的空间分布, 是实现精准农业关键环节。近年来, 光谱分析在土壤化学分析领域的应用得到了迅猛发展, 为耕层土壤养分的快速诊断提供新的思路^{[3, 4, 5]}。

有机质(soil organic matter, SOM)是田间耕层土壤的重要构成成分, 是作物的重要营养库, 为作物生长发育提供必需的微量、中量元素; 有机质在耕层土壤中的重金属转化、空气循环、结构改善等均具有重要作用, 是评价耕地质量的重要指标之一^{[6, 7, 8]}。耕层土壤有机质的理化性质主要受土壤类型^[9]、气候^[10]、轮作模式^[11]等多因子影响, 其含量具有动态平衡的特点, 因此, 实时、快速、精准获取土壤有机质含量具有重要的现实与理论意义。土壤有机质主要由胡敏酸、富里酸及胡敏素构成, 在可见光、近红外区域均有吸收作用; 胡敏酸与富里酸的化学结构相似, 它们具有相近的光谱特性, 而胡敏素则相对差异较大^{[12, 13, 14, 15]}。作物种植类型可影响土壤有机质的构成成分, 玉米秸秆可使土壤有机质中的脂族性成分增多, 增强有机质在1 450~1 625 nm的吸收作用^[13]; 土壤有机质与土壤光谱响应特征具有一定内在联系^{[16, 17, 18]}, 但在土壤有机质含量降低到一定限度后, 因受到水分^[19]、氧化铁^[20]等土壤组分的干扰, 有机质的光谱吸收特征较为微弱。为凸显或提取土壤光谱中的有机质信息, 采用数学变换^[21]、小波技术^[22]、主成分分析^[23]等对土壤光谱进行处理分析, 并利用偏最小二乘^[21]、随机森林^[24]、支持向量机^[25]、神经网络^[26]等回归分析算法构建土壤有机质含量估测模型。此外, 土壤有机质的光谱特性易受水分、氧化铁等土壤成分的干扰, 将水分、氧化铁等因素纳入到土壤有机质含量监测中可有效提升监测精度^{[20, 27]}。

受地理环境、气候条件与耕作模式的影响, 华北北部潮土土壤有机质含量大多在2%以下相对较低。有机质在土壤光谱中的吸收特征较为细弱, 与有机质相关的有益光谱信息多集中于高频数据内。因此, 如何将土壤光谱数据内的高频信息与低频信息分离, 并最大限度保留原始土壤光谱信息, 是一个亟待解决的技术难题。本工作利用二进制小波技术对土壤光谱进行的分解与重构, 实现将低频与高频信息的分离, 探索提升土壤有机质反演精度的方法。

1 实验部分

1.1 土样采集

以北京市密云区、通州区、顺义区为实验区, 选取面积较大、轮作模式为小麦/夏玉米的地块为研究对象, 采样时间为2011与2012年的10月, 共96份土样, 采样点的空间分布如图1所示。野外采样时, 利用四点混合法采集0~20 cm的表层土壤。将土样置于暗室内, 自然风干, 为消除土壤粒径大小对土壤光谱的影响, 对土样进行研磨、过筛20目处理, 最后ASD地物光谱仪测量土样室内光谱。采用重铬酸钾滴定法测定土壤有机质含量, 其统计性描述如表1所示。

	Figure Option View Download New Window
	图1 研究区域与样点空间分布Fig.1 The spatial distribution of study regions and samples

表1 土壤有机质含量的统计性描述 Table 1 The statistical description of soil organic matter content

1.2 光谱数据测量与处理

利用ASD公司生产的便携式光谱仪测量土样光谱数据, 光谱测量原理如图2所示: 利用1 000 W的卤光灯提供平行光束, 光束与垂线的夹角为15° ; 将土样置于黑色器皿内, 用直尺将其平面刮平, 并保持土壤厚度不低于2 cm; 利用直角装置将光谱仪的光纤置于土样上方30 cm处, 以获取垂直土样平面的光谱数据; 利用光谱常规测量流程采集土样光谱数据, 并将每个土样的光谱数据求平均后作为该土样的光谱。

	Figure Option View Download New Window
	图2 实验室内土壤光谱测量示意图Fig.2 The diagram of soil spectral measurement in laboratory

受外界环境、仪器暗电流等因素的影响, 获取的光谱数据内含较多噪声, 为削弱噪声的影响, 提升光谱数据的信噪比, 将信噪比较低的波段350~399和2 451~2 500 nm删除, 并采用海明窗低通滤波器对其余波段的光谱数据进行平滑处理, 最后将光谱分辨率降低至5 nm。

1.3 小波分析

小波分析是信号处理的一种新型技术, 具有多尺度分析与奇异点检测功能, 可利用低通与高通滤波器将信号分解为一系列高频与低频信号, 高频信号为原始信息中的细微信息, 低频信号为原始信息中的宏观信息, 两者可从细微与宏观角度深入分析原始信息, 小波分析的这一特点为光谱信号处理与分析提供了新的途径。

二进制小波是小波分析中的一类, 其可将原始信号中的高频信息(Detail, D)与低频信息(Approach, A)逐层分离(如图3所示)。由于潮土土壤有机质在土壤光谱上的表征较为微弱, 且易受外界环境的影响, 因此将二进制小波应用土壤光谱数据的处理与分析有助于提取土光谱中的有益信息。利用MATLAB语言实现二进制小波对土壤光谱数据分解与重构, 小波基为db5, 分解尺度为5。

	Figure Option View Download New Window
	图3 二进制小波的分解过程Fig.3 The decomposition process of binary wavelet

1.4 模型精度验证

光谱经二进制小波分解后, 将生成的高频信息与低频信息分别与土壤有机质进行相关性分析, 提取敏感波段, 并利用偏最小二乘算法(partial least squares regression, PLS)构建土壤有机质诊断模型。为更客观的反映建模精度, 选取2/3样本用于建模, 1/3样本用于验证, 并利用均方根误差(RMSE)与决定系数(R²)综合分析模型的精度, 其公式如下

$R^{2} = 1 - \frac{\overset{n}{\sum_{i = 0}} (SO M_{i} - SOM P_{i})^{2}}{\overset{n}{\sum_{i = 0}} (SO M_{i} {- \overset{M_{i}}{SO})}^{2}}$ (1)

$RMSE = \sqrt[]{\frac{\overset{n}{\sum_{i = 1}} (SO M_{i} - SOM P_{i})^{2}}{n}}$ (2)

式中, SOM为实测值, SOMP为基于诊断模型的预测值, $\bar{SOM}$ 为土壤有机质含量的均值。

2 结果与讨论

2.1 光谱分析

采用二进制小波将土壤光谱数据分解为低频与高频数据, 分解尺度为5, 然后基于不同尺度的低频(或高频)数据重构光谱曲线, 其结果如图4所示, 其中图4(a)为光谱曲线基于低频数据重构, 图4(b)基于高频数据。由图4(a)可知, 低频数据保存了原始光谱的形态特征, 但随着尺度增加, 光谱曲线呈逐渐平缓平滑, 吸收特征逐渐减弱或消失, 这表明随着分解尺度的增加, 二进制小波对光谱数据内的高频信息分离程度逐渐加深。从图4(b)分析知, 光谱曲线在400, 800, 1 400, 1 900及2 200 nm附近均有较强波动, 究其原因为原始光谱数据在这些波段附近具有相对较强的吸收或反射特征; 随着尺度的增加, 光谱曲线的取值范围逐渐变宽, 波峰、波谷宽度增长, 光谱波动强度呈先增强后减弱规律。综上分析可知, 二进制小波可有效分离光谱数据中的低频信息与高频信息, 能有效凸显土壤光谱中的吸收与反射特征。

	Figure Option View Download New Window
	图4 小波分析-低频(a)与高频(b)信息Fig.4 The wavelet analysis-low frequency (a) and high frequency (b) informations

2.2 相关性分析

图5为低频信息(a)、高频信息(b)分别与有机质含量的相关性分析结果; 由图5(a)可知, 低频信息与有机质含量的相关性曲线随分解尺度增加而逐渐光滑, 曲线在整体变化趋势方面相对一致, 其波动区间无明显变化, 而曲线的细微差异较为明显; 由图5(b)可知, 高频信息与土壤有机质含量的相关性曲线的波动频率随分解尺度的增加而逐渐降低(曲线趋于光滑), 相关系数R的变化区间较宽(最高达0.784, 最低达-0.757); 从(a)与(b)综合分析可知, 高频信息对有机质含量的敏感性明显优于低频信息, 高频信息与低频信息的相关性曲线均随分解尺度增加呈现相似的变化趋势, 究其原因是土壤有机质含量较低其对光谱响应较弱(多为细微信息), 而二进制小波变换可将弱信息进行分离, 从而呈现上述特征。

	Figure Option View Download New Window
	图5 低频信息(a)、高频信息(b)的相关性分析Fig.5 The correlation analysis of low frequency information (a) and high frequency information (b)

	Figure Option View Download New Window
	图6 4尺度模型的散点图Fig.6 The scatter plot of 4 scale model

2.3 传统数学变换建模

采用11种传统数学变换处理分析土壤光谱数据, 并构建土壤有机质含量诊断模型, 然后筛选出6类最优模型作为参照, 其中以对数变换模型的预测精度与稳定性最高, 为最优模型。该11种处理方法为: 倒数(1/R)、对数(LogR)、倒数的对数[log(1/R)]、一阶微分(R')、倒数的一阶微分[(1/R)']、对数的一阶微分[(LogR)']、倒数的对数的一阶微分[(log(1/R)']、除以R₉₃₀、除以R_(450~750)、吸收峰深度(H)、弓曲差(Gqc)^[21]。

2.4 二进制小波建模

利用二进制小波对土壤光谱数据分解为5个尺度的低频数据与高频数据, 然后利用各尺度低频(高频)数据重构土壤光谱; 通过将重构光谱数据与土壤有机质实测数据进行相关性分析, 提取敏感波段, 采用偏最小二乘算法构建土壤有机质含量诊断模型; 从建模与验证的决定系数R²与均方根误差(RMSE)综合评价模型的精度与稳定性。

表2 光谱反射率与土壤有机质含量的偏最小二乘回归模型列表 Table 2 The partial least squares regression model for spectral reflectance and soil organic matter content

表3为基于低频数据构建的土壤有机质含量诊断模型, 从表3对比分析可知, 基于重构光谱的模型的建模精度略低于原始光谱, 且建模精度随着尺度的增加而呈逐渐降低的趋势; 这是因为随着尺度的增加, 二进制小波变换对光谱中高频信息的分离程度持续加深所致。与原始光谱相比, 基于重构光谱的模型的验证精度略高, 各尺度的验证精度与建模精度的差距扩大, 模型的稳定性更差, 这表明随着高频信息的不断分离, 基于低频信息模型的建模精度不能正确表征模型的预测精度。

表3 基于低频数据土壤有机质含量诊断模型列表 Table 3 The list of models for estimating soil organic matter content based on low frequency data

基于高频信息构建的土壤有机质含量诊断模型如表4所示, 从模型列表分析可知, 各模型的建模精度和预测精度均较高, 模型的稳定性均相对较好, 其中以基于4尺度构建的模型精度与稳定性最高为最优模型, 其R²=0.685, RMSE=0.289; 随着尺度增加, 模型精度与稳定性均呈先增加后降低的趋势, 在第4尺度达到最高; 由构建模型的敏感波段分析可知, 在位于近红外区域的波段偏多, 而位于可见光区域的波段较少, 其原因主要为土壤有机质中的胡敏酸与富里酸在450 nm附近, 1 030~1 290, 1 400和2 100 nm均有较强的吸收作用, 研究区域的轮作模式多为小麦/玉米两季轮作, 土壤有机质中的脂类成分偏多, 增强了有机质对近红外区域的吸收作用。

表4 基于高频数据土壤有机质含量诊断模型列表 Table 4 The list of partial least squares regression analysis for the approach information of soil spectraland soil organic matter

2.5 二进制小波变换与传统数学变换建模的比较

由表2、表3、表4对比分析可知, 相对于传统数学变换中的微分模型, 基于高频信息构建的模型具有较高的预测精度与较好的稳定性, 表明利用二进制小波分离的高频信息可有效抑制噪声信息的干扰, 能有效提升光谱数据的信噪比; 与其他传统数学变换相比, 基于高频信息构建的模型较优, 高频信息的最优模型较传统数学变换的最优模型在预测精度上提升了31.5%, 稳定性提高了10.5%, 这表明二进制小波可有效提升光谱对土壤有机质含量的敏感性, 进而提升了模型的预测精度与稳定性; 基于高频信息构建的土壤有机质含量诊断模型优于低频信息, 这主要是因为有机质含量较低, 有益光谱信息多为细微信息。综上分析可知, 二进制小波既可抑制噪声信息的干扰, 也可有效提升光谱对土壤有机质含量的估测精度与可靠性。

在土壤有机质含量监测领域, 前人主要采用导数、对数、一阶微分等传统数学方法开展光谱数据的处理分析, 其在信息分离、噪声抑制、敏感度提升等具有较强的局限性, 且不易将土壤光谱内微弱有益信息进行整合, 因此未能取得理想效果; 本文采用二进制小波算法处理土壤光谱, 该算法与传统数学光谱处理方法相比, 可有效整合光谱内有益微弱信息, 并抑制高频噪声的影响, 提升光谱对土壤有机质含量的预测精度与稳定性, 最优模型精度模型提高了31.5%。土壤水分对土壤光谱具有较强的影响, 其对土壤有机质含量监测精度的提升具有较强的制约作用^[27], 研究分析土壤水分对土壤有机质含量估测的影响具有一定意义, 但由于本研究首先将土壤进行风干, 然后在实验室内测定其光谱及土壤参数, 故未考虑水分的影响。

目前报道较多的是利用神经网络、主成分分析、支持向量机等算法对土壤有机质含量定量反演特性, 及对土壤水分、铁等成分对土壤有机质含量监测的研究, 而对分离原始光谱数据的低频信息与高频信息的研究相对较少; 且已有研究表明土壤有机质的有益信息多集于高频信息内, 但高频信息内含噪声较多, 信噪比相对较低, 致使其构建的模型的预测精度与稳定性较低。本研究利用二进制小波处理分析土壤光谱数据, 研究分析二进制小波在信息分离与噪声抑制方面的作用, 结果表明二进制小波能有效分离光谱内的高频信息, 且分离的高频信息信噪比较高。本研究结论可为利用遥感技术快速、实时、精准、大尺度监测耕层土壤有机质含量的应用提供技术支撑, 但受实验条件制约与土壤有机质本身复杂性, 本研究仍存在一定不足, 需在以下几方面作改进:

(1)土壤光谱决定于土壤各构成成分的共同作用, 故耕层土壤内其他成分(特别是土壤水分与铁的含量)必然对土壤有机质含量的光谱响应存在一定影响或干扰, 该干扰的影响仍需开展进一步的深入研究。

(2)土壤光谱是在相对可控的环境条件下获取的, 但自然界不可控因素较多, 且复杂多变, 如何剔除外界因素的干扰, 将实验室研究成果转为现实可用技术, 仍需开展深入研究。

3 结论

利用二进制小波将土壤光谱分离为高频信息与低频信息, 将它们分别与土壤有机质含量进行相关性分析, 提取敏感波段, 并利用偏最小二乘算法构建土壤有机质含量诊断模型, 将其与11类数学变换方法进行对比分析得出如下结论:

(1)二进制小波处理中, 低频信息模型的建模精度与稳定性随分解尺度的增加而降低, 高频信息则呈先增加后降低的趋势, 其中基于4尺度高频信息构建的模型最优, R²=0.685和RMSE=0.289。

(2)二进制小波中, 高频信息的土壤有机质含量诊断精度与模型的稳定性明显优于低频信息。

(3)二进制小波可有效提升土壤有机质含量诊断精度与可靠性: 与传统数学变换相比, 二进制小波变换构建的模型精度较高, 稳定性较好, 其最优模型的预测精度提高了31.5%, 可靠性增加了10.5%。

(4)二进制小波可有效降低噪声, 增强模型的稳定性: 与传统微分变换相比, 经二进制小波分离后, 高频信息对土壤有机质含量诊断的稳定性明显较高, 最高增加了22.5%。

参考文献

文献列表

[1]	LIU Yan-sui, ZHANG Zi-wen, WANG Jie-yong(刘彦随, 张紫雯, 王介勇). Acta Geographica Sinica(地理学报), 2018, 73(2): 203. [本文引用:1]
[2]	Bongiovanni R, Lowenberg Deboer J. Precision Agricultural, 2004, (5): 359. [本文引用:1]
[3]	YANG Jian-feng, MA Jun-cheng, WANG Ling-chao(杨建锋, 马军成, 王令超). Geospatial Information(地理空间信息), 2015, 13(2): 47. [本文引用:1]
[4]	JI Wen-jun, SHI Zhou, ZHOU Qing, et al(纪文君, 史舟, 周清, 等). Journal of Infrared and Millimeter Waves(红外与毫米波学报), 2012, 31(3): 277. [本文引用:1]
[5]	CHEN Hong-yan, ZHAO Geng-xing, LI Xi-can, et al(陈红艳, 赵庚星, 李希灿, 等). Scientia Agricultura Sinica(中国农业科学), 2012, 45(7): 1425. [本文引用:1]
[6]	St Luce M, Ziadi N, Zebarth B J, et al. Geoderma, 2014, (232-234): 449. [本文引用:1]
[7]	Ahmed Z, Iqbal J. European Journal of Remote Sensing, 2014, 47(1): 557. [本文引用:1]
[8]	ZHAI Xing-yu, ZHANG Xing-yu, LI Hao, et al(翟星雨, 张兴义, 李浩, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2018, 34(19): 155. [本文引用:1]
[9]	LU Yan-li, BAI You-lu, YANG Li-ping, et al(卢艳丽, 白由路, 杨俐苹, 等). Journal of Plant Nutrition and Fertilizers(植物营养与肥料学报), 2011, 17(2): 456. [本文引用:1]
[10]	GUO Guang-fen, ZHANG Chen-yi, XU Ying(郭广芬, 张称意, 徐影). Chinese Journal of Ecology(生态学杂志), 2006, (4): 435. [本文引用:1]
[11]	LAN Yan, HUANG Guo-qin, YANG Bin-juan, et al(兰延, 黄国勤, 杨滨娟, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2014, 30(13): 146. [本文引用:1]
[12]	LI Wen-hui, LI Ai-min, ZHANG Shu-cai, et al(李文慧, 李爱民, 张树才, 等). Environmental Chemistry(环境化学), 2008, (4): 503. [本文引用:1]
[13]	SUN Xiang-ping, LI Guo-xue, XIAO Ai-ping, et al(孙向平, 李国学, 肖爱平, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2014, 34(9): 2413. [本文引用:2]
[14]	LI Ye, ZHOU Cong-cong, DAI Ling-xing, et al(黎烨, 周聪聪, 戴零星, 等). Acta Scientiae Circumstantiae(环境科学学报), 2017, 37(3): 1098. [本文引用:1]
[15]	JIANG Jie, LI Li, SUN Guo-xin(姜杰, 李黎, 孙国新). Environmental Chemistry(环境化学), 2012, 31(12): 2002. [本文引用:1]
[16]	Henderson T L, Baumgardner M F, Franzmeier D P, et al. Soil Science Society of America Journal, 1992, 56(3): 865. [本文引用:1]
[17]	Ben-Dor E, Banin A. Soil Science Society of America Journal, 1995, 59(2): 364. [本文引用:1]
[18]	Mccarty G W, Reeves J B, Reeves V B, et al. Soil Science Society of America Journal, 2002, 66(2): 640. [本文引用:1]
[19]	CHEN Hong-yan, ZHAO Geng-xing, ZHANG Xiao-hui, et al(陈红艳, 赵庚星, 张晓辉, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2014, 30(8): 91. [本文引用:1]
[20]	PENG Jie, LI Xi, ZHOU Qing, et al(彭杰, 李曦, 周清, 等). Journal of Remote Sensing(遥感学报), 2013, 17(6): 1396. [本文引用:2]
[21]	Di Nezio M S, Pistonesi M F, Fragoso W D, et al. Microchemical Journal, 2007, 85(2): 194. [本文引用:3]
[22]	YANG Chang-bao, LI Dong-hui, LIU Jin-yi, et al(杨长保, 李东辉, 刘津怿, 等). Journal of Basic Science and Engineering(应用基础与工程科学学报), 2017, 25(5): 869. [本文引用:1]
[23]	YE Qin, JIANG Xue-qin, LI Xi-can, et al(叶勤, 姜雪芹, 李西灿, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2017, 48(3): 164. [本文引用:1]
[24]	Wiesmeier M, Barthold F, Blank B, et al. Plant & Soil, 2011, 340(1/2): 7. [本文引用:1]
[25]	ZENG Yin, LU Yu-zhen, DU Chang-wen, et al(曾胤, 陆宇振, 杜昌文, 等). Acta Pedologica Sinica(土壤学报), 2014, 51(6): 1262. [本文引用:1]
[26]	ZHANG Juan-juan, TIAN Yong-chao, ZHU Yan, et al(张娟娟, 田永超, 朱艳, 等). Scientia Agricultural Sinica(中国农业科学), 2009, 42(9): 3154. [本文引用:1]
[27]	SI Hai-qing, YAO Yan-min, WANG De-ying, et al(司海青, 姚艳敏, 王德营, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2015, 31(9): 114. [本文引用:2]

2018

0.0

... 引言作为现代农业发展的重要方向, 精准农业是信息技术与传统农业相融合的结果, 其以现代农业科学知识为基础, 将农作过程进行精细设计、精准控制, 以实现节约资源、提升产出的目标^[1,2] ...

2004

0.0

2015

0.0

... 近年来, 光谱分析在土壤化学分析领域的应用得到了迅猛发展, 为耕层土壤养分的快速诊断提供新的思路^[3,4,5] ...

2012

0.0

... 近年来, 光谱分析在土壤化学分析领域的应用得到了迅猛发展, 为耕层土壤养分的快速诊断提供新的思路^[3,4,5] ...

2012

0.0

CHEN

Hong-yan

, ZHAO

Geng-xing

, LI

Xi-can

, et al(陈红艳, 赵庚星, 李希灿, 等). Scientia Agricultura Sinica(中国农业科学), 2012, 45(7): 1425.

【Objective】 This study aimed at improving the precision and practicability of the soil available potassium estimation model by removing the noise of soil hyperspectral reflectance. 【Method】 Seventy-six soil samples with similar soil organic matter, nitrogen and phosphorus element contents and different potassium element contents were selected. The first derivative spectrum of the soil sample logarithmic reflectance was decomposed to multiple levels by using four kinds of wavelet function, respectively. The low frequency wavelet coefficients were obtained, and the hyperspectral estimation models of soil available potassium content were built. 【Result】The results showed that the low frequency wavelet coefficients of 1-3 levels could represent the original spectrum. Based on the low frequency coefficients of different wavelet functions at the same level, the precise of soil available potassium estimation model showed a little difference. The model built with the low frequency coefficient of the second decomposition level using the Bior 1.3 function had comparatively high accuracy and was chosen as the best model. With the data reducing to 25% and reflecting 95.6% information of the input spectrum, the model building R2 reached 0.976 and RMSE was 10.66 mg•kg-1, which was validated to have fairly good forecast accuracy. 【Conclusion】Therefore, wavelet analysis for obtaining wavelet coefficients can not only extract the soil hyperspectral information, but also compress data, which is feasible to forecast soil potassium content in combination with partial least squares regression method.

1.山东农业大学资源与环境学院，山东泰安 271018 2.山东农业大学信息科学与工程学院，山东泰安 271018 3.山东农业大学园艺科学与工程学院，山东泰安 271018 4.山东省沾化县国土资源局，山东沾化 256800

【目的】对土壤高光谱数据去噪提纯，提高土壤速效钾含量高光谱估测模型的精度和实用性。【方法】选取土壤有机质、碱解氮、有效磷含量近似而速效钾含量差异较大的样本76个，对土壤样本反射率对数的一阶导数光谱分别基于4种函数进行多层小波离散分解；提取小波低频系数，构建土壤速效钾含量高光谱估测模型。【结果】小波分解1—3层获得的低频系数可用以代表原始光谱。基于各小波函数相同尺度的低频系数，土壤速效钾含量估测建模精度差异不大。其中基于Bior 1.3函数分解的第2层低频系数建模精度略高，作为最佳估测模型，在数据压缩到25%、反映输入光谱信息95.6%的基础上，建模R2达到0.976，RMSE为10.66 mg•kg-1，经验证模型具有较好的预测准确度。【结论】通过小波分析获得小波系数，既提取了土壤高光谱信息，又对数据进行了压缩，结合偏最小二乘回归预测土壤速效钾含量是可行的。

... 近年来, 光谱分析在土壤化学分析领域的应用得到了迅猛发展, 为耕层土壤养分的快速诊断提供新的思路^[3,4,5] ...

2014

0.0

... 有机质在耕层土壤中的重金属转化、空气循环、结构改善等均具有重要作用, 是评价耕地质量的重要指标之一^[6,7,8] ...

2014

0.0

... 有机质在耕层土壤中的重金属转化、空气循环、结构改善等均具有重要作用, 是评价耕地质量的重要指标之一^[6,7,8] ...

2018

0.0

ZHAI

Xing-yu

, ZHANG

Xing-yu

, LI

Hao

, et al(翟星雨, 张兴义, 李浩, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2018, 34(19): 155.

为了明晰等高垄作后对坡耕地土壤有机质的恢复作用，选取一块面积为1.4 hm2的典型黑土坡耕地，采用标准栅格法，同位大样点取样调查了改垄前和改垄10 a后土壤有机质和含水率等性状的变化。结果表明：1）与经典统计学相比，地统计学通过变程、块金值与基台值的比值以及插值绘制空间分布图，能够从全坡面更好地反映改垄前后性状的空间变化；2）等高改垄10 a后，垄台土壤含水率变程由510.7 m降低到193.2 m，块金值与基台值的比值由11.7%升至46.9%，空间相关性强度由强烈变为中等，水分再分配降低；3）耕层土壤有机质总体增加了2.61 g/kg，提升了8.4%，只在坡中上部西侧小区的部分区域降低了3.7%；4）土壤全氮含量减少了0.04 g/kg，降低了2.8%。上述结果表明，对于严重侵蚀的坡耕地，改顺坡垄为等高垄作，可弱化性状空间相关性，对土壤有机质具有恢复作用，但应适当增加化肥氮素的施用量，对东北黑土区坡耕地水土流失治理具有指导意义。

... 有机质在耕层土壤中的重金属转化、空气循环、结构改善等均具有重要作用, 是评价耕地质量的重要指标之一^[6,7,8] ...

2011

0.0

Yan-li

, BAI

You-lu

, YANG

Li-ping

, et al(卢艳丽, 白由路, 杨俐苹, 等). Journal of Plant Nutrition and Fertilizers(植物营养与肥料学报), 2011, 17(2): 456.

Organic matter (OM) content is an important index in reflecting soil fertility, and provides important information for soil digital management and precise fertilization if OM can be estimated using hyperspectral technology. In this experiment, NIR-Visible spectral reflectance of soil was measured using ASD FR 2500 hyperspectral meter. The spectral reflectance curve shapes and the OM prediction models of different soil types were analyzed. The results show that, according to the variation of spectral reflectance curves in the Northeast Plain, those soil types can be divided preliminarily into three categories, and their main differences are the curve slopes in 460–1350nm. Further, the sensitive bands to SOM were determined as 650–750nm by correlation analysis. The SOM regression models of different soil types were established using reflectance in the sensitive bands, and the homogeneity of variance is tested for regression coefficients and intercept. The results indicate that black and chernozem, and black and dark brown soil pass the homogeneity test respectively. Both of the two homogenous categories can be predicted using common models, but the predicting accuracy is slightly lower.

Institute of Agricultural Resources and Regional Planning/Ministry of Agriculture Key Laboratory of Crop Nutrition and Fertilization , Beijing 100081

土壤有机质含量是反映土壤肥力的重要特征，利用高光谱技术对耕层土壤有机质（OM）含量进行定量化反演可为土壤信息化管理和精准施肥提供重要的依据。本文利用ASD FR 2500高光谱仪在室内条件下测定了风干土壤样品的可见—近红外光谱，分析了东北平原不同土壤类型土壤光谱反射率曲线形状变化及土壤有机质含量的光谱反演模型，结果表明，根据土壤光谱反射率曲线变化特点可以初步将东北平原的几种土壤类型划分为三类，其主要区别在于460～1350nm范围内曲线斜率变化的差异。通过相关分析进一步确定了土壤有机质敏感波段为650～750nm，建立了不同土壤类型的有机质含量回归预测模型，并且对不同模型的回归系数和截距进行了方差同质性检验。结果显示，黑土和黑钙土、黑土和暗棕壤的模型没有显著差异，即可以分别构建同质性模型来简化有机质的预测，但是预测精度有所降低。

... 耕层土壤有机质的理化性质主要受土壤类型^[9]、气候^[10]、轮作模式^[11]等多因子影响, 其含量具有动态平衡的特点, 因此, 实时、快速、精准获取土壤有机质含量具有重要的现实与理论意义 ...

2006

0.0

GUO

Guang-fen

, ZHANG

Chen-yi

, XU

Ying

(郭广芬, 张称意, 徐影). Chinese Journal of Ecology(生态学杂志), 2006, (4): 435.

To understand the impacts of climate change on the soil organic carbon storage i n terrestrial ecosystems is of significance in predicting the changes of soil or ganic carbon in the future,and the interactions between terrestrial ecosystems and climate change.We presented the approaches of soil profile measurem ent,soil organic carbon modeling,and others to estimate the soil organic carbo n storage in terrestrial ecosystems,and the uncertainty of these estimations.T he research progress in understanding the effects of climate change on soil orga nic carbon storage in terrestrial ecosystems and the modeling of corresponding p rocesses were reviewed,and the estimations of soil organic carbon storage chang es under future climate changes were pointed out,with some questions in the est imations put forward.

1 Postgraduate Section,Chinese Academy of Meteorological Sciences,Beijing 100081,China; 2 Laboratory for Climate Studies,National Climate Center,Chin a Meteorological Administration,Beijing 100081,China

通过研究气候变化对土壤有机碳储藏的影响，对预测未来气候变化下土壤有机碳动态变化与深入理解陆地生态系统变化和气候变化之间的相互作用有着极其重要的意义。本文归纳了土壤类型法、模型模拟法等途径对土壤有机碳储量估算的结果并分析它们各自的不确定性，综述了气候变化对土壤碳贮藏影响机理的研究与相应过程模拟的模型研究进展，并综合分析了当前研究中还存在的问题与不足。

2014

0.0

LAN

Yan

, HUANG

Guo-qin

, YANG

Bin-juan

, et al(兰延, 黄国勤, 杨滨娟, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2014, 30(13): 146.

Abstract: The area of fall-planted cover crop in China is quite low in paddy field, and multiple cropping index declines yearly, resulting in tremendous waste of soil resources. This reduction in fall-planted cover crop can lead to decrease in soil fertility. Most paddy field in southern China is suitable for growing winter green manure because of abundant light and heat resources. Winter green manure in general contains 15% - 20% of the organic matter. It can be used to improve soil nutrients cycling and facilitate the mineralization of soil organic matter. Comprehensive evaluation of green manures in different crop rotations in paddy fields of southern China region is needed in order to develop sustainable multiple cropping systems. The objective of this study was to determine the effect of green manure on different cropping systems (A: fallow-early rice-late rice→fallow-early rice-late rice (CK); B: milk vetch- rice- rice→rapeseed/peanut-rice; C: rapeseed/peanut- rice→ potato/maize+ soybean- rice; D: potato/maize+soybean- rice→vegetable/peanut+maize- rice; E: vegetable/peanut+maize- rice→milk vetch- rice- rice) on soil nutrients, soil organic carbon, and carbon pool management indexes in southern China where two rice crops per year were common. The experiment was conducted during September 2011 - November 2013. Soil samples (at depths of 10-20 cm) were taken after rice was harvested in 2013 for analysis. The results showed that compared with CK, green manure rotation treatments had significantly (P<0.05) improved organic matter, alkaline hydrolytic nitrogen and available potassium by 8.73%-15.59%, 11.79%-19.64%, and 5.80%-37.19%, respectively. Soil organic of green manure rotation treatments of B and C were 15.59% and 11.19% higher (P<0.05) than CK. Soil active organic carbon of all green manure rotation treatments was 29.03%, 19.96%, 16.67% and 21.75% respectively higher (P<0.05) than CK. Compared with CK, the soil carbon pool index significantly (P<0.05) increased by 15.70% for the treatment B, and soil carbon management index significantly (P<0.05) increased by 32.07 for the treatment B, 21.98 for the treatment C, 18.64 for the treatment D and 24.47 for the treatment E, respectively. Among the treatments, milk vetch- rice- rice→rapeseed/peanut- rice had the highest values for the soil carbon pool index and soil carbon management index. The soil organic matter and total nitrogen, available phosphorus, total organic carbon, active organic carbon, and carbon pool management index were highly significantly (P<0.01) correlated. Soil total nitrogen, alkaline hydrolytic nitrogen, and available phosphorus had a significant (P<0.05) correlation with carbon pool management index. In addition, soil organic carbon was significantly (P<0.05) positively correlated with active organic carbon and carbon pool management indexes. Therefore, green manure rotation, especially milk vetch- rice- rice→rapeseed/peanut- rice could increase soil organic carbon contents and soil carbon pool management index, improve soil quality and soil fertility under the local soil fertility conditions, and was important for promoting sustainable development of agriculture.

综合评价中国南方不同稻田绿肥轮作模式，筛选与优化绿色高效的稻田多熟种植系统，有利于南方双季稻区农业可持续发展。通过田间试验，研究不同绿肥轮作模式（A冬闲-稻-稻→冬闲-稻-稻（对照）、B紫云英-稻-稻→油菜/花生-稻、C油菜/花生-稻→马铃薯/玉米+大豆-稻、D马铃薯/玉米+大豆-稻→蔬菜/花生+玉米-稻、E蔬菜/花生+玉米-稻→紫云英-稻-稻）对土壤养分中pH值、有机质、全氮、碱解氮、有效磷和速效钾及总有机碳库、活性有机碳、碳库管理指数的影响。结果表明：绿肥轮作各处理的土壤有机质、碱解氮和速效钾均显著高于对照处理（P<0.05），增加幅度分别为8.73%～15.59%、11.79%～19.64%和5.80%～37.19%。绿肥轮作处理B、C的总有机碳质量分数与对照相比有显著（P<0.05）增加，分别增加了15.59%、11.19%。与对照相比，绿肥轮作B、C、D、E均可以显著（P<0.05）提升土壤活性有机碳质量分数，提高幅度分别为29.03%、19.96%、16.67%和21.75%。与冬闲对照处理相比，绿肥轮作系统中处理B的土壤碳库指数显著（P<0.05）增加了15.70%。相比冬闲对照处理，绿肥轮作处理B、C、D、E的土壤碳库管理指数分别显著（P<0.05）增加了32.07、21.32、17.91和23.74，其中紫云英-稻-稻→油菜/花生-稻种植模式达到最大。土壤有机质与全氮、有效磷、总有机碳、活性有机碳、碳库管理指数存在极显著（P<0.01）的相关性，土壤全氮、碱解氮、有效磷均与碳库管理指数存在显著的相关性（P＜0.05）。此外，总有机碳与活性有机碳和碳库管理指数存在极显著的相关性（P<0.01）。可见在当地土壤肥力条件下，绿肥轮作尤其是紫云英-稻-稻-油菜/花生-稻能提高土壤有机碳质量分数和土壤碳库管理指数，有利于改善土壤质量，提高土壤肥力，为建立农田可持续的土壤管理和土地利用提供科学依据和参考价值。

2008

0.0

Wen-hui

, LI

Ai-min

, ZHANG

Shu-cai

, et al(李文慧, 李爱民, 张树才, 等). Environmental Chemistry(环境化学), 2008, (4): 503.

利用紫外-可见和荧光光谱研究土壤中腐殖酸的光谱特性.结果显示,富里酸在0-45mg·l-1浓度范围内,其浓度与荧光强度呈线性正相关,浓度大于45 mg·l-1时,二者的线性关系趋弱,非线性关系明显.pH值对荧光强度有显著的影响,酸性条件下荧光强度随pH值的增大而增强,pH值在4.5-10之间的变化不显著,在中性和碱性条件下有较高的荧光强度,当pH值超过10以后有所减弱.荧光光谱在酸性条件下呈现一定程度的蓝移,碱性条件下则呈现红移.

... 胡敏酸与富里酸的化学结构相似, 它们具有相近的光谱特性, 而胡敏素则相对差异较大^{[12,13,14,15]} ...

2014

0.0

SUN

Xiang-ping

, LI

Guo-xue

, XIAO

Ai-ping

, et al(孙向平, 李国学, 肖爱平, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2014, 34(9): 2413.

Using pig manure and corn straw as raw materials for high-temperature composting, setting three different treatments; C/N 15, C/N 25, and C/N 35. Composting period is 120 days, which contains 30 days for ventilation cycle by forced continuous ventilation. Sampled on 0, 22, 30, 60, 90, 120th days, they were analyzed by elemental analysis and IR spectroscopy to study effect of different lignin content on compost humic acid ( HA) composition and molecular structure. The results showed that the change in composting humic acid C focused on the first 30 days, while after composting, the O/C of compost HA increased, H/C decreased, and N content increased. Low C/N (15) and higher C/N ratio (35) had higher degree of oxidation than the C/N 25 in compost HA. FTIR indicated that the infrared spectrum shapes with different lignin content treatment are similar during the composting process, but the peak intensity is obviously different. Research results proved that the composting stage is more conducive to enhanced aromatic in compost HA. After composting, C/N 15 had less polysaccharide and fat ingredients and more aromatic structural components in compost HA, compared with C/N 25 and 35. In addition, compost HA of C/N 15 had higher degree of humification and its structure was more stable.

[Sun Xiang-ping; Xiao Ai-ping] Chinese Acad Agr Sci, Inst Bast Fiber Crops, Changsha 410125, Hunan, Peoples R China.;[Sun Xiang-ping; Li Guo-xue; Shi Hong; Wang Yi-ming; Li Yang-yang] China Agr Univ, Coll Resources & Environm Sci, Beijing 100193, Peoples R China.

... 胡敏酸与富里酸的化学结构相似, 它们具有相近的光谱特性, 而胡敏素则相对差异较大^{[12,13,14,15]} ...

... 作物种植类型可影响土壤有机质的构成成分, 玉米秸秆可使土壤有机质中的脂族性成分增多, 增强有机质在1 450~1 625 nm的吸收作用^[13] ...

2017

0.0

, ZHOU

Cong-cong

, DAI

Ling-xing

, et al(黎烨, 周聪聪, 戴零星, 等). Acta Scientiae Circumstantiae(环境科学学报), 2017, 37(3): 1098.

Soil samples at Dianchi Lake basin were collected in July 2014 and measured by ultraviolet-visible absorption spectroscopy and three-dimensional fluorescence spectroscopy (3DEEMs) in combination with carbon concentration. Based on these measured data, the spectrum characteristics of soil active humus as well as its compositions of humic acid and fulvic acid were studied. The results indicated that soil active humus in Dianchi Lake basin contained much more fulvic acid than humic acid. The distribution of soil active humus and its compositions were heterogeneous in this region, with the variation coefficient of soil active humus, humic acid and fulvic acid of 37.7%, 36.7% and 40.9%, respectively. The carbon concentration of active soil humus, humic acid and fulvic acid were in the same order of vegetable field < wasteland < grassland < wet land. The formation of active soil humus consisted of endogenous substances and partly degraded exogenic substances. A positive correlation between photochemical activity and endogenous characteristic was found in our study. Therefore, the photochemical and bio-availability activity in soil active humus was weak due to its weak endogenous characteristic in Dianchi Lake basin.

利用紫外-可见光谱和三维荧光光谱技术，结合土壤腐殖质及腐殖质组分（胡敏酸、富里酸）中的有机碳含量，研究了滇池流域土壤活性腐殖质及其组分胡敏酸和富里酸的光谱特征.结果表明，滇池流域土壤活性腐殖质中富里酸含量高于胡敏酸，流域范围内土壤活性腐殖质及其组分含量分布不均匀，活性腐殖质、胡敏酸和富里酸含量的变异系数分别达37.7%、36.4%、40.9%.土壤活性腐殖质及其组分胡敏酸和富里酸的碳含量均表现为菜地 < 荒地 < 草地 < 湿地.光谱特征结果表明，滇池流域土壤活性腐殖质主要由自生源物质和经一定程度降解后的外源物质共同组成.此外，自生源性与光化学及生物活性具有正相关关系.因此，具有弱自生源性的滇池流域土壤活性腐殖质所具有的光化学和生物活性较弱.

... 胡敏酸与富里酸的化学结构相似, 它们具有相近的光谱特性, 而胡敏素则相对差异较大^{[12,13,14,15]} ...

2012

0.0

... 胡敏酸与富里酸的化学结构相似, 它们具有相近的光谱特性, 而胡敏素则相对差异较大^{[12,13,14,15]} ...

1992

0.0

... 土壤有机质与土壤光谱响应特征具有一定内在联系^[16,17,18], 但在土壤有机质含量降低到一定限度后, 因受到水分^[19]、氧化铁^[20]等土壤组分的干扰, 有机质的光谱吸收特征较为微弱 ...

1995

0.0

2002

0.0

2014

0.0

CHEN

Hong-yan

, ZHAO

Geng-xing

, ZHANG

Xiao-hui

, et al(陈红艳, 赵庚星, 张晓辉, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2014, 30(8): 91.

Abstract: Soil moisture is a key issue in using spectrum analysis method to predict soil nutrients content. The purpose of this article is to explore a method of removing the effect of soil moisture and improving the hyperspectra estimation precision of soil organic matter (SOM) content. Firstly the soil samples were collected from agricultural fields of the brown soil in Daiyue county and the cinnamon soil in Huantai county, Shandong province, China. The hyperspectra of the moisture and sieved dry soil samples were measured using the ASD FieldSpec 3 and transformed to the first deviation. Because the soil moisture content and its coefficient of variation (CV) of the brown soil samples was relatively high, the brown soil samples were divided into two groups, additionally the all brown soil samples and the cinnamon soil samples, here there were four-group soil samples. Secondly, based on the difference between the moisture and dry spectra, the characteristic spectra of soil moisture were selected by singular value decomposition (SVD) in combination with correlation analysis, then the correcting coefficients of removing moisture factor from soil hyperspectra were built to reconstruct the corrected spectra of the wet samples. Finally the estimation models of the soil organic matter content were built using the partial least squares (PLS) regression based on the uncorrected and corrected spectra of the wet samples. The results indicated that using singular value decomposition to correct the moisture spectra could partly reduce the correlation coefficients between the soil moisture content and the hyperspectra in most range of spectra, and for the four-group soil samples including two for each brown soil grouped by the soil moisture content gradient, all brown soil and cinnamon soil, the coefficient of determination (R2) and relative prediction deviation (RPD) of models based on the corrected spectra were improved signally with the calibration R2 of 0.85, 0.82, 0.74 and 0.76 (an increase of 0.02-0.09), and the calibration root mean squares error (RMSE) of 0.19%, 0.20%, 0.23% and 0.19% (reduce of 0.01% - 0.03%), the validation R2 of 0.78, 0.77, 0.72, and 0.76 (an increase of 0.06 and 0.15) the validation RMSE of 0.21%, 0.15%, 0.21% and 0.15% (reduce of 0.01% - 0.08%) except for the cinnamon soil samples (increased 0.02 percentage), the validation RPD of 2.03, 2.02, 1.86 and 1.98 (an increase of 0.17 - 0.43), especially for the three-group samples with the smaller CV in soil moisture content. The models reached the good performance from needing improvement and could achieve better prediction accuracy of the soil organic matter content. Therefore, the experiments indicated that the method was effective to remove the soil moisture influence and improve the prediction accuracy of the soil organic matter content from hyperspectra. In addition, in order to achieve the better prediction accuracy, the soil samples should be grouped by the soil moisture content to reduce the dispersion degree of the soil moisture.

土壤水分的影响是当前采用光谱分析法预测土壤养分含量的关键问题，该文旨在探索去除土壤水分影响、提高有机质高光谱定量估测精度的方法。首先采用地物光谱仪进行湿土和过筛干土的高光谱测试，并进行一阶导数变换；然后，采用奇异值分解（singular value decomposition，SVD）结合相关分析筛选土壤水分特征光谱，构建去除水分因素的修正系数，形成湿土光谱的校正光谱；最后基于校正前后湿土光谱，应用偏最小二乘（partial least squares，PLS）回归构建土壤有机质含量的估测模型，并对模型进行验证和比较，分析评价校正前后光谱的预测精度。结果显示：按土壤水分含量梯度划分的2组和全部棕壤及褐土土样共4组样本校正后建模决定系数和均方根误差分别为0.85、0.82、0.74、0.76和0.19%、0.20%、0.23%、0.19%，决定系数提高了0.02～0.09，均方根误差降低了0.01～0.03百分点，验证决定系数、均方根误差和相对分析误差分别为0.78、0.77、0.72、0.76，0.21%、0.15%、0.21%、0.15%和2.03、2.02、1.86、1.98，决定系数提高了0.06～0.15，均方根误差除褐土土样提高0.02百分点外，其他样本组降低了0.01～0.08百分点，相对分析误差提高了0.17～0.43，模型决定系数和相对分析误差得到显著提升；尤其对于土壤水分含量变异系数较小的3组土样，模型从待改进级别提高到性能良好级别，对土壤有机质含量具有较好的预测准确性。说明该方法用于去除土壤水分因素影响和提高有机质含量高光谱估测精度的有效性。

2013

0.0

PENG

Jie

, LI

, ZHOU

Qing

, et al(彭杰, 李曦, 周清, 等). Journal of Remote Sensing(遥感学报), 2013, 17(6): 1396.

Spectral data of Soil Iron Oxide (SIO) content and Soil Organic Matter (SOM) content in 253 soil samples from Hunan, Zhejiang and Fujian provinces were analyzed to investigate the effect of SIO content on spectral characteristics and quantification by remote sensing inversion of SOM content. Results showed the following: the characteristic wave band of SIO content ranges from 600 nm to 1400 nm; if SIO content 30 g/kg, SI of SIO would cover up that of SOM content. When the ratio of SIO content to SOM content was less than 0.726, SI of SIO content would not affect that of SOM content; when SIO content/SOM content ranged from 1.05 to 2, HI in the range of 400 nm to 1300 nm band of SOM content would be highly influenced, whereas HI in the range of 1300 nm to 2400 nm would be slightly affected; when SIO content/SOM content ratio was more than 2.21, SI of SOM content would be completely covered by that of SIO content. SIO content affected quantificational remote sensing inversion of SOM. The inversion accuracy and stability of prediction model estimating SOM content decreased with the increase in SIO content or the ratio of SIO content to SOM content. However, SIO content slightly affected quantificational inversion of SOM when SIO content was less than 20 g/kg or SIO content/SOM content was less than 2.0.

通过分析湖南、浙江和福建三省不同氧化铁和有机质含量共253个土样的光谱数据，研究了氧化铁对有机质光谱特征及定量反演的影响。结果表明：氧化铁的光谱特征波段为600―1400 nm；氧化铁含量小于20 g/kg时对有机质的光谱信息没有影响，含量在20―30 g/kg时，对有机质可见光波段光谱信息的表现有影响，近红外波段的影响不大，含量大于30 g/kg时，氧化铁会掩盖有机质的光谱信息；氧化铁与有机质的比值小于0.726时对有机质的光谱信息没有影响，比值为1.05―2时，对有机质400―1300 nm波段光谱信息的表现有影响，1300―2400 nm波段的影响不大，比值大于2.21时，氧化铁会完全掩盖有机质的光谱信息；氧化铁对有机质的光谱定量反演有影响，随氧化铁含量的增加或氧化铁含量与有机质含量比值的增大，模型的稳定性与预测能力有所降低，但氧化铁含量小于20 g/kg、氧化铁含量与有机质含量比值小于2.0时，氧化铁的影响不明显。

... 此外, 土壤有机质的光谱特性易受水分、氧化铁等土壤成分的干扰, 将水分、氧化铁等因素纳入到土壤有机质含量监测中可有效提升监测精度^{[20, 27]} ...

2007

0.0

... 为凸显或提取土壤光谱中的有机质信息, 采用数学变换^[21]、小波技术^[22]、主成分分析^[23]等对土壤光谱进行处理分析, 并利用偏最小二乘^[21]、随机森林^[24]、支持向量机^[25]、神经网络^[26]等回归分析算法构建土壤有机质含量估测模型 ...

... 该11种处理方法为: 倒数(1/R)、对数(LogR)、倒数的对数[log(1/R)]、一阶微分(R')、倒数的一阶微分[(1/R)']、对数的一阶微分[(LogR)']、倒数的对数的一阶微分[(log(1/R)']、除以R₉₃₀、除以R_(450~750)、吸收峰深度(H)、弓曲差(Gqc)^[21] ...

2017

0.0

2017

0.0

2011

0.0

2014

0.0

2009

0.0

ZHANG

Juan-juan

, TIAN

Yong-chao

, ZHU

Yan

, et al(张娟娟, 田永超, 朱艳, 等). Scientia Agricultural Sinica(中国农业科学), 2009, 42(9): 3154.

【Objective】 The objectives of the present study were to determine the key spectral parameters and models for estimating SOM content. 【Method】 The dried sample of five different soil types in China were analyzed for SOM content and hyperspectral reflectance within 350-2 500 nm, quantitative models of SOM using spectral index and BP neural network were established, respectively. 【Result】The results showed that correlation between spectral indices which composed of first derivative and SOM content were obviously stronger than those composed of original reflectance, especially derivative with Norris smoothing filter. The correlation sequence of SOM to different index types was DI＞RI＞ND which composed of spectral reflectance or the first derivative spectra. DI composed of first derivative of 554 nm and 1 398 nm gave a better prediction performance, with equation as y=184.2×exp[-1297×DI(D554, D1398)], coefficient of determination was 0.90. Testing of the monitoring models with independent data from different soil types indicated that R2, RMSE and RPD of validation were 0.84, 3.64 and 2.98, respectively. In addition, the scores computed by PLS were applied as input of BP neural network developed with over 99.56% of cumulative proportion of correlation matrix. R2 of calibration model was 0.98, and R2, RMSE, RPD of validation were 0.96, 2.24 and 4.83, respectively. Compared with BP neural network model, DI(D554, D1398) had a little lower prediction precision, but it could meet need of estimating of SOM content. 【Conclusion】 It is concluded that both of methods based on DI(D554, D1398) and BP neural network can estimate SOM content accurately.

(南京农业大学农学院/江苏省信息农业高技术研究重点实验室)

【目的】构建适合土壤有机质含量估测的高光谱参数及定量反演模型。【方法】系统分析中国中、东部地区5种不同类型土壤风干样本有机质含量与350～2 500 nm波段范围高光谱反射率之间的关系,利用特征光谱参数和BP神经网络建立土壤有机质的定量估测模型。【结果】光谱一阶导数构成的两波段光谱参数与土壤有机质含量的相关性明显优于原始光谱,尤其采用Norris平滑滤波后导数光谱效果更好。光谱参数构成形式以差值指数最好,其次为比值和归一化指数。与土壤有机质含量相关程度最高的光谱参数是由可见光区554 nm和近红外区1 398 nm两个波段的一阶导数组合而成的差值指数DI(D554,D1398),两者呈显著指数曲线关系,拟合方程为y= 184.2 ×exp[-1297×DI(D554,D1398)],决定系数为0.90。经不同类型土壤的观测资料检验,模型预测决定系数为0.84,均方根误差RMSE为3.64,相对分析误差RPD为2.98,显示估测模型具有较好的预测精度。另外,利用BP神经网络结合偏最小二乘法(PLS)对导数光谱进行分析,提取贡献率达到99.56 %的前6个主成分建立了三层BP 神经网络模型,模型决定系数为0.98,经不同类型土壤的观测资料检验,模型预测决定系数为0.96,RMSE为2.24,相对偏差RPD为4.83。比较利用DI(D554,D1398)和BP网络进行土壤有机质含量的预测结果,前者精度低于后者,但可以满足土壤有机质监测的需要。【结论】利用差值光谱指数DI(D554,D1398)和BP神经网络模型均可实现对土壤有机质的精确估测。

2015

0.0

Hai-qing

, YAO

Yan-min

, WANG

De-ying

, et al(司海青, 姚艳敏, 王德营, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2015, 31(9): 114.

土壤含水率对有机质（soil organic matter，SOM）含量高光谱估算精度有很大的影响。为了探讨SOM高光谱估算中土壤含水率的影响，该文对烘干土、风干土和质量含水率为5%～40%（按5%递增）的土壤样本进行了室内高光谱测量，对光谱数据进行了反射率、反射率一阶导数和反射率倒数对数3种光谱数据变换，运用偏最小二乘回归法（partial least squares regression，PLSR）建立了相应的SOM估算模型。结果表明，风干土的SOM高光谱估算精度较好；当含水率水平小于25%时，SOM估算模型精度受含水率的影响较大，光谱数据进行反射率倒数对数变换后的模型精度最高；当含水率水平大于等于25%时，水分对土壤光谱反射率的影响要大于SOM，不适宜利用土壤光谱数据进行SOM含量高光谱估算。该研究可为大田环境不同含水率情况下光谱估算SOM提供参考。

... 土壤水分对土壤光谱具有较强的影响, 其对土壤有机质含量监测精度的提升具有较强的制约作用^[27], 研究分析土壤水分对土壤有机质含量估测的影响具有一定意义, 但由于本研究首先将土壤进行风干, 然后在实验室内测定其光谱及土壤参数, 故未考虑水分的影响 ...