WO2024099475A1 - Soil heavy metal stress identification method based on long time series ndvi - Google Patents

Soil heavy metal stress identification method based on long time series ndvi Download PDF

Info

Publication number
WO2024099475A1
WO2024099475A1 PCT/CN2024/070257 CN2024070257W WO2024099475A1 WO 2024099475 A1 WO2024099475 A1 WO 2024099475A1 CN 2024070257 W CN2024070257 W CN 2024070257W WO 2024099475 A1 WO2024099475 A1 WO 2024099475A1
Authority
WO
WIPO (PCT)
Prior art keywords
heavy metal
ndvi
metal stress
long time
time series
Prior art date
Application number
PCT/CN2024/070257
Other languages
French (fr)
Chinese (zh)
Inventor
李旭青
卢成乾
唐瑞尹
王延仓
何跃君
武建军
吴伶
吴艳萍
房红记
王棋
吴琼
孙肖
Original Assignee
北华航天工业学院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北华航天工业学院 filed Critical 北华航天工业学院
Publication of WO2024099475A1 publication Critical patent/WO2024099475A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/143—Sensing or illuminating at different wavelengths
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/62—Extraction of image or video features relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/188—Vegetation

Definitions

  • the invention relates to a soil heavy metal stress identification method, and relates to a soil heavy metal stress identification method based on long-time series NDVI, and belongs to the field of remote sensing technology.
  • the traditional heavy metal monitoring method is to conduct laboratory tests on samples collected on the spot.
  • the results are highly accurate, but slow, small in scope, and expensive, and cannot obtain continuous information on spatial distribution.
  • remote sensing technology has the characteristics of large-scale synchronous monitoring, fast transmission speed, and rich information, which provides a reference for heavy metal pollution monitoring.
  • the use of remote sensing satellite image data to monitor heavy metal pollution in farmland on a large scale has gradually become an important technical means.
  • domestic and foreign researchers have used remote sensing technology to study the status of heavy metal pollution using crop reflectance spectral response characteristics and spectral information analysis.
  • the optical remote sensing method has the advantage of high spatial resolution.
  • the optical vegetation index data is highly correlated with soil heavy metals and is the most commonly used data for soil heavy metal monitoring.
  • the Normalized Difference Vegetation Index (NDVI) is defined as the reflectivity difference between the near-infrared band and the red band divided by the reflectivity sum of the near-infrared band and the red band. It is one of the important parameters reflecting the growth of crops.
  • the complexity of the crop growth environment in natural farmland ecosystems causes them to be affected by multiple sources of compound stress factors such as soil moisture stress, pest and disease stress, and heavy metal stress during their growth.
  • compound stress factors such as soil moisture stress, pest and disease stress, and heavy metal stress during their growth.
  • EMD Empirical Mode Decomposition
  • EMD Empirical Mode Decomposition
  • the main limitations of current remote sensing monitoring of crop heavy metal stress are the lack of information in the remote sensing indicators used, the lack of stability of the indicators, and the lack of systematic analysis of the temporal characteristics of the spectral response parameters of crop heavy metal stress. If the stress spectral characteristic parameters are combined with the temporal characteristics of the crop growth stage, it will be possible to effectively distinguish heavy metal stress from other stress factors. Therefore, "time-spectrum" analysis is an important method for extracting heavy metal stress characteristics.
  • the purpose of the present invention is to provide a method for identifying soil heavy metal stress based on long-term NDVI.
  • the technical solution adopted by the present invention is:
  • a soil heavy metal stress identification method based on long-term NDVI includes the following steps:
  • Step 1 Remote sensing image preprocessing: Perform radiometric calibration, atmospheric correction and geometric correction preprocessing on N remote sensing image data, N>1;
  • Step 2 Construction of NDVI long time series: Calculate the NDVI long time series x(n) based on the preprocessed remote sensing image data, 1 ⁇ n ⁇ N:
  • ⁇ NIR (n) is the reflectance of the near-infrared band of the n-th remote sensing image
  • ⁇ RED (n) is the reflectance of the red band of the n-th remote sensing image
  • Step 3 EEMD decomposition: Decompose the NDVI long time series x(n) based on the EEMD algorithm.
  • the decomposition process is as follows:
  • M is the number of times the standard normal distribution white noise signal is added
  • c i,j (n) is the jth IMF component of the noisy NDVI long time series x i (n)
  • ri,j (n) is the EMD residual function of the noisy NDVI long time series
  • J is the number of IMF components
  • Step 4 Calculate statistical descriptive indicators: Calculate the number of extreme points K j and fluctuation period P j of the first to Jth IMF components of the EEMD decomposition of the NDVI long time series x(n):
  • Step 6 Identification of soil heavy metal stress series: According to the preset identification conditions, the IMF components of the EEMD decomposition of the NDVI long time series x(n) that meets the long-term soil heavy metal stress characteristics are selected and synthesized into the soil heavy metal stress series Yd (n);
  • Step 7 Extraction of soil heavy metal stress stability features: Calculate the heavy metal stress stability features of the soil heavy metal stress sequence Yd :
  • Y dt (n) is the value of the previous phase of the soil heavy metal stress sequence Y d (n)
  • Y dt+1 (n) is the value of the next phase of the soil heavy metal stress sequence Y d (n)
  • ⁇ DAY is the date interval between the previous phase and the next phase
  • Step 8 Ground data measurement: Use a portable XRF analyzer to measure the content of preset heavy metal elements in the soil;
  • Step 9 Construct a heavy metal prediction model: fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of the preset types of soil as a heavy metal prediction model;
  • Step 10 Monitor the degree of heavy metal stress in the soil: collect remote sensing images, perform the processing of steps 1 to 7, obtain the corresponding heavy metal stress stability characteristics, and use the heavy metal prediction model to predict the degree of heavy metal stress in the soil.
  • the preset screening condition of step 6 is that the fluctuation period Pr of the IMF component of the NDVI long time series is greater than T, where T is the common fluctuation period of the IMF component of the NDVI long time series and the IMF component of the seasonal average series of its corresponding NDVI long time series.
  • the preset identification condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 6.
  • the preset screening condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 12.
  • step 9 a quadratic curve is used to fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of a preset type in the soil.
  • the present invention utilizes long time series to avoid the drawback of a single phase lacking time characteristics, combines stress spectral characteristics with time characteristics, and combines crop phenology to establish stable characteristic indicators that effectively identify heavy metal stress in a complex soil environment, and the relationship between crop growth stages and heavy metal stress, thereby achieving the identification and monitoring of soil heavy metals;
  • the ensemble empirical mode decomposition algorithm of the present invention effectively eliminates the error during decomposition, and further improves the accuracy of the decomposition result;
  • the present invention performs analysis based on the time scale characteristics of winter wheat growth, and has stronger adaptability than existing heavy metal analysis methods, and has obvious advantages in processing non-stationary and nonlinear data.
  • Fig. 1 is a flow chart of the present invention
  • FIG. 2 is a long time series diagram of NDVI in Example 1 of the present invention.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • This embodiment constructs a long-term NDVI series of winter wheat based on multispectral remote sensing images, and performs EEMD decomposition on the constructed long-term series to obtain various IMF components; constructs a seasonal average model, combines the statistical descriptive indicators of each IMF component, identifies and extracts soil heavy metal stress subsequences, and then synthesizes the identified soil heavy metal stress subsequences to obtain a heavy metal stress sequence; calculates the first-order derivative of the soil heavy metal stress sequence to obtain the stable characteristics of soil heavy metal stress, and constructs an inversion model by combining the soil heavy metal stress data with the ground measured data to achieve large-scale soil heavy metal monitoring.
  • a soil heavy metal stress identification method based on long-term NDVI includes the following steps:
  • Step 1 Remote sensing image preprocessing: perform radiometric calibration and atmospheric correction on N remote sensing image data and geometric correction preprocessing, N>1;
  • Step 2 Construction of NDVI long time series: Calculate the NDVI long time series x(n) based on the preprocessed remote sensing image data, 1 ⁇ n ⁇ N:
  • ⁇ NIR (n) is the reflectance of the near infrared band of the n-th remote sensing image
  • ⁇ RED (n) is the reflectance of the red band of the n-th remote sensing image
  • Step 3 EEMD decomposition: Decompose the NDVI long time series x(n) based on the EEMD algorithm.
  • the decomposition process is as follows:
  • M is the number of times the standard normal distribution white noise signal is added
  • c i,j (n) is the jth IMF component of the noisy NDVI long time series x i (n)
  • ri,j (n) is the EMD residual function of the noisy NDVI long time series
  • J is the number of IMF components
  • Step 4 Calculate statistical descriptive indicators: Calculate the number of extreme points K j and fluctuation period P j of the first to Jth IMF components of the EEMD decomposition of the NDVI long time series x(n):
  • Step 6 Identification of soil heavy metal stress series: According to the preset identification conditions, the IMF components of the EEMD decomposition of the NDVI long time series x(n) that meets the long-term soil heavy metal stress characteristics are selected and synthesized into the soil heavy metal stress series Yd (n);
  • Step 7 Extraction of soil heavy metal stress stability features: Calculate the heavy metal stress stability features of the soil heavy metal stress sequence Yd :
  • Y dt (n) is the value of the previous phase of the soil heavy metal stress sequence Y d (n)
  • Y dt+1 (n) is the value of the next phase of the soil heavy metal stress sequence Y d (n)
  • ⁇ DAY is the date interval between the previous phase and the next phase
  • Step 8 Ground data measurement: Use a portable XRF analyzer to measure the content of preset heavy metal elements in the soil;
  • Step 9 Construct a heavy metal prediction model: fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of the preset types of soil as a heavy metal prediction model;
  • Step 10 Monitor the degree of heavy metal stress in the soil: collect remote sensing images, perform the processing of steps 1 to 7, obtain the corresponding heavy metal stress stability characteristics, and use the heavy metal prediction model to predict the degree of heavy metal stress in the soil.
  • the preset identification condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 12.
  • step 9 a quadratic curve is used to fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of the preset type of soil.
  • the fluctuation period of the IMF component of the seasonal average series is identified to be approximately equal to 12, which is consistent with the fluctuation period of the third IMF component of the EEMD decomposition of the NDVI long time series x(n);
  • the condition for identifying the soil heavy metal stress series is that the fluctuation period is greater than 12, and the fourth and fifth IMF components of the EEMD decomposition of the NDVI long time series x(n) are accumulated to form the soil heavy metal stress series;
  • the quadratic curve is used to construct the relationship between the stable characteristic Y df of soil heavy metal stress and the measured soil heavy metal element content
  • the functional relationship of is the optimal function model.
  • the vegetation index NDVI is selected as the winter wheat growth extraction parameter to construct a long NDVI series.
  • EEMD decomposition multiple groups of white noise are added to the original time series signal of the ensemble empirical mode decomposition, which effectively solves the problem of modal aliasing and achieves better decomposition results.
  • IMF components
  • Each IMF represents various oscillation models of the original data, which usually have different amplitudes and spectra that change with time.
  • a seasonal average model is constructed to obtain the spectral characteristics of the crop growth cycle, and the statistical descriptive indicators of each IMF component are combined to effectively identify the subsequences that meet the soil heavy metal stress.
  • Soil heavy metal stress is a long-term stress with a long duration and a long fluctuation period.
  • the fluctuation period of the third component IMF of the NDVI long-term series is consistent with the period of the IMF component of the seasonal average series of the NDVI long-term series representing the growth stage of winter wheat, which is 12 months, that is, the growth period of winter wheat.
  • the fluctuation periods of the fourth and fifth IMF components IMF4 and IMF5 of the seasonal average series of the NDVI long-term series are significantly greater than the period of the IMF component of the seasonal average series of the NDVI long-term series. It is determined that the fourth to fifth IMF components of the seasonal average series of the NDVI long-term series are soil heavy metal stress subseries.
  • the first-order derivative method is used to analyze the soil heavy metal stress series Yd to obtain the stable characteristics of heavy metal stress in the soil.
  • Example 2 This example constructs a long NDVI series for rice that is harvested twice a year based on multispectral remote sensing images. The difference from Example 1 is that the preset screening condition of step 6 is that the fluctuation period of the IMF component P r of the NDVI long time series is greater than 6.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Astronomy & Astrophysics (AREA)
  • Remote Sensing (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Investigation Of Foundation Soil And Reinforcement Of Foundation Soil By Compacting Or Drainage (AREA)

Abstract

A soil heavy metal stress identification method based on a long time series NDVI, which comprises the steps of remotely sensed image preprocessing, NDVI long time series construction, EEMD decomposition, statistical descriptive metric calculation, seasonal average model construction, soil heavy metal stress series identification, soil heavy metal stress stable feature extraction, actual measurement of ground data, and soil heavy metal stress level monitoring. The present method utilizes an NDVI long time series, and stress spectrum characteristics and time characteristics are combined, mitigating the defect of a lack of time characteristics in a single time phase; comprehensive EEMD decomposition and seasonal average model construction are integrated, and the problem of noise residue is solved, better facilitating the capture of the long-terms effects of soil heavy metal stress, and implementing identification and extraction of soil heavy metal stress; crop phenology is intregrated, stable characteristic metrics for effective identification of soil heavy metal stress is established in complex soil environments, and, by combining actual measurement data, large-area soil heavy metal stress monitoring is achieved.

Description

基于长时序NDVI的土壤重金属胁迫甄别方法A method for identifying soil heavy metal stress based on long-term NDVI 技术领域Technical Field
本发明涉及一种土壤重金属胁迫甄别方法，涉及一种基于长时序NDVI的土壤重金属胁迫甄别方法，属于遥感技术领域。The invention relates to a soil heavy metal stress identification method, and relates to a soil heavy metal stress identification method based on long-time series NDVI, and belongs to the field of remote sensing technology.
背景技术Background technique
传统的重金属监测方法是将实地采集的样本进行实验室化验，其结果精度高，但是速度慢、范围小、成本高，不能在空间分布上获取连续的信息。相较于传统方法，遥感技术具有大面积同步监测、传输速度快、信息丰富等特点，为其在重金属污染监测方面提供了借鉴。The traditional heavy metal monitoring method is to conduct laboratory tests on samples collected on the spot. The results are highly accurate, but slow, small in scope, and expensive, and cannot obtain continuous information on spatial distribution. Compared with traditional methods, remote sensing technology has the characteristics of large-scale synchronous monitoring, fast transmission speed, and rich information, which provides a reference for heavy metal pollution monitoring.
利用遥感卫星影像数据对农田重金属污染大范围监测，逐渐成为重要的技术手段。近年来，国内外学者借助遥感技术，利用作物反射光谱响应特征以及光谱信息分析方面研究重金属污染状况。其中，光学遥感方法具备高空间分辨率的优势，光学植被指数数据与土壤重金属高度相关，是进行土壤重金属监测最常用的一种数据。归一化植被指数(Normalized Difference Vegetation Index,NDVI)定义为近红波段与红光波段的反射率差值除以近红外波段与红光波段的反射率和值，是反映农作物长势的重要参数之一。The use of remote sensing satellite image data to monitor heavy metal pollution in farmland on a large scale has gradually become an important technical means. In recent years, domestic and foreign scholars have used remote sensing technology to study the status of heavy metal pollution using crop reflectance spectral response characteristics and spectral information analysis. Among them, the optical remote sensing method has the advantage of high spatial resolution. The optical vegetation index data is highly correlated with soil heavy metals and is the most commonly used data for soil heavy metal monitoring. The Normalized Difference Vegetation Index (NDVI) is defined as the reflectivity difference between the near-infrared band and the red band divided by the reflectivity sum of the near-infrared band and the red band. It is one of the important parameters reflecting the growth of crops.
自然农田生态系统中农作物生长环境的复杂性导致其在生长过程中受土壤水分胁迫、病虫害胁迫、重金属胁迫等多源复合胁迫因素的影响。在实现土壤重金属胁迫甄别时，需要对特定特征信号进行处理，从而提取出表征特定特征的分量。集合经验模态分解(Ensemble Empirical Mode Decomposition，EEMD)算法在经验模态分解(Empirical Mode Decomposition，EMD)算法的基础上进行了改进，能将原始长时间序列数据进行不断分解，获取不同本征模态函数(Intrinsic Mode Function，IMF)分量，最终分解为若干单一频率的序列以及一个残差组分的形式，为土壤重金属胁迫甄别提供技术支撑。The complexity of the crop growth environment in natural farmland ecosystems causes them to be affected by multiple sources of compound stress factors such as soil moisture stress, pest and disease stress, and heavy metal stress during their growth. When realizing the identification of soil heavy metal stress, it is necessary to process specific characteristic signals to extract components that characterize specific characteristics. The Ensemble Empirical Mode Decomposition (EEMD) algorithm is improved on the basis of the Empirical Mode Decomposition (EMD) algorithm. It can continuously decompose the original long-time series data to obtain different intrinsic mode function (IMF) components, and finally decompose them into several single-frequency sequences and a residual component, providing technical support for the identification of soil heavy metal stress.
当前遥感监测作物重金属胁迫的主要局限在于采用的遥感指标的信息量匮乏、指标的稳定性不足，并且缺乏对农作物重金属胁迫光谱响应参数时间特征的系统分析。如果将胁迫光谱特征参数与作物生长阶段的时间特征相结合，将能够有效区分重金属胁迫和其他胁迫因子。因此，基于“时-谱”分析是提取重金属胁迫特征的重要方法。The main limitations of current remote sensing monitoring of crop heavy metal stress are the lack of information in the remote sensing indicators used, the lack of stability of the indicators, and the lack of systematic analysis of the temporal characteristics of the spectral response parameters of crop heavy metal stress. If the stress spectral characteristic parameters are combined with the temporal characteristics of the crop growth stage, it will be possible to effectively distinguish heavy metal stress from other stress factors. Therefore, "time-spectrum" analysis is an important method for extracting heavy metal stress characteristics.
发明内容 Summary of the invention
本发明的目的是提供一种基于长时序NDVI的土壤重金属胁迫甄别方法。The purpose of the present invention is to provide a method for identifying soil heavy metal stress based on long-term NDVI.
为解决上述技术问题，本发明采用的技术方案是：In order to solve the above technical problems, the technical solution adopted by the present invention is:
一种基于长时序NDVI的土壤重金属胁迫甄别方法，包括以下步骤：A soil heavy metal stress identification method based on long-term NDVI includes the following steps:
步骤1：遥感影像预处理：对N幅遥感影像数据进行辐射定标、大气校正以及几何校正预处理，N＞1；Step 1: Remote sensing image preprocessing: Perform radiometric calibration, atmospheric correction and geometric correction preprocessing on N remote sensing image data, N>1;
步骤2：NDVI长时间序列构建：基于预处理后的遥感影像数据计算NDVI长时间序列x(n),1≤n≤N：
Step 2: Construction of NDVI long time series: Calculate the NDVI long time series x(n) based on the preprocessed remote sensing image data, 1≤n≤N:
其中ρNIR(n)为第n幅遥感影像近红外波段的反射率，ρRED(n)为第n幅遥感影像红光波段的反射率；Where ρ NIR (n) is the reflectance of the near-infrared band of the n-th remote sensing image, and ρ RED (n) is the reflectance of the red band of the n-th remote sensing image;
步骤3：EEMD分解：基于EEMD算法对NDVI长时间序列x(n)进行分解，其分解流程如下：Step 3: EEMD decomposition: Decompose the NDVI long time series x(n) based on the EEMD algorithm. The decomposition process is as follows:
将具有标准正态分布的白噪声信号zi(n)加入到NDVI长时间序列x(n)上,产生加噪NDVI长时间序列xi(n)：
xi(n)＝x(n)+zi(n)，1≤i≤MAdd the white noise signal z i (n) with standard normal distribution to the NDVI long time series x(n) to generate the noisy NDVI long time series x i (n):
x i (n) = x (n) + z i (n), 1 ≤ i ≤ M
式中，M为加入标准正态分布的白噪声信号的次数；Where M is the number of times the standard normal distribution white noise signal is added;
对加噪NDVI长时间序列xi(n)分别进行EMD分解，得到各自IMF和的形式：
Perform EMD decomposition on the noisy NDVI long time series x i (n) to obtain the form of their respective IMF sums:
式中，ci,j(n)为加噪NDVI长时间序列xi(n)的第j个IMF分量，ri,j(n)是加噪NDVI长时间序列的EMD残余函数，J是IMF分量的数量；Where c i,j (n) is the jth IMF component of the noisy NDVI long time series x i (n), ri,j (n) is the EMD residual function of the noisy NDVI long time series, and J is the number of IMF components;
计算NDVI长时间序列x(n)的EEMD分解的第j个IMF分量：
Calculate the jth IMF component of the EEMD decomposition of the NDVI long time series x(n):
步骤4：计算统计性描述指标：计算NDVI长时间序列x(n)的EEMD分解的第一个至第J个IMF分量的极值点个数Kj、波动周期Pj：
Step 4: Calculate statistical descriptive indicators: Calculate the number of extreme points K j and fluctuation period P j of the first to Jth IMF components of the EEMD decomposition of the NDVI long time series x(n):
步骤5：季节平均模型构建：对NDVI长时间序列各年同月数据求取平均值，得到NDVI长时间序列的季节平均序列；对NDVI长时间序列的季节平均序列使用所述步骤3进行EEMD分解，得到NDVI长时间序列的季节平均序列的IMF分量imfr，计算NDVI长时间序列的季节平均序列的IMF分量imfr的周期，r＝1,2,…R；Step 5: seasonal average model construction: average the data of the same month of each year of the NDVI long-term series to obtain the seasonal average series of the NDVI long-term series; perform EEMD decomposition on the seasonal average series of the NDVI long-term series using the step 3 to obtain the IMF component imfr of the seasonal average series of the NDVI long-term series, and calculate the period of the IMF component imfr of the seasonal average series of the NDVI long-term series, r = 1, 2, ... R;
步骤6：土壤重金属胁迫序列甄别：按照预设甄别条件选择符合长周期土壤重金属胁迫特征的NDVI长时间序列x(n)的EEMD分解的各IMF分量累加合成为土壤重金属胁迫序列Yd(n)；Step 6: Identification of soil heavy metal stress series: According to the preset identification conditions, the IMF components of the EEMD decomposition of the NDVI long time series x(n) that meets the long-term soil heavy metal stress characteristics are selected and synthesized into the soil heavy metal stress series Yd (n);
步骤7：土壤重金属胁迫稳定特征提取：计算土壤重金属胁迫序列Yd的重金属胁迫稳定特征：
Step 7: Extraction of soil heavy metal stress stability features: Calculate the heavy metal stress stability features of the soil heavy metal stress sequence Yd :
式中：Ydt(n)为土壤重金属胁迫序列Yd(n)的前一时相的数值，Ydt+1(n)为土壤重金属胁迫序列Yd(n)的后一时相的数值，ΔDAY为前一时相对于后一时相对的日期间隔；Where: Y dt (n) is the value of the previous phase of the soil heavy metal stress sequence Y d (n), Y dt+1 (n) is the value of the next phase of the soil heavy metal stress sequence Y d (n), ΔDAY is the date interval between the previous phase and the next phase;
步骤8：地面数据实测：使用便携式XRF分析仪对土壤中的预设种类的重金属元素含量测定；Step 8: Ground data measurement: Use a portable XRF analyzer to measure the content of preset heavy metal elements in the soil;
步骤9：构建重金属预测模型：拟合重金属胁迫稳定特征Ydf(n)与土壤的预设种类的重金属元素含量间的关系，作为重金属预测模型；Step 9: Construct a heavy metal prediction model: fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of the preset types of soil as a heavy metal prediction model;
步骤10：监测土壤重金属胁迫程度：采集遥感影像，进行所述步骤1-步骤7的处理，得到对应的重金属胁迫稳定特征，使用所述重金属预测模型预测土壤重金属胁迫程度。Step 10: Monitor the degree of heavy metal stress in the soil: collect remote sensing images, perform the processing of steps 1 to 7, obtain the corresponding heavy metal stress stability characteristics, and use the heavy metal prediction model to predict the degree of heavy metal stress in the soil.
进一步，所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量的波动周期Pr为大于T，T为NDVI长时间序列的IMF分量与其对应的NDVI长时间序列的季节平均序列的IMF分量的共同波动周期。 Furthermore, the preset screening condition of step 6 is that the fluctuation period Pr of the IMF component of the NDVI long time series is greater than T, where T is the common fluctuation period of the IMF component of the NDVI long time series and the IMF component of the seasonal average series of its corresponding NDVI long time series.
进一步，所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量的波动周期Pr为大于6。Furthermore, the preset identification condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 6.
进一步，所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量的波动周期Pr大于12。Furthermore, the preset screening condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 12.
进一步，所述步骤9中使用二次方曲线拟合重金属胁迫稳定特征Ydf(n)与土壤的预设种类的重金属元素含量间的关系。Furthermore, in step 9, a quadratic curve is used to fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of a preset type in the soil.
采用上述技术方案，本发明的有益效果是：By adopting the above technical solution, the beneficial effects of the present invention are:
1.本发明利用长时间序列，避免了单一时相缺乏时间特征的弊端，将胁迫光谱特征与时间特征相结合，结合作物物候期在复杂的土壤环境中建立有效甄别重金属胁迫的稳定特征指标、作物生长阶段与受重金属胁迫之间的关系，实现了土壤重金属的甄别与监测；1. The present invention utilizes long time series to avoid the drawback of a single phase lacking time characteristics, combines stress spectral characteristics with time characteristics, and combines crop phenology to establish stable characteristic indicators that effectively identify heavy metal stress in a complex soil environment, and the relationship between crop growth stages and heavy metal stress, thereby achieving the identification and monitoring of soil heavy metals;
2.本发明集合经验模态分解算法对比其前身经验模态分解，有效地消除了分解时的误差，使分解结果的精确的进一步提高；2. Compared with its predecessor, the empirical mode decomposition, the ensemble empirical mode decomposition algorithm of the present invention effectively eliminates the error during decomposition, and further improves the accuracy of the decomposition result;
3.本发明依据冬小麦长势自身的时间尺度特征进行分析，对比已有的重金属分析方法具有更强的适应性，在处理非平稳、非线性数据上具有明显的优势。3. The present invention performs analysis based on the time scale characteristics of winter wheat growth, and has stronger adaptability than existing heavy metal analysis methods, and has obvious advantages in processing non-stationary and nonlinear data.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
下面结合附图和具体实施方式对本发明作进一步详细的说明。The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.
图1是本发明的流程图；Fig. 1 is a flow chart of the present invention;
图2是本发明的实施例1中的NDVI长时间序列图。FIG. 2 is a long time series diagram of NDVI in Example 1 of the present invention.
具体实施方式Detailed ways
实施例1：Embodiment 1:
本实施例基于多光谱遥感影像对冬小麦进行NDVI长时间序列构建，对所构建的长时间序列进行EEMD分解得到各IMF分量；构建季节平均模型，结合各IMF分量的统计性描述指标，甄别提取出土壤重金属胁迫子序列，然后将甄别出的土壤重金属胁迫子序列进行合成获得重金属胁迫序列；对土壤重金属胁迫序列进行一阶导数计算，获得土壤重金属胁迫稳定特征，结合土壤重金属胁迫数据与地面实测数据构建反演模型，实现大范围土壤重金属的监测。This embodiment constructs a long-term NDVI series of winter wheat based on multispectral remote sensing images, and performs EEMD decomposition on the constructed long-term series to obtain various IMF components; constructs a seasonal average model, combines the statistical descriptive indicators of each IMF component, identifies and extracts soil heavy metal stress subsequences, and then synthesizes the identified soil heavy metal stress subsequences to obtain a heavy metal stress sequence; calculates the first-order derivative of the soil heavy metal stress sequence to obtain the stable characteristics of soil heavy metal stress, and constructs an inversion model by combining the soil heavy metal stress data with the ground measured data to achieve large-scale soil heavy metal monitoring.
一种基于长时序NDVI的土壤重金属胁迫甄别方法，包括以下步骤：A soil heavy metal stress identification method based on long-term NDVI includes the following steps:
步骤1：遥感影像预处理：对N幅遥感影像数据进行辐射定标、大气校正 以及几何校正预处理，N＞1；Step 1: Remote sensing image preprocessing: perform radiometric calibration and atmospheric correction on N remote sensing image data and geometric correction preprocessing, N>1;
步骤2：NDVI长时间序列构建：基于预处理后的遥感影像数据计算NDVI长时间序列x(n),1≤n≤N：
Step 2: Construction of NDVI long time series: Calculate the NDVI long time series x(n) based on the preprocessed remote sensing image data, 1≤n≤N:
其中ρNIR(n)为第n幅遥感影像近红外波段的反射率，ρRED(n)为第n幅遥感影像红光波段的反射率；Where ρ NIR (n) is the reflectance of the near infrared band of the n-th remote sensing image, and ρ RED (n) is the reflectance of the red band of the n-th remote sensing image;
步骤3：EEMD分解：基于EEMD算法对NDVI长时间序列x(n)进行分解，其分解流程如下：Step 3: EEMD decomposition: Decompose the NDVI long time series x(n) based on the EEMD algorithm. The decomposition process is as follows:
将具有标准正态分布的白噪声信号zi(n)加入到NDVI长时间序列x(n)上,产生加噪NDVI长时间序列xi(n)：
xi(n)＝x(n)+zi(n)，1≤i≤MAdd the white noise signal z i (n) with standard normal distribution to the NDVI long time series x(n) to generate the noisy NDVI long time series x i (n):
x i (n) = x (n) + z i (n), 1 ≤ i ≤ M
式中，M为加入标准正态分布的白噪声信号的次数；Where M is the number of times the standard normal distribution white noise signal is added;
对加噪NDVI长时间序列xi(n)分别进行EMD分解，得到各自IMF和的形式：
Perform EMD decomposition on the noisy NDVI long time series x i (n) to obtain the form of their respective IMF sums:
式中，ci,j(n)为加噪NDVI长时间序列xi(n)的第j个IMF分量，ri,j(n)是加噪NDVI长时间序列的EMD残余函数，J是IMF分量的数量；Where c i,j (n) is the jth IMF component of the noisy NDVI long time series x i (n), ri,j (n) is the EMD residual function of the noisy NDVI long time series, and J is the number of IMF components;
计算NDVI长时间序列x(n)的EEMD分解的第j个IMF分量：
Calculate the jth IMF component of the EEMD decomposition of the NDVI long time series x(n):
步骤4：计算统计性描述指标：计算NDVI长时间序列x(n)的EEMD分解的第一个至第J个IMF分量的极值点个数Kj、波动周期Pj：
Step 4: Calculate statistical descriptive indicators: Calculate the number of extreme points K j and fluctuation period P j of the first to Jth IMF components of the EEMD decomposition of the NDVI long time series x(n):
步骤5：季节平均模型构建：对NDVI长时间序列每年同月数据求取平均值；对NDVI长时间序列的季节平均序列使用所述步骤3进行EEMD分解，得到NDVI长时间序列的季节平均序列的IMF分量imfr，计算NDVI长时间序列的季节平均序列的IMF分量Pr的周期，r＝1,2,…R；Step 5: Construction of seasonal average model: Calculate the average value of the data of the same month of the NDVI long-term series every year; Perform EEMD decomposition on the seasonal average series of the NDVI long-term series using the above step 3 to obtain the IMF component imf r of the seasonal average series of the NDVI long-term series, and calculate the period of the IMF component P r of the seasonal average series of the NDVI long-term series, r = 1, 2, ... R;
步骤6：土壤重金属胁迫序列甄别：按照预设甄别条件选择符合长周期土壤重金属胁迫特征的NDVI长时间序列x(n)的EEMD分解的各IMF分量累加合成为土壤重金属胁迫序列Yd(n)；Step 6: Identification of soil heavy metal stress series: According to the preset identification conditions, the IMF components of the EEMD decomposition of the NDVI long time series x(n) that meets the long-term soil heavy metal stress characteristics are selected and synthesized into the soil heavy metal stress series Yd (n);
步骤7：土壤重金属胁迫稳定特征提取：计算土壤重金属胁迫序列Yd的重金属胁迫稳定特征：
Step 7: Extraction of soil heavy metal stress stability features: Calculate the heavy metal stress stability features of the soil heavy metal stress sequence Yd :
式中：Ydt(n)为土壤重金属胁迫序列Yd(n)的前一时相的数值，Ydt+1(n)为土壤重金属胁迫序列Yd(n)的后一时相的数值，ΔDAY为前一时相对于后一时相对的日期间隔；Where: Y dt (n) is the value of the previous phase of the soil heavy metal stress sequence Y d (n), Y dt+1 (n) is the value of the next phase of the soil heavy metal stress sequence Y d (n), ΔDAY is the date interval between the previous phase and the next phase;
步骤8：地面数据实测：使用便携式XRF分析仪对土壤中的预设种类的重金属元素含量测定；Step 8: Ground data measurement: Use a portable XRF analyzer to measure the content of preset heavy metal elements in the soil;
步骤9：构建重金属预测模型：拟合重金属胁迫稳定特征Ydf(n)与土壤的预设种类的重金属元素含量间的关系，作为重金属预测模型；Step 9: Construct a heavy metal prediction model: fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of the preset types of soil as a heavy metal prediction model;
步骤10：监测土壤重金属胁迫程度：采集遥感影像，进行所述步骤1-步骤7的处理，得到对应的重金属胁迫稳定特征，使用所述重金属预测模型预测土壤重金属胁迫程度。Step 10: Monitor the degree of heavy metal stress in the soil: collect remote sensing images, perform the processing of steps 1 to 7, obtain the corresponding heavy metal stress stability characteristics, and use the heavy metal prediction model to predict the degree of heavy metal stress in the soil.
所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量的波动周期Pr大于12。The preset identification condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 12.
所述步骤9中使用二次方曲线拟合重金属胁迫稳定特征Ydf(n)与土壤的预设种类的重金属元素含量间的关系。In step 9, a quadratic curve is used to fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of the preset type of soil.
在本实施例中，甄别出季节平均序列的IMF分量的波动周期约等于12，与它一致的为NDVI长时间序列x(n)的EEMD分解的第三IMF分量的波动周期一致；甄别出土壤重金属胁迫序列的条件为波动周期大于12，NDVI长时间序列x(n)的EEMD分解的第四和第五IMF分量累加合成为土壤重金属胁迫序列；使用二次方曲线构建土壤重金属胁迫稳定特征Ydf与实测土壤重金属元素含量间 的的函数关系为最优函数模型。In this embodiment, the fluctuation period of the IMF component of the seasonal average series is identified to be approximately equal to 12, which is consistent with the fluctuation period of the third IMF component of the EEMD decomposition of the NDVI long time series x(n); the condition for identifying the soil heavy metal stress series is that the fluctuation period is greater than 12, and the fourth and fifth IMF components of the EEMD decomposition of the NDVI long time series x(n) are accumulated to form the soil heavy metal stress series; the quadratic curve is used to construct the relationship between the stable characteristic Y df of soil heavy metal stress and the measured soil heavy metal element content The functional relationship of is the optimal function model.
本实施例选取植被指数NDVI作为冬小麦长势提取参数，构建NDVI长时间序列。使用EEMD分解，在集合经验模态分解的原始时间序列信号中加入多组白噪声，有效解决了模态混叠问题，达到更好的分解结果。EEMD分解后得到若干分量IMF，各个IMF代表了原始数据的各种振荡模型，它们通常带有随时刻而改变的不同振幅和频谱。构建季节平均模型，获得作物生长周期的光谱特征，结合各IMF分量的统计性描述指标，有效甄别出符合土壤重金属胁迫的子序列。土壤重金属胁迫属于长期胁迫，胁迫持续时间长，波动周期长。NDVI长时间序列的第三分量IMF的波动周期与代表冬小麦生长阶段的NDVI长时间序列的季节平均序列的IMF分量的周期一致为12个月，即冬小麦生长周期，NDVI长时间序列的季节平均序列的第四和第五IMF分量IMF4与IMF5的波动周期均明显大于NDVI长时间序列的季节平均序列的IMF分量的周期，判定NDVI长时间序列的季节平均序列的第四至和第五IMF分量为土壤重金属胁迫子系列。将它们合成为表征冬小麦生长周期内受到的土壤重金属胁迫的影响。为了进一步消除其他高频信号的干扰，采用一阶导数方法对土壤重金属胁迫序列Yd进行分析，获取土壤中的重金属胁迫稳定特征。In this embodiment, the vegetation index NDVI is selected as the winter wheat growth extraction parameter to construct a long NDVI series. Using EEMD decomposition, multiple groups of white noise are added to the original time series signal of the ensemble empirical mode decomposition, which effectively solves the problem of modal aliasing and achieves better decomposition results. After EEMD decomposition, several components IMF are obtained. Each IMF represents various oscillation models of the original data, which usually have different amplitudes and spectra that change with time. A seasonal average model is constructed to obtain the spectral characteristics of the crop growth cycle, and the statistical descriptive indicators of each IMF component are combined to effectively identify the subsequences that meet the soil heavy metal stress. Soil heavy metal stress is a long-term stress with a long duration and a long fluctuation period. The fluctuation period of the third component IMF of the NDVI long-term series is consistent with the period of the IMF component of the seasonal average series of the NDVI long-term series representing the growth stage of winter wheat, which is 12 months, that is, the growth period of winter wheat. The fluctuation periods of the fourth and fifth IMF components IMF4 and IMF5 of the seasonal average series of the NDVI long-term series are significantly greater than the period of the IMF component of the seasonal average series of the NDVI long-term series. It is determined that the fourth to fifth IMF components of the seasonal average series of the NDVI long-term series are soil heavy metal stress subseries. They are synthesized to characterize the influence of soil heavy metal stress during the growth period of winter wheat. In order to further eliminate the interference of other high-frequency signals, the first-order derivative method is used to analyze the soil heavy metal stress series Yd to obtain the stable characteristics of heavy metal stress in the soil.
实施例2：本实施例基于多光谱遥感影像对一年两熟的水稻进行NDVI长时间序列构建，与实施例1的区别在于所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量Pr的波动周期为大于6。 Example 2: This example constructs a long NDVI series for rice that is harvested twice a year based on multispectral remote sensing images. The difference from Example 1 is that the preset screening condition of step 6 is that the fluctuation period of the IMF component P r of the NDVI long time series is greater than 6.

Claims (5)

  1. 一种基于长时序NDVI的土壤重金属胁迫甄别方法，包括以下步骤：A soil heavy metal stress identification method based on long-term NDVI includes the following steps:
    步骤1：遥感影像预处理：对N幅遥感影像数据进行辐射定标、大气校正以及几何校正预处理，N>1；Step 1: Remote sensing image preprocessing: Perform radiometric calibration, atmospheric correction and geometric correction preprocessing on N remote sensing image data, N>1;
    步骤2：NDVI长时间序列构建：基于预处理后的遥感影像数据计算NDVI长时间序列x(n),1≤n≤N：
    Step 2: Construction of NDVI long time series: Calculate the NDVI long time series x(n) based on the preprocessed remote sensing image data, 1≤n≤N:
    其中ρNIR(n)为第n幅遥感影像近红外波段的反射率，ρRED(n)为第n幅遥感影像红光波段的反射率；Where ρ NIR (n) is the reflectance of the near infrared band of the n-th remote sensing image, and ρ RED (n) is the reflectance of the red band of the n-th remote sensing image;
    步骤3：EEMD分解：基于EEMD算法对NDVI长时间序列x(n)进行分解，其分解流程如下：Step 3: EEMD decomposition: Decompose the NDVI long time series x(n) based on the EEMD algorithm. The decomposition process is as follows:
    将具有标准正态分布的白噪声信号zi(n)加入到NDVI长时间序列x(n)上,产生加噪NDVI长时间序列xi(n)：
    xi(n)＝x(n)+zi(n)，1≤i≤M    Add the white noise signal z i (n) with standard normal distribution to the NDVI long time series x(n) to generate the noisy NDVI long time series x i (n):
    x i (n) = x (n) + z i (n), 1 ≤ i ≤ M
    式中，M为加入标准正态分布的白噪声信号的次数；Where M is the number of times the standard normal distribution white noise signal is added;
    对加噪NDVI长时间序列xi(n)分别进行EMD分解，得到各自IMF和的形式:
    Perform EMD decomposition on the noisy NDVI long time series x i (n) to obtain the form of their respective IMF sums:
    式中，ci,j(n)为加噪NDVI长时间序列xi(n)的第j个IMF分量，ri,j(n)是加噪NDVI长时间序列的EMD残余函数，J是IMF分量的数量；Where c i,j (n) is the jth IMF component of the noisy NDVI long time series x i (n), ri,j (n) is the EMD residual function of the noisy NDVI long time series, and J is the number of IMF components;
    计算NDVI长时间序列x(n)的EEMD分解的第j个IMF分量：
    Calculate the jth IMF component of the EEMD decomposition of the NDVI long time series x(n):
    步骤4：计算统计性描述指标：计算NDVI长时间序列x(n)的EEMD分解的第一个至第J个IMF分量的极值点个数Kj、波动周期Pj：
    Step 4: Calculate statistical descriptive indicators: Calculate the number of extreme points K j and fluctuation period P j of the first to Jth IMF components of the EEMD decomposition of the NDVI long time series x(n):
    步骤5：季节平均模型构建：对NDVI长时间序列各年同月数据求取平均值，得到NDVI长时间序列的季节平均序列；对NDVI长时间序列的季节平均序列使用所述步骤3进行EEMD分解，得到NDVI长时间序列的季节平均序列的IMF分量imfr，计算NDVI长时间序列的季节平均序列的IMF分量imfr的周期，r＝1,2,…R；Step 5: seasonal average model construction: average the data of the same month of each year of the NDVI long-term series to obtain the seasonal average series of the NDVI long-term series; perform EEMD decomposition on the seasonal average series of the NDVI long-term series using the step 3 to obtain the IMF component imfr of the seasonal average series of the NDVI long-term series, and calculate the period of the IMF component imfr of the seasonal average series of the NDVI long-term series, r = 1, 2, ... R;
    步骤6：土壤重金属胁迫序列甄别：按照预设甄别条件选择符合长周期土壤重金属胁迫特征的NDVI长时间序列x(n)的EEMD分解的各IMF分量累加合成为土壤重金属胁迫序列Yd(n)；Step 6: Identification of soil heavy metal stress series: According to the preset identification conditions, the IMF components of the EEMD decomposition of the NDVI long time series x(n) that meets the long-term soil heavy metal stress characteristics are selected and synthesized into the soil heavy metal stress series Yd (n);
    步骤7：土壤重金属胁迫稳定特征提取：计算土壤重金属胁迫序列Yd的重金属胁迫稳定特征：
    Step 7: Extraction of soil heavy metal stress stability features: Calculate the heavy metal stress stability features of the soil heavy metal stress sequence Yd :
    式中：Ydt(n)为土壤重金属胁迫序列Yd(n)的前一时相的数值，Ydt+1(n)为土壤重金属胁迫序列Yd(n)的后一时相的数值，ΔDAY为前一时相对于后一时相对的日期间隔；Where: Y dt (n) is the value of the previous phase of the soil heavy metal stress sequence Y d (n), Y dt+1 (n) is the value of the next phase of the soil heavy metal stress sequence Y d (n), ΔDAY is the date interval between the previous phase and the next phase;
    步骤8：地面数据实测：使用便携式XRF分析仪对土壤中的预设种类的重金属元素含量测定；Step 8: Ground data measurement: Use a portable XRF analyzer to measure the content of preset heavy metal elements in the soil;
    步骤9：构建重金属预测模型：拟合重金属胁迫稳定特征Ydf(n)与土壤的预设种类的重金属元素含量间的关系，作为重金属预测模型；Step 9: Construct a heavy metal prediction model: fit the relationship between the heavy metal stress stability characteristic Y df (n) and the content of heavy metal elements of the preset types of soil as a heavy metal prediction model;
    步骤10：监测土壤重金属胁迫程度：采集遥感影像，进行所述步骤1-步骤7的处理，得到对应的重金属胁迫稳定特征，使用所述重金属预测模型预测土壤重金属胁迫程度。Step 10: Monitor the degree of heavy metal stress in the soil: collect remote sensing images, perform the processing of steps 1 to 7, obtain the corresponding heavy metal stress stability characteristics, and use the heavy metal prediction model to predict the degree of heavy metal stress in the soil.
  2. 根据权利要求1所述的基于长时序NDVI的土壤重金属胁迫甄别方法，其特征在于：所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量的波动周期Pr为大于T，T为NDVI长时间序列的IMF分量与其对应的NDVI长时间序列的季节平均序列的IMF分量的共同波动周期。 The soil heavy metal stress identification method based on long-term NDVI according to claim 1 is characterized in that: the preset identification condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than T, and T is the common fluctuation period of the IMF component of the NDVI long time series and the IMF component of the seasonal average series of the corresponding NDVI long time series.
  3. 根据权利要求1所述的基于长时序NDVI的土壤重金属胁迫甄别方法，其特征在于：所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量的波动周期Pr为大于6。The soil heavy metal stress identification method based on long-term NDVI according to claim 1 is characterized in that the preset identification condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 6.
  4. 根据权利要求1所述的基于长时序NDVI的土壤重金属胁迫甄别方法，其特征在于：所述步骤6的预设甄别条件为NDVI长时间序列的IMF分量的波动周期Pr大于12。The soil heavy metal stress identification method based on long-term NDVI according to claim 1 is characterized in that the preset identification condition of step 6 is that the fluctuation period P r of the IMF component of the NDVI long time series is greater than 12.
  5. 根据权利要求1或2所述的基于长时序NDVI的土壤重金属胁迫甄别方法，其特征在于：所述步骤9中使用二次方曲线拟合重金属胁迫稳定特征Ydf(n)与土壤的预设种类的重金属元素含量间的关系。 The soil heavy metal stress identification method based on long-term NDVI according to claim 1 or 2 is characterized in that: in step 9, a quadratic curve is used to fit the relationship between the heavy metal stress stability feature Y df (n) and the heavy metal element content of a preset type of soil.
PCT/CN2024/070257 2022-11-07 2024-01-03 Soil heavy metal stress identification method based on long time series ndvi WO2024099475A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211382540.1 2022-11-07
CN202211382540.1A CN115578640A (en) 2022-11-07 2022-11-07 Soil heavy metal stress discrimination method based on long-time-sequence NDVI

Publications (1)

Publication Number Publication Date
WO2024099475A1 true WO2024099475A1 (en) 2024-05-16

Family

ID=84589148

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2024/070257 WO2024099475A1 (en) 2022-11-07 2024-01-03 Soil heavy metal stress identification method based on long time series ndvi

Country Status (3)

Country Link
CN (1) CN115578640A (en)
LU (1) LU506698B1 (en)
WO (1) WO2024099475A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115578640A (en) * 2022-11-07 2023-01-06 北华航天工业学院 Soil heavy metal stress discrimination method based on long-time-sequence NDVI

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113111799A (en) * 2021-04-19 2021-07-13 北华航天工业学院 Cultivated land soil fertility level monitoring method based on ensemble empirical mode decomposition
CN115578640A (en) * 2022-11-07 2023-01-06 北华航天工业学院 Soil heavy metal stress discrimination method based on long-time-sequence NDVI

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113111799A (en) * 2021-04-19 2021-07-13 北华航天工业学院 Cultivated land soil fertility level monitoring method based on ensemble empirical mode decomposition
CN115578640A (en) * 2022-11-07 2023-01-06 北华航天工业学院 Soil heavy metal stress discrimination method based on long-time-sequence NDVI

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Chinese Master Thesis", 15 August 2018, CHINA UNIVERSITY OF GEOSCIENCES (BEIJING, CN, article TIAN LINGWEN 级: "Extraction of Rice Heavy Metal Stress Features Based on Long Time Series LAI and EEMD Method", pages: 1 - 90, XP093170608 *

Also Published As

Publication number Publication date
CN115578640A (en) 2023-01-06
LU506698B1 (en) 2024-06-28

Similar Documents

Publication Publication Date Title
CN113192117B (en) Method for improving estimation of leaf area index of early growth stage of wheat based on Sentinel-2 satellite image red edge band
Zeng et al. A review of vegetation phenological metrics extraction using time-series, multispectral satellite data
WO2024099475A1 (en) Soil heavy metal stress identification method based on long time series ndvi
Kowalik et al. Yield estimation using SPOT-VEGETATION products: A case study of wheat in European countries
JP5021293B2 (en) Crop growth status analysis method, crop growth status analysis device, and crop growth status analysis program
Giuffrida et al. A simple model for nondestructive leaf area estimation in bedding plants
CN113237836A (en) Flue-cured tobacco leaf moisture content estimation method based on hyperspectral image
CN113111799B (en) Cultivated land soil fertility level monitoring method based on ensemble empirical mode decomposition
WO2024098587A1 (en) Ceemdan-based farmland soil water stress discrimination and monitoring method
CN110579186A (en) Crop growth monitoring method based on inversion of leaf area index by inverse Gaussian process
Zhou et al. Increasing temperature shortened the carbon uptake period and decreased the cumulative net ecosystem productivity in a maize cropland in Northeast China
CN117077868A (en) Model-based grain yield prediction system and method
CN112129709A (en) Apple tree canopy scale nitrogen content diagnosis method
CN116757332B (en) Leaf vegetable yield prediction method, device, equipment and medium
CN114527082A (en) Sweet corn seed germination prediction method and device
Karakoc et al. Ratio-based vegetation indices for biomass estimation depending on grassland characteristics
CN113552096A (en) Spectrum-based pineapple leaf nitrogen content estimation method
CN111223002A (en) Corn region dry matter yield or silage yield evaluation method and system
Henke et al. Reconstructing leaf growth based on non-destructive digitizing and low-parametric shape evolution for plant modelling over a growth cycle
Luo et al. Evergreen broadleaf greenness and its relationship with leaf flushing, aging, and water fluxes
CN116187859A (en) Method, device, equipment and medium for monitoring and diagnosing low-temperature cold injury of crops in irrigation area
CN115266720A (en) Multi-scale spectral imaging corn growth period nutrient information detection method
CN114529838A (en) Soil nitrogen content inversion model construction method and system based on convolutional neural network
CN113390797A (en) Method for estimating total phosphorus content of pineapple leaves based on hyperspectral data
Cao et al. A spatiotemporal shape model fitting method for within-season crop phenology detection

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24725031

Country of ref document: EP

Kind code of ref document: A1