CN114946485B - Method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA - Google Patents
Method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA Download PDFInfo
- Publication number
- CN114946485B CN114946485B CN202210521422.8A CN202210521422A CN114946485B CN 114946485 B CN114946485 B CN 114946485B CN 202210521422 A CN202210521422 A CN 202210521422A CN 114946485 B CN114946485 B CN 114946485B
- Authority
- CN
- China
- Prior art keywords
- eucalyptus
- opls
- yellowing
- healthy
- disease
- 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.)
- Active
Links
- 201000010099 disease Diseases 0.000 title claims abstract description 70
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 28
- 206010022971 Iron Deficiencies Diseases 0.000 title claims abstract description 12
- 244000166124 Eucalyptus globulus Species 0.000 title abstract 8
- 238000004383 yellowing Methods 0.000 claims abstract description 54
- 230000003595 spectral effect Effects 0.000 claims abstract description 44
- 238000004458 analytical method Methods 0.000 claims abstract description 30
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000010200 validation analysis Methods 0.000 claims abstract description 3
- 241000219927 Eucalyptus Species 0.000 claims description 56
- 238000001228 spectrum Methods 0.000 claims description 40
- 238000012795 verification Methods 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000002310 reflectometry Methods 0.000 claims description 5
- -1 50W halogen Chemical class 0.000 claims description 3
- 229910052736 halogen Inorganic materials 0.000 claims description 3
- 238000012886 linear function Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000012300 Sequence Analysis Methods 0.000 claims 1
- 208000003152 Yellow Fever Diseases 0.000 claims 1
- 241000196324 Embryophyta Species 0.000 abstract description 9
- 238000011835 investigation Methods 0.000 abstract description 6
- 238000011161 development Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 239000000523 sample Substances 0.000 description 24
- 238000010239 partial least squares discriminant analysis Methods 0.000 description 15
- 239000002689 soil Substances 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 208000015181 infectious disease Diseases 0.000 description 3
- 238000011551 log transformation method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000003337 fertilizer Substances 0.000 description 2
- 230000012010 growth Effects 0.000 description 2
- 239000011572 manganese Substances 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000282994 Cervidae Species 0.000 description 1
- 241000238631 Hexapoda Species 0.000 description 1
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 1
- PWHULOQIROXLJO-UHFFFAOYSA-N Manganese Chemical compound [Mn] PWHULOQIROXLJO-UHFFFAOYSA-N 0.000 description 1
- 241000607479 Yersinia pestis Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 229910052748 manganese Inorganic materials 0.000 description 1
- 238000004204 optical analysis method Methods 0.000 description 1
- 238000012567 pattern recognition method Methods 0.000 description 1
- 230000008635 plant growth Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002023 wood Substances 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
- A01G13/00—Protecting plants
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
- A01G7/00—Botany in general
- A01G7/04—Electric or magnetic or acoustic treatment of plants for promoting growth
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
- A01G7/00—Botany in general
- A01G7/06—Treatment of growing trees or plants, e.g. for preventing decay of wood, for tingeing flowers or wood, for prolonging the life of plants
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA, and belongs to the technical field of plant disease monitoring. The method comprises the following technical steps: (1) leaf sample collection; (2) spectral data acquisition; (3) spectral data processing; (4) constructing an OPLS-DA discriminant analysis model; (5) independent sample validation. According to the invention, the VNIR technology is combined with the OPLS-DA discriminant analysis model, so that the disease condition of the eucalyptus yellow can be predicted efficiently, accurately and objectively, the problem that the eucalyptus yellow is difficult to identify accurately due to the fact that the eucalyptus yellow has burstiness, concealment and randomness is solved, and the error caused by subjectivity of artificial investigation is avoided. According to the method, when the eucalyptus leaves do not show the characteristic of the yellowing disease, the disease condition of the eucalyptus yellowing disease can be predicted efficiently, accurately and objectively, so that corresponding measures are taken in time to prevent the occurrence of the eucalyptus yellowing disease, and the method has great significance in guaranteeing the high-quality development of artificial forests and downstream industries thereof.
Description
Technical Field
The invention belongs to the technical field of plant disease monitoring, and particularly relates to a method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA.
Background
Eucalyptus is an important economic tree species in the southern area of China and makes an important contribution to promoting the economic development in the southern area. However, due to the long-term adoption of short-cycle, high-strength and extensive management modes, the problem of soil productivity decline causes the occurrence of eucalyptus yellowing diseases in recent years, and the development of eucalyptus artificial forests and downstream industries thereof is severely limited. Eucalyptus yellowing is a special physiological disease, and is characterized in that plants lose green after the occurrence of the disease, yellow leaves grow, during the period of yellowing, the plant growth stagnates, the new leaf extraction speed is extremely slow, and the annual growth quantity is reduced by about 20-40%. The method for applying the effective iron (Fe) and manganese (Mn) fertilizers and spraying the foliar fertilizer in the early stage of the occurrence of the yellowing disease can obviously reduce the economic loss caused by the yellowing disease.
Limited by the space variability of forest soil height and higher soil investigation cost, the cost for preventing eucalyptus yellow disease by large-scale soil investigation and laboratory analysis is higher. More importantly, eucalyptus yellowing disease is affected by the change of hydrothermal conditions, sudden and random occurrence is caused when the eucalyptus yellowing disease occurs, and often woodland operators find that serious economic loss is caused after the woods turn yellow, and how to accurately identify the yellowing disease when eucalyptus plants do not show diseases is a problem to be solved urgently in the forestry industry at present.
Visible-near infrared spectrum (VNIR) is a relatively rapidly developing optical analysis technique in recent years, which has been applied to study wood properties and to acquire physiological information of plants. The VNIR detection technology mainly utilizes absorption of light quanta when hydrogen-containing groups in substance structural molecules in a wave band of 350-2500 nm vibrate, has stronger penetrability, extracts spectral response characteristics of plants or tissues and cells stressed by diseases and insect pests, carries out formalization and quantitative expression on the spectral response characteristics of the plants, and can judge the infection degree, infection type, infection stage and the like of the plants. Compared with the traditional monitoring mode of field investigation by means of naked eye identification, the VNIR detection technology is more efficient, accurate and objective, allows the growth condition of plants to be monitored, analyzed and evaluated in real time in the whole growing season, and reasonable prevention and treatment measures are formulated, but the technology has less research on screening and monitoring of eucalyptus yellowing diseases.
PLS-DA (Partial least squares Discriminant Analysis) discriminant analysis is a supervised pattern recognition method based on Partial Least Squares (PLS), taking into account auxiliary matrices in constructing factors, and representing class member information in the form of codes. The OPLS-DA discriminant analysis is an extension of the partial least squares method, and by adding Orthogonal verification (orthoonal) to PLS to eliminate the complexity of the uncorrelated variable reduction model, the model built has better fitness.
Because the eucalyptus yellow disease has burstiness, concealment and randomness, the invention designs an effective method for predicting the disease condition of the eucalyptus yellow disease for more efficient, accurate and objective prediction, and further adopts corresponding measures for prevention and control, thereby having great significance for guaranteeing high-quality development of artificial forests and downstream industries thereof.
Disclosure of Invention
Aiming at the problems, the invention provides a method for predicting eucalyptus iron deficiency yellow based on VNIR and OPLS-DA, which can efficiently, accurately and objectively predict the disease condition of eucalyptus yellow.
The invention is realized by the following technical scheme:
a method for predicting eucalyptus iron deficiency yellow based on VNIR and OPLS-DA comprises the following steps:
(1) Blade sample collection: collecting a yellowing leaf sample and a healthy leaf sample which does not show diseases in a eucalyptus yellowing forest zone, and collecting a healthy leaf sample in a healthy eucalyptus forest zone with consistent standing conditions of the same forest Ban Na;
(2) And (3) spectrum data acquisition: collecting spectral data of a yellowing leaf sample in a eucalyptus yellowing forest area, a healthy leaf sample which does not show diseases and a healthy leaf sample in a healthy eucalyptus forest area, wherein the yellowing leaf sample is collected in the step (1);
(3) Spectral data processing: preprocessing and normalizing the spectrum data acquired in the step (2);
(4) Constructing an OPLS-DA discriminant analysis model: constructing an OPLS-DA discriminant analysis model by adopting an OPLS method on the spectral data processed in the step (3), and evaluating the constructed OPLS-DA discriminant analysis model;
(5) Independent sample validation: and (3) randomly selecting spectral data of the blade sample to be tested, inputting the spectral data of the blade sample to be tested into the OPLS-DA discriminant analysis model constructed in the step (4) for verification, and rapidly judging the yellowing disease condition of the blade sample to be tested.
In the step (4), the spectral data after the processing is preferably subjected to the construction of an OPLS-DA discriminant analysis model by Simca software.
As a preferred aspect, in the step (2), the specific step of collecting the spectrum data includes: and (3) carrying out spectrum data acquisition on the acquired yellow leaf samples in the eucalyptus yellow forest region, the healthy leaf samples which do not show diseases and the healthy leaf samples in the healthy eucalyptus forest region by adopting an ASD (automatic service digital) field spec4 ground object reflection hyperspectral meter, wherein each leaf sample acquires 10 times of spectrum data, taking an average value, and calibrating the spectrometer 1 time by adopting a standard white board for every 5 leaf samples.
As a preferred embodiment, the basic parameters of the spectrometer are: the spectrum band range is 350-2500 nm, the light source is a 50W halogen lamp, the spectrum resolution is 1nm, the scanning time is 100ms, the wavelength precision is 0.5nm, and the wavelength repeatability is 0.1nm.
As a preferred embodiment, in the step (3), the specific step of preprocessing the spectrum data includes: the ViewSpecPro6.0 software is adopted to preprocess the spectrum data, the data of which the two ends of the spectrum measuring range are smaller than 400nm and larger than 2400nm are removed, and the original spectrum reflectivity is subjected to logarithmic transformation, so that the influence of noise is reduced.
As a preference of the technical scheme, in the original spectral reflectance, spectral characteristic wave bands of the yellowing blade and the healthy blade which does not show diseases are in the ranges of 800-1260 nm, 1400-1720 nm and 2000-2400 nm.
As a preferred aspect, in the logarithmic converted spectral reflectance, spectral characteristic band positions of the yellowing leaf and healthy leaf which does not exhibit disease are 550nm, 640nm, 1508nm, 2000nm and 2280 nm.
As a preferred embodiment, in the step (3), the normalization formula used for the normalization processing of the spectrum data is:
linear function normalized X' = (X-X) min )/(X max -X min ) Wherein X is original data, X max 、X min Respectively the maximum and minimum of the original data set.
As a preferred technical scheme, in the step (1), the specific method for collecting the disease sample is as follows: and randomly laying 20 x 20m standard places in a eucalyptus yellowing forest area, respectively collecting fresh leaf samples from the top end to the bottom of each living standing tree, collecting all living standing tree yellowing leaves in the standard places, simultaneously collecting healthy leaves which do not show diseases in the yellowing forest and healthy eucalyptus artificial Lin She with consistent standing conditions of the same forest Ban Na as a control group, and placing the samples into a 4 ℃ ice box for storage after the sample collection is completed for indoor spectral data information collection.
Compared with the prior art, the invention has the advantages that:
1. the VNIR technology is combined with the OPLS-DA discriminant analysis method, so that the disease condition of the eucalyptus yellow can be predicted efficiently, accurately and objectively, the problem that the eucalyptus yellow is difficult to identify accurately due to the fact that the eucalyptus yellow has the sudden property, the concealment property and the randomness is solved, the labor cost and the time cost are saved greatly, and the error caused by the subjectivity of the artificial investigation is avoided. According to the method, when the eucalyptus leaves do not show the characteristic of the yellowing disease, the disease condition of the eucalyptus yellowing disease can be predicted efficiently, accurately and objectively, so that corresponding measures are taken in time to prevent the occurrence of the eucalyptus yellowing disease, and the method has great significance in guaranteeing the high-quality development of artificial forests and downstream industries thereof.
2. The invention collects the spectrum of the plant leaves, and does not need to damage the leaves; because eucalyptus yellowing disease is higher in concealment, compared with PLS-DA discriminant analysis for judging whether eucalyptus leaves are yellowing or not, the model accuracy is lower, so that the method adopts an OPLS-DA discriminant analysis method, and quadrature verification is added into a traditional PLS-DA model, so that the discrimination accuracy and the model stability are remarkably improved, and the effect is good. The method solves the problems of destructive test samples, manual investigation, visual identification and discrimination of eucalyptus yellowing diseases, low model fitting precision and the like in the existing model, and has the advantages of rapidness, accuracy, objectivity, effectiveness, greenness, no damage and the like.
Drawings
FIG. 1 is a graph of a portion of eucalyptus yellow leaf, yellow forest healthy leaf, and healthy forest healthy leaf samples collected in example 1. In fig. 1, the left graph is a yellow leaf map, the middle graph is a yellow forest health leaf map, and the right graph is a healthy leaf map of a healthy forest region.
FIG. 2 is a plot of raw spectral reflectance versus log transformation for leaf samples of different disease levels in example 1. In fig. 2, a is an original spectral reflectance map, and B is a logarithmic transformation map.
FIG. 3 is a plot of PLS-DA and OPLS-DA spectra of leaf samples of different disease levels in example 1. In FIG. 3, A is an R-PLS-DA scattergram, B is a Log-PLS-DA scattergram, C is an R-OPLS-DA scattergram, and D is a Log-OPLS-DA scattergram.
FIG. 4 is a Log-OPLS-DA discriminant analysis lattice plot of the yellow leaf blade of example 1.
Detailed Description
The present invention is further illustrated by the following examples, which are only intended to illustrate the present invention and not to limit the scope of the present invention.
Example 1
The embodiment provides a method for predicting eucalyptus iron deficiency yellow based on VNIR and OPLS-DA.
(one) experimentally profile:
the test is located in Guangxi national yellow crown forest farm (109 DEG 43 '46' in east longitude and 24 DEG 37 '25' in North latitude) and the farm is located in deer village county, the forest farm belongs to medium and subtropical climate, the average temperature of the forest farm is 19 ℃ in the same season as warm, rainy and hot, the average annual rainfall is 1750mm, the average annual evaporation is 1426mm, and the relative height difference of the low mountain landform is 200-400 m. The woodland soil is mainly red soil developed from sand shale, and the texture is light soil.
The test method (II) comprises the following steps:
(1) Blade sample collection: in the yellowing forest zone, 3 standard parts of 20 x 20m with consistent site conditions are randomly set up, all the live stumps in the standard part are collected to obtain yellow leaves (chlorsis, C), 10 fresh leaf samples are collected from the top to the bottom of each live stump, and meanwhile, healthy leaves (chlorsis-Normal, C-N) which do not show diseases in the yellowing forest and healthy eucalyptus artificial forest leaves (Normal, N) with consistent site conditions of the same forest Ban Na are collected as a control group, and 30 typical leaf samples are taken in total. A diagram of a part of collected eucalyptus yellow leaf, yellow forest healthy leaf and healthy leaf sample of a healthy forest zone is shown in fig. 1. And after the sample is collected, the sample is put into an ice box at the temperature of 4 ℃ for storage and is used for collecting indoor hyperspectral information.
(2) And (3) spectrum data acquisition: spectral data were collected using an ASD FieldSpec4 ground object reflectance hyperspectral meter on samples of the yellow leaves collected in eucalyptus yellow forest, healthy leaves showing no disease, and healthy leaves in healthy eucalyptus forest. The spectrum band is between visible-near infrared band (350 nm-2500 nm), the resolution is 1nm, the angle of view of the probe is 15 degrees, the light source adopts a 50W halogen lamp matched with the instrument, the scanning time is 100ms, the wavelength precision is 0.5nm, and the wavelength repeatability is 0.1nm. In order to ensure the accuracy of the spectrum data, 10 times of spectrum data are collected for each blade sample, and an arithmetic average value is taken as the spectrum data of a final sample.
(3) Spectral data processing: spectral data were preprocessed using viewspecpro 6.0. Because the self performance of the instrument causes spectrum data noise, the two ends of the spectrum instrument range (less than 400nm and more than 2400 nm) are removed, and the original spectrum reflectivity is subjected to logarithmic transformation, so that the influence of noise is reduced. To improve the running speed of the modelThe rate and the model accuracy rate are normalized, and all spectrum data are normalized by using a formula of linear function normalization X' = (X-X) min )/(X max -X min ) Wherein X is original data, X max 、X min Respectively the maximum and minimum of the original data set.
(4) Constructing a PLS-DA and OPLS-DA discriminant analysis model: and respectively constructing PLS-DA and OPLS-DA discriminant analysis models on the processed spectrum data by using Simca software, and evaluating the constructed PLS-DA and OPLS-DA discriminant analysis models.
(III) results and analysis:
(1) Spectral reflectance characterization: the average values of the original spectral reflectances (R) and Log pretreatment (Log) of the yellowing leaf (C), the non-diseased leaf (C-N) and the normal leaf (N) are shown in FIG. 2, wherein FIG. 2 is a graph of the original spectral reflectances and Log transformation of leaf samples with different disease degrees, in FIG. 2, A is a graph of the original spectral reflectances, and B is a graph of the Log transformation. The difference is observed, the spectral reflectance curves of the blades with different disease degrees show the same trend, but the reflectance is obviously different, the original spectral reflectance of the blade affected by the disease is obviously higher than that of a normal blade, and the change rule after logarithmic transformation is opposite.
As can be seen from the graph A of FIG. 2, the absorption peaks of the original spectral reflectance (R) are mainly 5, and the reflectance of the yellowing leaf (C) is obviously higher than that of the non-onset leaf (C-N) and the normal leaf (N) in the visible light green light (550 nm), near infrared 1180nm, 1288nm, 1630nm and 2200nm regions respectively, and the reflectance of the normal leaf (N) is obviously lower than that of the affected leaf in the near infrared wave ranges of 800-1260 nm, 1400-1720 nm and 2000-2400 nm.
As can be seen from the graph B of fig. 2, after logarithmic transformation, the differences of spectral reflectivities of the yellowing leaf (C), the non-diseased leaf (C-N) and the normal leaf (N) are reduced, but the peak shapes are more sharp, the difference peaks mainly appear at 640nm, 1508nm and 2000nm, the trough appears at 550nm and 2280nm, and these difference peaks can be used as spectral characteristic wave bands for identifying eucalyptus leaf yellowing.
(2) Blade discriminant analysis of different disease degrees: 70% of 30 samples are randomly selected as modeling samples and 30% of the 30 samples are used as verification samples, the original spectral reflectance and the logarithm preprocessed reflectance of the samples are input into a model, different disease degree blades are subjected to discriminant analysis by using PLS-DA and OPLS-DA, as shown in FIG. 3, and FIG. 3 is a PLS-DA and OPLS-DA scatter diagram of the spectral indexes of the blade samples with different disease degrees. In FIG. 3, A is an R-PLS-DA scattergram, B is a Log-PLS-DA scattergram, C is an R-OPLS-DA scattergram, and D is a Log-OPLS-DA scattergram. From the same index, the discriminant analysis clustering effect based on the OPLS-DA method is better than that of PLS-DA, and the sample discrete degree is lower.
The accumulated interpretation rate of the PLS-DA on the X value of the original spectrum and the logarithmic spectrum is 91.4 percent and 85.2 percent respectively, but the model R is distinguished 2 Only 0.607 and 0.665 show that PLS-DA has a lower effect than OPLS-DA in extracting the band with obvious correlation to the yellowing disease, and meanwhile R-PLS-DA cannot accurately judge samples of the yellowing blades 2 and 3 in the original spectrum and the logarithmic spectrum.
The OPLS-DA can better distinguish samples with different morbidity degrees, and the accumulated interpretation rate of the X value of the OPLS-DA on the original spectrum and the log spectrum is 56.14 percent and 50.5 percent respectively, so that the analysis model R is judged 2 Respectively reaching 0.778 and 0.801, the method improves 28.17 percent and 20.45 percent of PLS-DA, and has more efficient and accurate pre-judging effect on the yellowing disease relative to PLS-DA.
(IV) independent sample verification
And screening an optimal model Log-OPLS-DA from the OPLS-DA discriminant analysis model to verify, wherein the verification samples are yellowing blades, the number of the verification set samples is 8, and the verification set samples account for 27.5% of the total number of the samples and meet the model verification requirement. Fig. 4 is a Log-OPLS-DA discriminant analysis lattice diagram of the yellowing blade, and the lattice diagram shows that in the Log-OPLS-DA model, the accumulated variance of the principal components reaches 98.7%, the number of discriminant error samples is 2 (22, 23), the number of correct samples is 6 (24, 25, 26, 27, 28, 29), the recognition rate reaches more than 75%, the model precision is at a higher level, so that the OPLS-DA discriminant analysis model established by the invention can accurately recognize the disease condition of eucalyptus yellowing disease, can effectively meet the early identification and recognition of the disease condition of eucalyptus yellowing disease, and has important practical significance for efficiently, accurately and objectively predicting the disease condition of eucalyptus yellowing disease.
Claims (5)
1. The method for predicting the eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA is characterized by comprising the following steps:
(1) Blade sample collection: collecting a yellowing leaf sample and a healthy leaf sample which does not show diseases in a eucalyptus yellowing forest zone, and collecting a healthy leaf sample in a healthy eucalyptus forest zone with consistent standing conditions of the same forest Ban Na;
(2) And (3) spectrum data acquisition: collecting spectral data of a yellowing leaf sample in a eucalyptus yellowing forest area, a healthy leaf sample which does not show diseases and a healthy leaf sample in a healthy eucalyptus forest area, wherein the yellowing leaf sample is collected in the step (1); the specific steps of spectrum data acquisition include: carrying out spectrum data acquisition on the acquired yellow leaf samples in the eucalyptus yellow forest region, the healthy leaf samples which do not show diseases and the healthy leaf samples in the healthy eucalyptus forest region by adopting an ASD (automatic sequence analysis) field spec4 ground object reflection hyperspectral meter, acquiring 10 times of spectrum data for each leaf sample, taking an average value, and calibrating the spectrometer for 1 time by adopting a standard white board for every 5 leaf samples to be detected;
(3) Spectral data processing: preprocessing and normalizing the spectrum data acquired in the step (2); the spectrum data preprocessing specifically comprises the following steps: preprocessing spectral data by using ViewSpecPro6.0 software, removing data with the range both ends of a spectrum instrument being less than 400nm and more than 2400nm, and carrying out logarithmic transformation on the original spectral reflectivity at the same time so as to reduce the influence of noise; in the original spectral reflectance, spectral characteristic wave bands of the yellowing blades and the healthy blades which do not show diseases are in the ranges of 800-1260 nm, 1400-1720 nm and 2000-2400 nm; in the logarithmic transformation of the spectral reflectivities, spectral characteristic band positions of the yellowing blades and the healthy blades which do not show diseases are 550nm, 640nm, 1508nm, 2000nm and 2280 nm;
(4) Constructing an OPLS-DA discriminant analysis model: constructing an OPLS-DA discriminant analysis model by adopting an OPLS method on the spectral data processed in the step (3), and evaluating the constructed OPLS-DA discriminant analysis model;
(5) Independent sample validation: and (3) randomly selecting spectral data of the blade sample to be tested, inputting the spectral data of the blade sample to be tested into the OPLS-DA discriminant analysis model constructed in the step (4) for verification, and rapidly judging the yellowing disease condition of the blade sample to be tested.
2. The method for predicting eucalyptus iron deficiency yellow based on VNIR and OPLS-DA according to claim 1, wherein in step (4), the processed spectral data is subjected to OPLS-DA discriminant analysis model construction by Simca software.
3. The method for predicting eucalyptus iron deficiency yellow based on VNIR and OPLS-DA according to claim 1, wherein in step (2), the basic parameters of the spectrometer are: the spectrum band range is 350-2500 nm, the light source is a 50W halogen lamp, the spectrum resolution is 1nm, the scanning time is 100ms, the wavelength precision is 0.5nm, and the wavelength repeatability is 0.1nm.
4. The method for predicting eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA according to claim 1, wherein in step (3), the normalization formula used for the normalization processing of the spectrum data is:
linear function normalized X' = (X-X) min ) / (X max - X min ) Wherein X is original data, X max 、X min Respectively the maximum and minimum of the original data set.
5. The method for predicting eucalyptus iron deficiency yellow fever based on VNIR and OPLS-DA according to claim 1, wherein in step (1), the specific method for collecting the disease sample is as follows: and randomly laying 20 x 20m standard places in a eucalyptus yellowing forest area, respectively collecting fresh leaf samples from the top end to the bottom of each living standing tree, collecting all living standing tree yellowing leaves in the standard places, simultaneously collecting healthy leaves which do not show diseases in the yellowing forest and healthy eucalyptus artificial Lin She with consistent standing conditions of the same forest Ban Na as a control group, and placing the samples into a 4 ℃ ice box for storage after the sample collection is completed for indoor spectral data information collection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210521422.8A CN114946485B (en) | 2022-05-13 | 2022-05-13 | Method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210521422.8A CN114946485B (en) | 2022-05-13 | 2022-05-13 | Method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114946485A CN114946485A (en) | 2022-08-30 |
| CN114946485B true CN114946485B (en) | 2023-07-07 |
Family
ID=82984106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210521422.8A Active CN114946485B (en) | 2022-05-13 | 2022-05-13 | Method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114946485B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103134758A (en) * | 2013-01-29 | 2013-06-05 | 华南农业大学 | Rice leaf blast disease resistance identification grading method based on multi-scale hyperspectral image processing |
| CN106018332A (en) * | 2016-07-21 | 2016-10-12 | 华南农业大学 | Near-infrared-spectrum citrus yellow shoot disease field detection method |
| CN110007017A (en) * | 2019-03-04 | 2019-07-12 | 山东省中医药研究院 | It is a kind of to identify mulberry leaf whether through white detection method |
| CN110363125A (en) * | 2019-07-04 | 2019-10-22 | 浙江大学山东工业技术研究院 | Using the method for Model Transfer identification different cultivars Citrus Huanglongbing pathogen |
| WO2020213672A1 (en) * | 2019-04-16 | 2020-10-22 | 花王株式会社 | Method of predicting soybean yield |
| CN112179870A (en) * | 2020-10-15 | 2021-01-05 | 湖北中烟工业有限责任公司 | Cigarette classification recognition model construction method based on near infrared spectrum and OPLS-DA |
| CN114418027A (en) * | 2022-01-26 | 2022-04-29 | 浙大城市学院 | Hyperspectral image characteristic waveband selection method based on waveband attention mechanism |
-
2022
- 2022-05-13 CN CN202210521422.8A patent/CN114946485B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103134758A (en) * | 2013-01-29 | 2013-06-05 | 华南农业大学 | Rice leaf blast disease resistance identification grading method based on multi-scale hyperspectral image processing |
| CN106018332A (en) * | 2016-07-21 | 2016-10-12 | 华南农业大学 | Near-infrared-spectrum citrus yellow shoot disease field detection method |
| CN110007017A (en) * | 2019-03-04 | 2019-07-12 | 山东省中医药研究院 | It is a kind of to identify mulberry leaf whether through white detection method |
| WO2020213672A1 (en) * | 2019-04-16 | 2020-10-22 | 花王株式会社 | Method of predicting soybean yield |
| CN110363125A (en) * | 2019-07-04 | 2019-10-22 | 浙江大学山东工业技术研究院 | Using the method for Model Transfer identification different cultivars Citrus Huanglongbing pathogen |
| CN112179870A (en) * | 2020-10-15 | 2021-01-05 | 湖北中烟工业有限责任公司 | Cigarette classification recognition model construction method based on near infrared spectrum and OPLS-DA |
| CN114418027A (en) * | 2022-01-26 | 2022-04-29 | 浙大城市学院 | Hyperspectral image characteristic waveband selection method based on waveband attention mechanism |
Non-Patent Citations (3)
| Title |
|---|
| 双包络去除和OPLS的土壤有机碳含量反演;丛麟骁;黄旻;刘祥磊;齐云松;;光谱学与光谱分析(第03期);全文 * |
| 基于GC-MS技术的不同产地稻米代谢组学研究;冯玉超;富天昕;李雪;王长远;张东杰;;中国生物制品学杂志(第08期);全文 * |
| 桉树杂交种与其亲本的近红外光谱判别;卢万鸿;李鹏;王楚彪;林彦;罗建中;;光谱学与光谱分析(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114946485A (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110082300B (en) | Method for establishing winter wheat canopy water content monitoring model based on spectral parameters | |
| Loh et al. | Using the SPAD 502 meter to assess chlorophyll and nitrogen content of benjamin fig and cottonwood leaves | |
| CN102426153B (en) | A kind of Wheat plant moisture monitoring method based on canopy high spectral index | |
| CN107796764B (en) | Method for constructing wheat leaf area index estimation model based on three-band vegetation index | |
| CN110376167A (en) | Rice leaf nitrogen content monitoring method based on unmanned plane EO-1 hyperion | |
| CN110160967A (en) | A kind of total nitrogen content evaluation method of crop canopies blade | |
| CN109187398A (en) | A kind of EO-1 hyperion measuring method of wheat plant nitrogen content | |
| CN101382488A (en) | Method for detecting nitrogen content in fresh tea by visible light-near infrared diffuse reflection spectrum technology | |
| CN111062251A (en) | Monitoring method of farmland cotton aphid pest grade model based on unmanned aerial vehicle imaging | |
| CN110069895B (en) | Method for establishing winter wheat nitrogen content full-growth period spectrum monitoring model | |
| CN110082309B (en) | Method for establishing SPAD value comprehensive spectrum monitoring model of winter wheat canopy | |
| CN110987830A (en) | Model, method and application for rapidly determining chlorophyll content of plant canopy leaves | |
| CN110779875B (en) | Method for detecting moisture content of winter wheat ear based on hyperspectral technology | |
| Jin et al. | Estimation of wheat nitrogen status under drip irrigation with canopy spectral indices | |
| CN114946485B (en) | Method for prejudging eucalyptus iron deficiency yellow disease based on VNIR and OPLS-DA | |
| Karakoc et al. | Ratio-based vegetation indices for biomass estimation depending on grassland characteristics | |
| CN107796779A (en) | The near infrared spectrum diagnostic method of rubber tree LTN content | |
| Starks et al. | Estimation of nitrogen concentration and in vitro dry matter digestibility of herbage of warm‐season grass pastures from canopy hyperspectral reflectance measurements | |
| Subedi et al. | Determination of optimum maturity stages of mangoes using fruit spectral signatures | |
| Li et al. | Chlorophyll assessment and sensitive wavelength exploration for tea (Camellia sinensis) based on reflectance spectral characteristics | |
| CN115184322A (en) | Rice leaf water content monitoring method | |
| Jia et al. | Inversion of chlorophyll fluorescence parameters on vegetation indices at leaf scale | |
| Cushnahan et al. | Identifying grass species using hyperspectral sensing | |
| Abdel-Rahman et al. | Imaging spectroscopy for estimating sugarcane leaf nitrogen concentration | |
| CN113252592B (en) | Subtilis parameter calibration method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20220830 Assignee: Wuzhou Ruifeng Agricultural Technology Co.,Ltd. Assignor: GUANGXI ZHUANG AUTONOMOUS REGION FORESTRY Research Institute Contract record no.: X2023980045120 Denomination of invention: A Method for Predicting Iron Deficiency Yellowing Disease in Eucalyptus Based on VNIR and OPLS-DA Granted publication date: 20230707 License type: Common License Record date: 20231103 |
|
| EE01 | Entry into force of recordation of patent licensing contract |