CN110987830A - Model, method and application for rapidly determining chlorophyll content of plant canopy leaves - Google Patents
Model, method and application for rapidly determining chlorophyll content of plant canopy leaves Download PDFInfo
- Publication number
- CN110987830A CN110987830A CN201911338503.9A CN201911338503A CN110987830A CN 110987830 A CN110987830 A CN 110987830A CN 201911338503 A CN201911338503 A CN 201911338503A CN 110987830 A CN110987830 A CN 110987830A
- Authority
- CN
- China
- Prior art keywords
- leaves
- chlorophyll content
- model
- training set
- spectral data
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 229930002875 chlorophyll Natural products 0.000 title claims abstract description 80
- 235000019804 chlorophyll Nutrition 0.000 title claims abstract description 80
- ATNHDLDRLWWWCB-AENOIHSZSA-M chlorophyll a Chemical compound C1([C@@H](C(=O)OC)C(=O)C2=C3C)=C2N2C3=CC(C(CC)=C3C)=[N+]4C3=CC3=C(C=C)C(C)=C5N3[Mg-2]42[N+]2=C1[C@@H](CCC(=O)OC\C=C(/C)CCC[C@H](C)CCC[C@H](C)CCCC(C)C)[C@H](C)C2=C5 ATNHDLDRLWWWCB-AENOIHSZSA-M 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000003595 spectral effect Effects 0.000 claims abstract description 57
- 238000012549 training Methods 0.000 claims abstract description 57
- 238000012795 verification Methods 0.000 claims abstract description 25
- 238000001514 detection method Methods 0.000 claims abstract description 19
- 238000005070 sampling Methods 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 238000010801 machine learning Methods 0.000 claims abstract description 8
- 238000001228 spectrum Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 14
- 239000000126 substance Substances 0.000 claims description 12
- 238000005259 measurement Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000009499 grossing Methods 0.000 claims description 7
- 238000010238 partial least squares regression Methods 0.000 claims description 7
- 230000002860 competitive effect Effects 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000010987 Kennard-Stone algorithm Methods 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000002310 reflectometry Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 239000003814 drug Substances 0.000 claims description 2
- 238000001914 filtration Methods 0.000 claims description 2
- 235000013311 vegetables Nutrition 0.000 claims description 2
- 235000013305 food Nutrition 0.000 claims 1
- 238000010200 validation analysis Methods 0.000 claims 1
- 238000003745 diagnosis Methods 0.000 abstract description 2
- 229940126680 traditional chinese medicines Drugs 0.000 abstract description 2
- 241000196324 Embryophyta Species 0.000 description 35
- 238000012544 monitoring process Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 235000014698 Brassica juncea var multisecta Nutrition 0.000 description 3
- 235000006008 Brassica napus var napus Nutrition 0.000 description 3
- 235000006618 Brassica rapa subsp oleifera Nutrition 0.000 description 3
- 241001301148 Brassica rapa subsp. oleifera Species 0.000 description 3
- 235000004977 Brassica sinapistrum Nutrition 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 241000238631 Hexapoda Species 0.000 description 2
- 241000607479 Yersinia pestis Species 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 235000003715 nutritional status Nutrition 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 229930091051 Arenine Natural products 0.000 description 1
- 241000499436 Brassica rapa subsp. pekinensis Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 238000001479 atomic absorption spectroscopy Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- XTLNYNMNUCLWEZ-UHFFFAOYSA-N ethanol;propan-2-one Chemical compound CCO.CC(C)=O XTLNYNMNUCLWEZ-UHFFFAOYSA-N 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000029553 photosynthesis Effects 0.000 description 1
- 238000010672 photosynthesis Methods 0.000 description 1
- 230000000243 photosynthetic effect Effects 0.000 description 1
- 230000008288 physiological mechanism Effects 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0098—Plants or trees
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Pathology (AREA)
- Immunology (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Botany (AREA)
- Wood Science & Technology (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention belongs to the technical field of plant chlorophyll content determination methods, and particularly relates to a method for rapidly determining the chlorophyll content of plant canopy leaves. The method comprises the steps of formulating a sampling scheme, collecting spectral data and measuring the chlorophyll content; preprocessing the spectral data; dividing a training set and a verification set; and constructing a training set model and a verification set model and the like. The invention creatively adopts a spectral diagnosis method and a machine learning method to construct a prediction model of the chlorophyll content of the plant canopy leaves, realizes the rapid acquisition of data, and can be used for the rapid detection of the chlorophyll content of the plant canopy leaves of traditional crops, traditional Chinese medicines and the like.
Description
Technical Field
The invention belongs to the technical field of plant chlorophyll content determination methods, and particularly relates to a model for rapidly determining the chlorophyll content of plant canopy leaves, a determination method and application.
Background
Chlorophyll content in plant leaves is an important parameter for understanding plant physiological mechanisms and productivity, and is an important index for evaluating plant photosynthetic capacity, development status and nutritional status. The change of the chlorophyll content in different growth periods can reflect the photosynthesis intensity of crops and the growth period conditions of the crops, so that the research of estimating the chlorophyll content in plant leaves has important significance for monitoring vegetation growth, estimating crop yield and preventing plant diseases and insect pests. Traditional measuring methods for chlorophyll content comprise an atomic absorption spectrometry method and a spectrophotometer method, wherein fresh plant leaves or dry materials need to be collected, complicated pretreatment and chemical experiments are needed, time and labor are wasted, environmental pollution to a certain degree is caused, and quick acquisition of vegetation chlorophyll information cannot be realized. It is difficult to meet the requirements of real-time, rapid, non-destructive and large-area monitoring. The hyperspectral remote sensing technology has the characteristics of time saving, low price, small sample amount, no damage to the sample structure, rich information and the like, and can realize the monitoring of the crop growth and the nutritional status under the condition of ensuring the integrity of the plant structure.
The remote sensing technology has unique advantages in chlorophyll content monitoring. The chlorophyll content of the canopy can reflect the information of a single plant and the growth condition of the whole vegetation, and is an important component of remote sensing monitoring. Therefore, real-time monitoring of the chlorophyll content in the canopy is of great significance for remote monitoring of crop growth, estimation of crop yield and prevention of plant diseases and insect pests. Since chlorophyll has a unique biochemical structure, spectral absorption due to its electron transition is generally located in the visible region. In recent years, researchers at home and abroad have conducted a great deal of research on hyperspectral inversion of chlorophyll content of leaves in a visible light range. While the study of inverting chlorophyll over the entire band is rare. The near infrared spectrum has the incomparable characteristic and has the advantages of high speed, accuracy, simplicity and low cost. When the spectrum in the visible light range is simply detected, the spectral data of the blade cannot be comprehensively obtained, so that the measurement error is generated. Therefore, a rapid and nondestructive method for detecting the chlorophyll content of plant canopy leaves based on spectrum needs to be established again.
Disclosure of Invention
In order to solve the technical problems, the invention provides a model for rapidly measuring the chlorophyll content of plant canopy leaves, a measuring method and application.
The invention aims to provide a method for rapidly measuring the chlorophyll content of plant canopy leaves, which comprises the following steps:
step 1, formulating a sampling scheme, collecting spectral data and measuring chlorophyll content
Randomly selecting a plurality of undamaged complete plant canopy leaves as detection object leaves;
measuring the spectral reflectivity of each detected object blade in the near infrared and visible light ranges to obtain blade spectral data for later use;
collecting the detected object leaves, and measuring the chlorophyll content of the detected object leaves by a chemical method for later use;
step 2, preprocessing the spectral data
Step 2.1, removing abnormal values of the spectral data of the blades, and performing nine-point smoothing processing on the spectral data by using a filtering fitting method;
step 2.2, performing standard normal transformation, reciprocal logarithm and first derivative processing on the spectrum data subjected to the nine-point smoothing processing in sequence to obtain preprocessed spectrum data;
step 3, dividing the training set and the verification set
Adopting a Kennard-Stone algorithm to randomly divide the preprocessed spectral data of the detected object blade and the corresponding chlorophyll data into a training set and a verification set;
step 4, constructing a training set model and a verification set model
The training set model and the verification set model have the same construction steps, specifically as follows:
step 4.1, extracting spectral data characteristic information
Extracting characteristic information of the preprocessed spectral data by jointly adopting a continuous projection algorithm and a competitive adaptive reweighting sampling algorithm;
step 4.2, model construction
Constructing a training set model for the extracted characteristic information of the training set by adopting a partial least squares regression machine learning method; constructing a verification set model for the extracted verification set characteristic information;
step 4.3, verifying and evaluating model stability and prediction capability
Comparing and analyzing the chlorophyll content result of the training set model with the chlorophyll content result of the training set model, and verifying and evaluating the stability and the prediction capability of the training set model; and selecting a training set model close to the measurement result of the chemical chlorophyll content as an optimal chlorophyll content prediction model for measuring the chlorophyll content of plant canopy leaves.
Preferably, in the method for rapidly determining the chlorophyll content in the plant canopy leaves, in step 1, the sampling scheme is as follows: selecting 12 test areas in total, and randomly selecting ten undamaged complete rape canopy leaves in each test area as detection object leaves; the ten detection object leaves are positioned in the canopy of ten different rape plants, and the height positions and the horizontal positions of the ten detection object leaves are different.
Preferably, in the method for rapidly determining the chlorophyll content of the plant canopy leaves, the boundary effect is removed during sampling, namely the rape plants at the boundary of the test area and the boundary leaves in the rape plants are not taken as sampling objects.
Preferably, in the method for rapidly measuring the chlorophyll content in the plant canopy leaves, in step 1, the leaf spectral data measuring method of a single detection object leaf is as follows: spectral reflectance measurements are performed on the blade to be detected in four directions by using a portable field spectrometer ASDFieldspec4, and average spectral data is taken as blade spectral data of the blade to be detected.
Preferably, in the method for rapidly measuring the chlorophyll content in the plant canopy leaves, in the step 1, the wave bands in the near infrared and visible light ranges are 350-2500 nm.
Preferably, in the method for rapidly measuring the chlorophyll content in the plant canopy leaves, in step 3, 70% of the preprocessed spectral data of the leaves to be detected and the corresponding chlorophyll data thereof are selected as training set samples, and the leaf spectral data of the remaining 30% of the leaves to be detected are selected as verification set samples.
Preferably, in the method for rapidly determining the chlorophyll content of the plant canopy leaves, in step 4.3, the stability and the prediction capability of a training set model are verified and evaluated by adopting a root mean square error, a decision coefficient and a relative analysis deviation; r2The larger the RPD value is, the better the model prediction capability is, and the stronger the stability is; the evaluation criteria for RPD are as follows: RPD>3 represents that the model has excellent prediction capability; 2<RPD<3 that the model has limited predictive power; RPD<2 indicates that the model has no predictive capability.
The invention also provides a chlorophyll content prediction model constructed by the method for rapidly determining the chlorophyll content of the plant canopy leaves.
The invention also provides application of the chlorophyll content prediction model in measuring the chlorophyll content of the canopy leaves of vegetables, grain crops and traditional Chinese medicine plants.
Compared with the prior art, the model, the method and the application for rapidly determining the chlorophyll content of the plant canopy leaves have the following beneficial effects:
the invention creatively adopts a spectral diagnosis method and a machine learning method to construct a prediction model of the chlorophyll content of the plant canopy leaves, realizes the rapid acquisition of data, and can be used for the rapid detection of the chlorophyll content of the plant canopy leaves of traditional crops, traditional Chinese medicines and the like.
In order to effectively extract the relevant information of the wave bands in the spectral data and establish a reliable model, the selected training set has good representativeness. The invention adopts a spectral analysis algorithm Kennard-Stone (KS) to randomly divide the acquired data into a training set and a verification set, thereby effectively avoiding the problems caused by uneven distribution of chlorophyll and spectral values of the vegetation leaves.
Because the spectral data has higher resolution and is easy to generate redundancy problem, the invention comprehensively adopts the continuous projection algorithm and the competitive adaptive reweighting sampling algorithm to extract the characteristic information of the rape spectrum, thereby effectively reducing the data redundancy problem and improving the representativeness of the characteristic information of the spectral data.
Aiming at the chlorophyll content of rape, the embodiment of the invention constructs a training set model based on PLSR machine learning; by comparing the prediction result of the training set model with the verification set, the modeling precision and the prediction precision of the training set model established after the spectral data preprocessing are superior to those of the training set model based on reflectivity (namely, the training set model is not subjected to the spectral data preprocessing).
In addition, compared with the traditional chemical method, the data processing method and the machine learning method provided by the invention can effectively save cost, avoid the use of chemical reagents in the conventional measurement method, ensure the integrity of vegetation, and have obvious advantages in the aspects of prediction precision and universality.
Drawings
FIG. 1 is blade spectrum data collected from a part of blades of an inspection object in example 1;
FIG. 2 is a comparison of predicted and chemically measured values for a model of canola canopy leaves.
Detailed Description
In order that those skilled in the art will better understand the technical solutions of the present invention to be implemented, the present invention will be further described with reference to the following specific embodiments and accompanying drawings. The following examples, as well as test methods not specifically identified in the summary of the invention, were conducted according to methods and conditions conventional in the art.
The rape sample planting area is located in the west of the plain in the Guanzhong of Shaanxi province and belongs to the administrative district in the Mei county. Its south part is the mountain range of Qinling mountain, and its north part is the Weishui river, belonging to valley region of midstream yellow river. Mei county belongs to warm temperate continental semi-humid climate with 107 degrees 39-108 degrees 00 'for east longitude and 33 degrees 59-34 degrees 19' for north latitude. The altitude is 442-3767m, the average annual temperature is about 12.9 ℃, the average annual precipitation is 609.5mm, the average annual sunshine time is 2015.2h, and the frost-free period is 21 d.
Example 1
A method for rapidly measuring the chlorophyll content of plant canopy leaves comprises the following steps:
step 1, formulating a sampling scheme, collecting spectral data and measuring chlorophyll content
A total of 12 test zones were selected for this experiment. Randomly selecting ten undamaged complete rape canopy leaves in each test area as a detection object leaf; the ten detection object leaves are positioned in the canopy of ten different rape plants, and the height positions and the horizontal positions of the ten detection object leaves are different, so that canopy leaf information with different heights and horizontal positions can be conveniently reflected; the 12 test areas total 120 test object blades. It should be noted that the boundary effect should be removed during sampling, i.e. the rape plants at the boundary of the test area and the boundary leaves in the rape plants are not taken as the sampling objects.
The blade spectral data measuring method of the single blade to be detected comprises the following steps: the spectral reflectance measurement in the near infrared and visible light ranges is performed on each blade to be detected by using a portable geophysical spectrometer ASDFieldspec4, and the average spectral data is taken as the blade spectral data (R) of the blade to be detected.
FIG. 1 is blade spectral data collected from a portion of a blade of an inspection object.
Collecting the leaves of a detection object, putting the leaves into a fresh-keeping bag, storing the leaves in a laboratory, and measuring the chlorophyll content of the leaves of the detection object by utilizing a conventional acetone-ethanol mixing method (Lvcez, L ü peize, influence of salicylic acid-soaked seeds on partial physiological indexes of the brassica rapa pekinensis type Gansu agricultural science and technology, 2017(4): 41-45.).
Step 2, preprocessing the spectral data
And 2.1, removing abnormal values of the spectral data of the blades, and performing nine-point smoothing on the spectral data by using a filter fitting method (Savitzky-Golay, SG). The filter parameters determined by our experiments are the best when the filter parameters are nine, so that the spectral data are subjected to nine-point smoothing.
And 2.2, sequentially carrying out standard normal transformation (SNV), reciprocal Logarithm (LOG) and First Derivative (FD) processing on the spectrum data subjected to the nine-point smoothing processing to obtain preprocessed spectrum data.
Step 3, dividing the training set and the verification set
Randomly dividing collected preprocessing spectrum data of the detected object leaves and corresponding chlorophyll data thereof into a training set and a verification set by adopting a Kennard-Stone (KS) algorithm, wherein 70% of preprocessing spectrum data of the detected object leaves and corresponding chlorophyll data thereof are selected as training set samples, and 84 samples in total are used for constructing a training set model; and (3) taking the preprocessed spectral data of the remaining 30 percent of the leaves of the detection object and the corresponding chlorophyll data thereof as verification set samples, and using 36 samples in total to construct a verification set model.
The training set selected by the KS algorithm has good representativeness, and relevant information can be well extracted to establish a reliable prediction model, so that the problem caused by uneven distribution of chlorophyll chemical values and spectral values is effectively avoided.
Step 4, constructing a training set model and a verification set model
The training set model and the verification set model have the same construction steps, and the following description of the modeling method takes the training set as an example, and specifically includes the following steps:
step 4.1, extracting spectral data characteristic information: and extracting the characteristic information of the rape preprocessing spectrum data of the training set by adopting a method of a Successive Projection Algorithm (SPA) + competitive adaptive re-weighted sampling algorithm (CARS). The characteristic information of rape light preprocessing spectrum data in the training set is extracted by respectively adopting a continuous projection algorithm and a competitive self-adaptive re-weighting sampling algorithm, and the coincident characteristic information in the continuous projection algorithm and the competitive self-adaptive re-weighting sampling algorithm is taken as effective characteristic information to be used for next modeling, so that the problem of data redundancy is effectively reduced, and the representativeness of the characteristic information of the spectrum data is improved.
Step 4.2, constructing a model: adopting a Partial Least Squares Regression (PLSR) machine learning method to construct a training set model for the extracted training set characteristic information;
step 4.3, verifying and evaluating model stability and prediction capability
Comparing and analyzing the chlorophyll content result of the training set model with the chlorophyll content result of the training set model, and adopting Root Mean Square Error (RMSE) and Coefficient of determination (R)2) And the relative analysis deviation (RPD) and other parameters to verify and evaluate the stability and the prediction capability of the training set model.
In formulae (1) to (5), yiChemical measurements representing samples of the training set are shown,
the model prediction values representing the training set samples,
chemical assays representing training set samplesThe average of the magnitudes, n represents the number of samples.
In the formula, R2The larger the RPD value is, the better the model prediction ability is, and the stronger the stability is. The evaluation criteria for RPD are as follows: RPD>3 represents that the model has excellent prediction capability; 2<RPD<3 that the model has limited predictive power; RPD<2 indicates that the model has no predictive capability.
The model construction method of the above embodiment is briefly described as "R + SG + SNV + LOG + FD".
Aiming at the chlorophyll content of rape in a test area, a training set model based on PLSR machine learning is constructed; by comparing the PLSR prediction result of the training set model with the verification set, the modeling precision and the prediction precision of the training set model established after the spectral data preprocessing are superior to those of the training set model based on reflectivity (namely, the training set model is not subjected to the spectral data preprocessing).
By comparing the indices, the training set of example 1 resulted in R20.98, RPD 7.08, RMSE 3.12, SE 3.16, Slope 0.98, Offset 1.61; verification set result is R2The model constructed by the method of R + SG + SNV + LOG + FD is regarded as an excellent model because the model is 0.98, RPD is 7.52, RMSE is 2.94, SE is 2.98, slope is 0.98, and offset is 1.43.
In the test area, some rape canopy leaves are selected as detection objects again, the model constructed by the method in the embodiment 1 is used for predicting the chlorophyll content value of the newly-taken rape canopy leaves, and the chemical method in the embodiment 1 is used for measuring the actual chlorophyll content value. FIG. 2 is a comparison of chlorophyll content values of freshly taken canola canopy leaves (canola samples in test zones) measured using the model constructed in accordance with the method of example 1 of the present invention and their corresponding chemical measurements. The result shows that the measured value of the model is close to the actual chemical measured value, which shows that the model constructed by the method has high accuracy and strong practicability.
It should be noted that, when the present invention relates to a numerical range, it should be understood that two endpoints of each numerical range and any value between the two endpoints can be selected, and since the steps and methods adopted are the same as those in the embodiment, in order to prevent redundancy, the present invention describes a preferred embodiment. While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (9)
1. A method for rapidly measuring the chlorophyll content of plant canopy leaves is characterized by comprising the following steps:
step 1, formulating a sampling scheme, collecting spectral data and measuring chlorophyll content
Randomly selecting a plurality of undamaged complete plant canopy leaves as detection object leaves;
measuring the spectral reflectivity of each detected object blade in the near infrared and visible light ranges to obtain blade spectral data for later use;
collecting the detected object leaves, and measuring the chlorophyll content of the detected object leaves by a chemical method for later use;
step 2, preprocessing the spectral data
Step 2.1, removing abnormal values of the spectral data of the blades, and performing nine-point smoothing processing on the spectral data by using a filtering fitting method;
step 2.2, performing standard normal transformation, reciprocal logarithm and first derivative processing on the spectrum data subjected to the nine-point smoothing processing in sequence to obtain preprocessed spectrum data;
step 3, dividing the training set and the verification set
Adopting a Kennard-Stone algorithm to randomly divide the preprocessed spectral data of the detected object blade and the corresponding chlorophyll data into a training set and a verification set;
step 4, constructing a training set model and a verification set model
The training set model and the verification set model have the same construction steps, specifically as follows:
step 4.1, extracting spectral data characteristic information
Extracting characteristic information of the preprocessed spectral data by jointly adopting a continuous projection algorithm and a competitive adaptive reweighting sampling algorithm;
step 4.2, model construction
Constructing a training set model for the extracted characteristic information of the training set by adopting a partial least squares regression machine learning method; constructing a verification set model for the extracted verification set characteristic information;
step 4.3, verifying and evaluating model stability and prediction capability
Comparing and analyzing the chlorophyll content result of the training set model with the chlorophyll content result of the training set model, and verifying and evaluating the stability and the prediction capability of the training set model; and selecting a training set model close to the measurement result of the chemical chlorophyll content as an optimal chlorophyll content prediction model for measuring the chlorophyll content of plant canopy leaves.
2. The method for rapidly measuring the chlorophyll content in the plant canopy leaves according to claim 1, wherein in the step 1, the sampling scheme is as follows: selecting 12 test areas in total, and randomly selecting ten undamaged complete rape canopy leaves in each test area as detection object leaves; the ten detection object leaves are positioned in the canopy of ten different rape plants, and the height positions and the horizontal positions of the ten detection object leaves are different.
3. The method for rapidly determining the chlorophyll content in the plant canopy leaves as claimed in claim 2, wherein the boundary effect is removed during sampling, i.e. the rape plants at the boundary of the test area and the boundary leaves in the rape plants are not taken as the sampling objects.
4. The method for rapidly measuring the chlorophyll content in the leaves of the plant canopy according to claim 1, wherein in the step 1, the leaf spectral data of the leaves of a single detection object are measured by the following method: spectral reflectance measurements are performed on the blade to be detected in four directions by using a portable field spectrometer ASDFieldspec4, and average spectral data is taken as blade spectral data of the blade to be detected.
5. The method for rapidly measuring the chlorophyll content in the leaves of the plant canopy as claimed in claim 1, wherein in step 1, the wavelength ranges of the near infrared and visible light are 350-2500 nm.
6. The method for rapidly measuring the chlorophyll content in the plant canopy leaves as claimed in any one of claims 1-5, wherein in step 3, 70% of the pre-processed spectral data of the leaves to be detected and the corresponding chlorophyll data thereof are selected as training set samples, and the remaining 30% of the leaf spectral data of the leaves to be detected are selected as validation set samples.
7. The method for rapidly measuring the chlorophyll content in the plant canopy leaves as claimed in claim 6, wherein in step 4.3, the stability and the prediction ability of the training set model are verified and evaluated by adopting the root mean square error, the decision coefficient and the relative analysis deviation; r2The larger the RPD value is, the better the model prediction capability is, and the stronger the stability is; the evaluation criteria for RPD are as follows: RPD>3 represents that the model has excellent prediction capability; 2<RPD<3 that the model has limited predictive power; RPD<2 indicates that the model has no predictive capability.
8. The chlorophyll content prediction model constructed by the plant canopy leaf chlorophyll content rapid determination method according to claim 1.
9. The use of the chlorophyll content prediction model according to claim 8 for determining chlorophyll content in canopy leaves of vegetables, food crops, and traditional Chinese medicine plants.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911338503.9A CN110987830A (en) | 2019-12-23 | 2019-12-23 | Model, method and application for rapidly determining chlorophyll content of plant canopy leaves |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911338503.9A CN110987830A (en) | 2019-12-23 | 2019-12-23 | Model, method and application for rapidly determining chlorophyll content of plant canopy leaves |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110987830A true CN110987830A (en) | 2020-04-10 |
Family
ID=70075725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911338503.9A Pending CN110987830A (en) | 2019-12-23 | 2019-12-23 | Model, method and application for rapidly determining chlorophyll content of plant canopy leaves |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110987830A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112129709A (en) * | 2020-09-16 | 2020-12-25 | 西北农林科技大学 | Apple tree canopy scale nitrogen content diagnosis method |
| CN114324215A (en) * | 2021-12-31 | 2022-04-12 | 重庆市农业科学院 | Lemon leaf chlorophyll content and two-dimensional distribution detection method thereof |
| CN114460015A (en) * | 2021-10-29 | 2022-05-10 | 东北农业大学 | Method and model for predicting chlorophyll content of canopy of rice in cold region |
| CN116660206A (en) * | 2023-05-31 | 2023-08-29 | 浙江省农业科学院 | Crop yield estimation method and system |
| CN113686821B (en) * | 2021-08-27 | 2024-06-11 | 广西壮族自治区中国科学院广西植物研究所 | Karst plant leaf TN content nondestructive monitoring method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101762463A (en) * | 2009-12-16 | 2010-06-30 | 中国烟草总公司郑州烟草研究院 | Method for measuring chlorophyll content of fresh tobacco leaf of flue-cured tobacco based on canopy multi-spectra |
| CN102507453A (en) * | 2011-10-25 | 2012-06-20 | 南京林业大学 | Hyperspectral monitoring method of pine wood nematode disease of black pine |
| CN103345707A (en) * | 2013-06-04 | 2013-10-09 | 中国科学院遥感与数字地球研究所 | Crop maturation stage remote sensing prediction method based on multi-source remote sensing data |
| CN105277491A (en) * | 2015-09-24 | 2016-01-27 | 中国农业科学院农业资源与农业区划研究所 | Chlorophyll content measurement method and apparatus thereof |
| CN108195794A (en) * | 2017-11-21 | 2018-06-22 | 山东大学 | A kind of method of Lumbrokinase potency near infrared ray earthworm extraction intermediate |
| CN109030378A (en) * | 2018-06-04 | 2018-12-18 | 沈阳农业大学 | Japonica rice canopy chlorophyll content inverse model approach based on PSO-ELM |
| CN109657653A (en) * | 2019-01-21 | 2019-04-19 | 安徽大学 | A kind of wheat seed head blight recognition methods based on Imaging Hyperspectral Data |
| CN110160967A (en) * | 2019-04-16 | 2019-08-23 | 安徽大学 | A kind of total nitrogen content evaluation method of crop canopies blade |
-
2019
- 2019-12-23 CN CN201911338503.9A patent/CN110987830A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101762463A (en) * | 2009-12-16 | 2010-06-30 | 中国烟草总公司郑州烟草研究院 | Method for measuring chlorophyll content of fresh tobacco leaf of flue-cured tobacco based on canopy multi-spectra |
| CN102507453A (en) * | 2011-10-25 | 2012-06-20 | 南京林业大学 | Hyperspectral monitoring method of pine wood nematode disease of black pine |
| CN103345707A (en) * | 2013-06-04 | 2013-10-09 | 中国科学院遥感与数字地球研究所 | Crop maturation stage remote sensing prediction method based on multi-source remote sensing data |
| CN105277491A (en) * | 2015-09-24 | 2016-01-27 | 中国农业科学院农业资源与农业区划研究所 | Chlorophyll content measurement method and apparatus thereof |
| CN108195794A (en) * | 2017-11-21 | 2018-06-22 | 山东大学 | A kind of method of Lumbrokinase potency near infrared ray earthworm extraction intermediate |
| CN109030378A (en) * | 2018-06-04 | 2018-12-18 | 沈阳农业大学 | Japonica rice canopy chlorophyll content inverse model approach based on PSO-ELM |
| CN109657653A (en) * | 2019-01-21 | 2019-04-19 | 安徽大学 | A kind of wheat seed head blight recognition methods based on Imaging Hyperspectral Data |
| CN110160967A (en) * | 2019-04-16 | 2019-08-23 | 安徽大学 | A kind of total nitrogen content evaluation method of crop canopies blade |
Non-Patent Citations (3)
| Title |
|---|
| 刘燕德 等: "基于高光谱成像技术的脐橙叶片的叶绿素含量及其分布测量", 《发光学报》 * |
| 贺英 等: "基于数码相机的玉米冠层 SPAD 遥感估算", 《中国农业科学》 * |
| 马文君 等: "棉花全生育期叶片 SPAD 值的遥感估算模型", 《干旱地区农业研究》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112129709A (en) * | 2020-09-16 | 2020-12-25 | 西北农林科技大学 | Apple tree canopy scale nitrogen content diagnosis method |
| CN113686821B (en) * | 2021-08-27 | 2024-06-11 | 广西壮族自治区中国科学院广西植物研究所 | Karst plant leaf TN content nondestructive monitoring method |
| CN114460015A (en) * | 2021-10-29 | 2022-05-10 | 东北农业大学 | Method and model for predicting chlorophyll content of canopy of rice in cold region |
| CN114324215A (en) * | 2021-12-31 | 2022-04-12 | 重庆市农业科学院 | Lemon leaf chlorophyll content and two-dimensional distribution detection method thereof |
| CN116660206A (en) * | 2023-05-31 | 2023-08-29 | 浙江省农业科学院 | Crop yield estimation method and system |
| CN116660206B (en) * | 2023-05-31 | 2024-05-28 | 浙江省农业科学院 | Crop yield estimation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110987830A (en) | Model, method and application for rapidly determining chlorophyll content of plant canopy leaves | |
| CN110376167A (en) | Rice leaf nitrogen content monitoring method based on unmanned plane EO-1 hyperion | |
| CN107796764B (en) | Method for constructing wheat leaf area index estimation model based on three-band vegetation index | |
| CN102435564B (en) | Method for estimating plant nitrogen content based on three-band spectral index | |
| CN102426153B (en) | A kind of Wheat plant moisture monitoring method based on canopy high spectral index | |
| Lamb et al. | Estimating leaf nitrogen concentration in ryegrass (Lolium spp.) pasture using the chlorophyll red-edge: theoretical modelling and experimental observations | |
| CN108875913B (en) | Tricholoma matsutake rapid nondestructive testing system and method based on convolutional neural network | |
| CN113268923B (en) | Summer corn yield estimation method based on simulated multispectral | |
| CN109187398A (en) | A kind of EO-1 hyperion measuring method of wheat plant nitrogen content | |
| CN110567892B (en) | Summer corn nitrogen hyperspectral prediction method based on critical nitrogen concentration | |
| CN111062251A (en) | Monitoring method of farmland cotton aphid pest grade model based on unmanned aerial vehicle imaging | |
| Song et al. | Using multi-angle hyperspectral data to monitor canopy leaf nitrogen content of wheat | |
| CN103868880A (en) | Wheat leaf nitrogen content monitoring method based on spectrum double-peak index and method for establishing monitoring model | |
| CN103472009A (en) | Method for monitoring wheat plant water content under different plant nitrogen content levels | |
| Yang et al. | Rapid determination of leaf water content for monitoring waterlogging in winter wheat based on hyperspectral parameters | |
| CN114140695B (en) | Prediction method and system for tea tree nitrogen diagnosis and quality index determination based on unmanned aerial vehicle multispectral remote sensing | |
| Huang et al. | Detection of scab in wheat ears using in situ hyperspectral data and support vector machine optimized by genetic algorithm | |
| CN111882242A (en) | Evaluation method of vegetation index in herbage biomass estimation research | |
| CN115184322A (en) | Rice leaf water content monitoring method | |
| CN112129709A (en) | Apple tree canopy scale nitrogen content diagnosis method | |
| CN112924401A (en) | Semi-empirical inversion method for chlorophyll content of vegetation canopy | |
| Wang et al. | Monitoring model for predicting maize grain moisture at the filling stage using NIRS and a small sample size | |
| Zhang et al. | A comparative study on wheat leaf area index by different measurement methods | |
| Ling et al. | Winter wheat biomass estimation based on canopy spectra | |
| CN113504186B (en) | Method for estimating utilization rate of nitrogen fertilizer in wheat by remote sensing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |