CN110414729A - The potential maximum photosynthetic capacity prediction technique of plant based on characteristic wavelength - Google Patents
The potential maximum photosynthetic capacity prediction technique of plant based on characteristic wavelength Download PDFInfo
- Publication number
- CN110414729A CN110414729A CN201910655457.9A CN201910655457A CN110414729A CN 110414729 A CN110414729 A CN 110414729A CN 201910655457 A CN201910655457 A CN 201910655457A CN 110414729 A CN110414729 A CN 110414729A
- Authority
- CN
- China
- Prior art keywords
- plant
- photosynthetic capacity
- characteristic wavelength
- value
- potential maximum
- 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
- 230000000243 photosynthetic effect Effects 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000013528 artificial neural network Methods 0.000 claims abstract description 13
- 238000012360 testing method Methods 0.000 claims abstract description 13
- 230000002068 genetic effect Effects 0.000 claims abstract description 12
- 230000008569 process Effects 0.000 claims abstract description 10
- 238000005457 optimization Methods 0.000 claims abstract description 9
- 238000000985 reflectance spectrum Methods 0.000 claims abstract description 9
- 238000000342 Monte Carlo simulation Methods 0.000 claims abstract description 5
- 230000005856 abnormality Effects 0.000 claims abstract description 4
- 238000002474 experimental method Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 35
- 239000010410 layer Substances 0.000 claims description 34
- 238000001228 spectrum Methods 0.000 claims description 22
- 210000002569 neuron Anatomy 0.000 claims description 18
- 238000002310 reflectometry Methods 0.000 claims description 14
- 210000000349 chromosome Anatomy 0.000 claims description 9
- 230000035772 mutation Effects 0.000 claims description 9
- 108090000623 proteins and genes Proteins 0.000 claims description 9
- 230000006978 adaptation Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000009792 diffusion process Methods 0.000 claims description 5
- 239000011229 interlayer Substances 0.000 claims description 3
- 238000010186 staining Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 241000196324 Embryophyta Species 0.000 description 31
- 239000000523 sample Substances 0.000 description 25
- 229930002875 chlorophyll Natural products 0.000 description 15
- 235000019804 chlorophyll Nutrition 0.000 description 15
- 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 description 15
- 244000061458 Solanum melongena Species 0.000 description 13
- 235000002597 Solanum melongena Nutrition 0.000 description 13
- 238000001514 detection method Methods 0.000 description 9
- 230000003595 spectral effect Effects 0.000 description 7
- 230000004300 dark adaptation Effects 0.000 description 6
- 230000029553 photosynthesis Effects 0.000 description 5
- 238000010672 photosynthesis Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 238000010521 absorption reaction Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000001537 neural effect Effects 0.000 description 3
- 230000037039 plant physiology Effects 0.000 description 3
- 238000004611 spectroscopical analysis Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 238000012614 Monte-Carlo sampling Methods 0.000 description 2
- 241000209140 Triticum Species 0.000 description 2
- 235000021307 Triticum Nutrition 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 229910002092 carbon dioxide Inorganic materials 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000012010 growth Effects 0.000 description 2
- 229910052736 halogen Inorganic materials 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000008635 plant growth Effects 0.000 description 2
- 150000003839 salts Chemical class 0.000 description 2
- 229910052721 tungsten Inorganic materials 0.000 description 2
- 239000010937 tungsten Substances 0.000 description 2
- -1 tungsten halogen Chemical class 0.000 description 2
- 241001116389 Aloe Species 0.000 description 1
- 208000005156 Dehydration Diseases 0.000 description 1
- 235000007688 Lycopersicon esculentum Nutrition 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 240000003768 Solanum lycopersicum Species 0.000 description 1
- 244000061456 Solanum tuberosum Species 0.000 description 1
- 235000002595 Solanum tuberosum Nutrition 0.000 description 1
- 231100000768 Toxicity label Toxicity 0.000 description 1
- 235000009754 Vitis X bourquina Nutrition 0.000 description 1
- 235000012333 Vitis X labruscana Nutrition 0.000 description 1
- 240000006365 Vitis vinifera Species 0.000 description 1
- 235000014787 Vitis vinifera Nutrition 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 235000011399 aloe vera Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006353 environmental stress Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000007850 fluorescent dye Substances 0.000 description 1
- 238000010353 genetic engineering Methods 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 230000035882 stress Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
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
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6486—Measuring fluorescence of biological material, e.g. DNA, RNA, cells
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Mining
-
- 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
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N2021/6495—Miscellaneous methods
Abstract
Plant potential photosynthetic capacity prediction technique based on characteristic wavelength, is arranged mutually synthermal, humidity, CO under six lighting gradients2The culture environment of concentration generates difference to the plant under different lighting processes, randomly selects plant leaf as experiment sample, the dark fluorescence parameter of plant leaf blade and Visible-to-Near InfaRed reflectance spectrum is measured respectively, as sample data；Using Monte Carlo method rejecting abnormalities sample, by 4:1 random division training set and test set；Characteristic wavelength is extracted using correlation coefficient process and successive projection method；It is input with the corresponding reflectance spectrum of characteristic wavelength, the potential maximum photosynthetic capacity of plant is output, establishes the potential maximum photosynthetic capacity prediction model of plant using the radial basis function neural network of genetic algorithm optimization；Using the model, maximum photosynthetic capacity potential to plant is predicted, the present invention is the quick of the potential maximum photosynthetic capacity of plant, and lossless, low cost monitoring provides theoretical foundation.
Description
Technical field
The invention belongs to reading intelligent agriculture technical fields, are related to the research of plant maximum potential photosynthetic capacity, in particular to one
The potential maximum photosynthetic capacity prediction technique of plant of the kind based on characteristic wavelength.
Background technique
Photosynthesis is the approach of plant material accumulation, directly affects plant growth and fruit yield.Plant Light cooperation
With not only being influenced by plant growth environment, also influenced by its physiological status.The chlorophyll of plant leaf blade under different physiological status
Content, efficiency of light energy utilization etc. directly affect its photosynthetic rate there are notable difference.The luminous energy of plant absorption is disappeared in the form of three kinds
Consumption: photosynthesis, chlorophyll fluorescence and heat.Therefore, chlorophyll fluorescence can reflect photosynthesis of plant.Currently, chlorophyll fluorescence
Technology be widely used in photosynthesis of plant energy absorption, transmitting, dissipation, distribution non-destructive testing (Zhang Shouren 1999).
Chlorophyll fluorescence parameters Fv/Fm is using frequency for potential maximum photosynthetic capacity and PSII reaction center maximum photosynthetic efficiency
One of highest parameter of rate is chiefly used in characterizing influence of the environment-stress for plant photosynthesis, and Rong Zhou etc. (2015) is used
The early detection of Fv/Fm progress Tomato Heat Tolerance；Hazrati and Saeid's (2016) research shows that water stress and illumination the side of body
The Fv/Fm for compeling will cause aloe is reduced；Qin Hongyan etc. (2013) carries out various concentration salt treatment to grape seedling, finds its blade
Fv/Fm aggravates with salt stress degree and is reduced.Therefore, realize the real non-destructive detection of Fv/Fm for characterizing plant physiology state
It is of great significance.However traditional Fv/Fm detection mode need to carry out dark adaptation to blade to be measured, cannot achieve real-time measurement,
Expensive additionally, due to chlorophyll fluorescence instrument, Fv/Fm testing cost is high, is not able to satisfy actually detected demand.
Spectrum detection technique has quick, lossless, advantage of lower cost advantage, in recent years in plant physiology status monitoring
Aspect is quickly grown.Relationship of the domestic and foreign scholars to plant chlorophyll fluorescence parameter in reflectance spectrum has made intensive studies, Zhu
Gorgeous equal (2007) under different nitrogen amount applieds, the chlorophyll fluorescence of the various position leaves wheat leaf blade of different cultivars and growthdevelopmental stage is joined
Several and hyper spectral reflectance is analyzed, at the top of discovery wheat the fluorescence parameter of two blades with its difference vegetation index DVI (750,
550) correlation highest, wherein Fv/Fm and the exponential dependence are up to 0.68；Ibarakip etc. (2011) built a set of PRI at
As system, PRI index is calculated after the reflective light intensity of 530nm and 570nm is acquired to potato leaf, finds blade
After dark adaptation, there are linear relationships with Fv/Fm for the index under low light condition.Hao Zhang etc. (2011) is using principal component point
The method of analysis extracts the characteristic wave bands of rice leaf chlorophyll fluorescence parameters, selects multiple vegetation indexs and builds with fluorescence parameter
Mould, the results showed that the fitting of the normalized differential vegetation index NDSI ((R680-R935)/(R680+R935)) of 680nm and 935nm is imitated
Fruit is best, the coefficient of determination R with the regression model of Fv/Fm2It is 0.669, root-mean-square error RMSE is 0.03.The above research
Demonstrate the feasibility of reflection spectrum detection Fv/Fm.However current research mostly uses vegetation index to model, it includes reflection
Spectral information amount is limited, and modeling pattern is simple, and model accuracy is restricted, can not Accurate Prediction Fv/Fm.
Summary of the invention
In order to overcome the disadvantages of the above prior art, the purpose of the present invention is to provide a kind of plants based on characteristic wavelength
Potential maximum photosynthetic capacity prediction technique, measures plant visible light-near infrared spectrum data and chlorophyll fluorescence parameters Fv/Fm
(i.e. potential maximum light and ability), using multiple spectrum data preprocessing method, with correlation coefficient analysis combination successive projection
The mode of method screens characteristic spectrum, and it is pre- to establish the plant seedlings Fv/Fm based on characteristic spectrum using radial basis function neural network
Model is surveyed, and model parameter is optimized using genetic algorithm, the final real-time Accurate Prediction for realizing Fv/Fm.
To achieve the goals above, the technical solution adopted by the present invention is that:
The potential maximum photosynthetic capacity prediction technique of plant based on characteristic wavelength characterized by comprising
Step 1, mutually synthermal, humidity, CO under multiple and different lighting gradients are set2The culture environment of concentration, to not share the same light
Difference is generated according to the plant under processing, plant leaf blade is randomly selected as experiment sample, measures the potential maximum of plant leaf blade respectively
Light and ability and Visible-to-Near InfaRed reflectance spectrum, as sample data；
Step 2, random division training set and test set；
Step 3, characteristic wavelength is extracted；
It step 4, is input, the potential maximum photosynthetic capacity of plant with the corresponding reflectivity of training set sample spectrum characteristic wavelength
For output, the potential maximum photosynthetic capacity prediction model of plant is established；
Step 5, using the model, maximum photosynthetic capacity potential to plant is predicted.
In the step 1, six lighting gradients, respectively 50,90,140,220,280,340 μm of olm are set-2s-1。
In the step 2, using Monte Carlo method rejecting abnormalities sample, by 4:1 random division training set and test set.
In the step 3, unrelated wavelength is first removed using correlation coefficient process, is then selected wherein most using successive projection method
Representative wavelength calculates each wavelength reflection of spectrum with latent as characteristic wavelength, using Pearson correlation coefficient calculation formula
Under the related coefficient of maximum photosynthetic capacity, calculation formula:
Wherein, riIt is i-th of wave band and the related coefficient of Fv/Fm, n is number of samples, xijIt is i-th of j-th of sample
The reflectivity of wave band, yjFor the Fv/Fm value of j-th of sample；
The unrelated wavelength refers to that related coefficient absolute value is less than the wavelength of setting value, will be removed using successive projection method unrelated
The remaining wavelength of wavelength further compresses, and extracts characteristic wavelength.
In the step 4, it is photosynthetic that the potential maximum of plant is established using the radial basis function neural network of genetic algorithm optimization
Ability prediction model.
The establishment step of the potential maximum photosynthetic capacity prediction model of the plant is as follows:
1), by initialization of population, stochastic parameter is set and generates initial population, each individual in population corresponds to radial base
One spread value of function generates individual chromosome using binary coding；
2) individual adaptation degree in initial population, is calculated, fitness function is radial basis function neural network, and fitness value is
The average error value of potential maximum photosynthetic capacity predicted value and measured value, that is, use the corresponding spread value of each individual of initial population
Potential maximum photosynthetic capacity prediction model is constructed, and calculates the average error value of each model as its fitness value.
3), selection operation is eliminated and is selected, its lower quilt of individual adaptation degree by stake turntable progress initial population individual
Choose probability bigger；
4), crossover operation selects two individuals from initial population, randomly chooses some chromosome locations and swap,
To generate new individual；
5), mutation operation, an optional individual from initial population, a little making a variation to generate in selective staining body
More excellent individual；
6), by selection, intersection, variation, heredity, population of new generation is obtained, the fitness of each generation population at individual is calculated, note
Record minimum fitness and its corresponding spread value；
7) step 3) -6, is repeated) after iterative evolution is multiple, fitness converges to minimum, exits circulation, obtains minimum fit
The corresponding spread value of response, i.e., optimal spread value；
8), the radial base diffusion velocity using optimal spread value as radial basis function neural network, establish it is potential most
Big photosynthetic capacity prediction model.
In the step 1), according to genetic algorithm optimization radial basis function neural network spread parameter, spread is set
The index range of value is to be translated into individual by binary coding for [1,30], and number of individuals is set as 30, forms one
Population, i.e. population scale are 30, and Evolution of Population algebra is set as 50 generations, crossover probability 0.5, mutation probability 0.03, by this
Stochastic parameter generates initial population.
Formula is encoded in the step 1) are as follows:
Decoding formula are as follows:
In formula, b is the binary string after coding；M takes binary string character number by chromosome；A is required coding ten
System number；aminFor space encoder minimum decimal number；amaxFor space encoder maximum decimal number.
In the step 2), the learning procedure of radial basis function is, it is first determined in hidden layer neuron radial function
The heart, if training set sample input matrix is P, output matrix T, Q are training set sample number, then Q hidden layer neuron is corresponding
Radial basis function center be C=P ', determine hidden layer and output interlayer weight and threshold value be b1=[b11,b12,...,b1Q] ',
Wherein,After the radial basis function center of hidden layer neuron and threshold value determine, hidden layer mind
Output through member are as follows:
Di=exp (- | | C-pi||2bi), i=1,2,, Q
Wherein, pi=[pi1,pi2,...,piM] ', is i-th of training sample vector, and is denoted as D=[d1,d2,...,dQ],
The output of each hidden neuron is always denoted as vector D, if the connection weight W of hidden layer and output layer is
Wherein, wijThe connection weight between j-th of hidden layer neuron and i-th of output layer neuron is indicated, if N number of output
The threshold value b of layer neuron2=[b21,b22,...,b2N] ', and satisfaction: [w, b2].[D；I]=T, wherein I=[1,1 ...,
1]1×Q。
In the step 3), the selected probability of each individual is Si,F in formulaiFor individual i in population
Fitness inverse, N indicate Population Size.
In the step 5), j-th of gene a of i-th of individual is chosenijIt makes a variation, mutation operation method is as follows:
In formula, amaxFor gene aijThe upper bound；aminFor gene aijLower bound；F (g)=o2(1-g/Gmax)2； O2It is one
Random number, g are current evolutionary generation；GmaxFor maximum evolution number；Random number of the O between [0,1].
Compared with prior art, the beneficial effects of the present invention are:
1) the potential maximum photosynthetic capacity prediction model of the plant based on characteristic wavelength is proposed, with plant reflection spectrum characteristic wave
The long potential maximum photosynthetic capacity Fv/Fm of reflectivity prediction.Compared to conventional method, it can measure without dark adaptation, reduce inspection
The time is surveyed, and cost is relatively low for detection device.
2) characteristic wavelength is extracted using related coefficient combination successive projection method, Effective selection goes out related to Fv/Fm and conllinear
The smallest characteristic wavelength of property.In this, as the input of model, the complexity of model is reduced, reduces model and calculates the time.
Compared to vegetation index, using more spectral wavelengths as mode input, hence it is evident that improve Fv/Fm precision of forecasting model.
3) compared to conventional linear or nonlinear regression, radial basis function neural network performance is more preferable.Using heredity
Algorithm finds optimal radial basic function diffusion velocity as modeling parameters, provides effective foundation for model parameter selection.With this
The Fv/Fm precision of forecasting model of foundation is high, and generalization ability is strong.The model provides for the detection of the potential maximum photosynthetic capacity of plant
New paragon, rapidity and low cost are more suitable for actually detected needs.
Detailed description of the invention
Fig. 1 is prediction technique flow chart of the present invention.
Fig. 2 is eggplant sample mean spectral schematic in the embodiment of the present invention.
Fig. 3 is related coefficient of each wavelength reflection of spectrum with Fv/Fm.
Fig. 4 is GA-RBF algorithm modeling process in the embodiment of the present invention.
Fig. 5 is that fitness and evolutionary generation change schematic diagram in the embodiment of the present invention.
Fig. 6 is predicted value and measured value contrast schematic diagram in the embodiment of the present invention.
Fig. 7 is the potential maximum photosynthetic capacity prediction model verifying schematic diagram of plant of the present invention.
Specific embodiment
The embodiment that the present invention will be described in detail with reference to the accompanying drawings and examples.
The present invention establishes a kind of potential photosynthetic energy of maximum of the plant based on characteristic wavelength using eggplant leaf as research object
Power prediction model measures the potential maximum photosynthetic capacity value of eggplant leaf and Visible-to-Near InfaRed reflected light under different lighting gradients
Spectrum is used as sample data, using Monte Carlo method rejecting abnormalities sample, divides training set and test set by 4:1, utilizes phase relation
Number method combination successive projection method extracts characteristic wavelength, is input with the corresponding reflectivity of training set sample spectrum characteristic wavelength, dives
It is output in maximum photosynthetic capacity, potential maximum photosynthetic capacity is established using the radial basis function neural network of genetic algorithm optimization
Prediction model verifies model accuracy and generalization ability using test set sample.
As shown in Figure 1, detailed process is as follows:
1 testing program
It is tested in November, 2018 in January, 2019 in Xibei Univ. of Agricultural & Forest Science & Technology's agricultural rural area portion agricultural Internet of Things emphasis
Laboratory carries out.It is purplish red long eggplant F1 seedling for experimental material.In November, 2018 is by healthy growth, the consistent eggplant seedling of growing way
It is placed in culture in 6 carbon dioxide growth cabinets, environment parameter setting in climate box are as follows: photoperiod day night is 14h/10h,
Temperature day night is 25 DEG C/16 DEG C, and relative air humidity day night is 60%/50%, CO2Concentration is 400 μm of olmol-1.Setting
6 intensity of illumination gradients (table 1), remaining environmental factor and cultivation management are consistent.After cultivating 15 days, in seedling stage to leaf morphology
The apparent plant of difference carries out secretly lower fluorescence parameter Fv/Fm and reflection spectrum measuring.Wherein dark fluorescence parameter Fv/Fm meaning is latent
In maximum light and ability.
Using fluorescence parameter under portable modulated chlorophyll fluorescence instrument (Mini-Pam II, WALZ, Germany) measurement secretly
Fv/Fm.Before measurement, blade is flattened, clamps blade to be measured using dark adaptation blade folder, the leaf of vein is avoided in the selection of clamping position
Medium film section position.Abundant dark adaptation after twenty minutes, by the probe entrance of fluorescent probe insertion dark adaptation blade folder, keep probe with
Blade surface is vertical.It opens blade and presss from both sides plectrum, be measured.Visible-to-Near InfaRed reflectance spectrum is measured in blade same position.Light
Modal data acquisition system includes that wave-length coverage is 350-1100nm, and resolution ratio is 1.5nm spectrometer (OFS-1100, Ocean
Optics, America), tungsten halogen lamp (HL-2000, Ocean Optics, America), integral ball-type blade folder
(SpectroClip-TR, Ocean Optics, America) and computer.Every time before measurement, tungsten halogen lamp is preheated 30 minutes,
To guarantee that the intensity of light source is uniform.To avoid this experimental precision of crop lunch break effects, only in 9:00-11:00 and 14:00-17:
It is tested in 00 two periods.Experiment obtains Fv/Fm data and corresponding 322 groups of spectroscopic data altogether.
1 CO of table2Growth cabinet intensity of illumination gradient
| Light levels | L1 | L2 | L3 | L4 | L5 | L6 |
| PPFD(μmol·m-2s-1) | 50 | 90 | 140 | 220 | 280 | 340 |
2 exceptional samples are rejected
Monte Carlo method be it is a kind of with Probability Statistics Theory be guidance very important numerical computation method, it use
Random number (or more common pseudo random number) solves many computational problems.By the method for certain " test ", with this event
The probability of this chance event of the Frequency Estimation of appearance, or certain numerical characteristics of this stochastic variable are obtained, and made
For the solution of problem.
Since spectrum head and the tail wave band noise is relatively low, thus choose 400~1000nm wave-length coverage in 1358 wave bands into
Row spectrum analysis.Using Monte Carlo sampling method (Monte-Carlo sampling method, MCS) to 332 groups of Fv/Fm
Value and corresponding spectroscopic data are analyzed, and 29 groups of obvious exceptional values are removed.293 groups of data of remainder are further analyzed, Fv/Fm is taken
Being worth section is [0.65,0.90], is divided into 5 subintervals by step-length for 0.05, Fv/Fm is fallen within to the eggplant in each subinterval
Blades reflectance spectrum is averaged, and averaged spectrum is as shown in Figure 2.It can be seen that eggplant leaf reflectance spectrum and plant spectral
Universals be consistent.In visible light wave range, since blue light can be absorbed in chlorophyll, reflectivity is formed in the region 400~500nm
Low ebb.And it is increased since chlorophyll is small for green light absorption, therefore in 500~600nm reflectivity, and deposited in 550nm or so
In reflectance peak, i.e., " green peak "；The chlorophyll of absorption due to to(for) feux rouges again, therefore reflected being formed in the region 600~700nm
Rate low ebb, and there is reflectivity minimum " red paddy " in 680nm or so.Between visible light and near infrared band, reflectivity is anxious
Play rises, and forms " red side " phenomenon.
Eggplant leaf average reflectance spectra difference corresponding to the different sections Fv/Fm is obvious.With eggplant leaf Fv/Fm liter
Height, spectrum are integrally on a declining curve.This is because Fv/Fm value reflects the maximum light energy conversion efficiency of PSII reaction center.
Sample luminous energy high conversion efficiency, absorbing ability is strong, therefore reflected light is weak.Wherein, the phenomenon is the most at 500~700nm wave band
Obviously.
3 characteristic wavelengths extract
Successive projection algorithm (successive projections algorithm, SPA) is selected before one kind to circulation
Method, it is extracted effective wave band containing minimum redundancy and minimum synteny by the Projection Analysis of vector.The party
Method is since a wave band, and circulation all calculates it in the projection not being selected on wave band every time, and the maximum wave band of projection vector is drawn
Enter to band combination.Each wave band being newly selected into, it is all minimum with previous linear relationship.Successive projection method advantageously reduces mould
Type complexity improves model calculation speed.It is low that this method can extract information repeatability in spectrum, most representative wavelength
Combination, but the wavelength unrelated with target can be also filtered out simultaneously, influence modeling accuracy.Therefore the present invention is removed using correlation coefficient process
Unrelated wavelength, so using successive projection method select wherein most representative wavelength as characteristic wavelength.
First using Pearson correlation coefficient calculation formula calculate each wavelength reflection of spectrum with Fv/Fm related coefficient,
Its calculation formula is as shown in Equation 1:
Wherein, riIt is i-th of wave band and the related coefficient of Fv/Fm, n is number of samples, xijIt is i-th of j-th of sample
The reflectivity of wave band, yjFor the Fv/Fm value of j-th of sample.Calculated result is as shown in Figure 3.
Red-label is related coefficient of the absolute value greater than 0.5 in Fig. 3.As can be seen that in 535nm to 645nm and
In 695nm to 700nm wave-length coverage, spectral reflectivity and Fv/Fm reach moderate negative correlation (- 0.8~-0.5).To correlation
Wave band of the absolute value greater than 0.5 is counted, and the results are shown in Table 2.Related coefficient peak value present in 590.66nm, for-
0.53880, correlation absolute value greater than 0.5 number of wavelengths be 251, the 18.5% of the total number of wavelengths of Zhan.This is the result shows that eggplant
There are significant correlativities for blades spectrum and its potential maximum photosynthetic capacity, for potential photosynthetic capacity value prediction model hereafter
Foundation provide foundation.
2 eggplant leaf spectrum of table and its potential photosynthetic capacity correlation statistics
| Peak wavelength (nm) | Peak value related coefficient | Absolute value is greater than 0.5 number of wavelengths |
| 590.66 | -0.53886 | 251 |
Synteny to eliminate between numerous wavelength variables influences, and is further pressed above-mentioned 251 wave bands using SPA algorithm
Characteristic wave bands are extracted in contracting, and the results are shown in Table 3.Characteristic wavelength is distributed near " green peak " and " red paddy ", this subband then has
Effect reflects reflection of the chlorophyll to green light and the absorption to feux rouges.Characteristic wavelength number is 6, only the 0.4% of the total number of wavelengths of Zhan,
This shows that correlation coefficient process combination SPA effectively reduces the dimension of spectroscopic data, reduces its data redundancy.
3 characteristic wavelength of table statistics
| Characteristic wavelength | Wavelength number |
| 536.97、583.91、598.30、615.35、637.72、629.22 | 6 |
The 4 eggplant potential photosynthetic capacity prediction models based on characteristic wavelength
Radial basis function (Radical Basis Function, RBF) network structure is simple, and training is succinct and study restrains
Speed is fast, can approach any non-linear continuous function function.It belongs to feedforward neural network, by input layer, hidden layer and defeated
Layer forms out.Wherein, hidden layer converts input vector by radial basis function, and the input data of low-dimensional is transformed to height
In dimension space, so that can divide in higher dimensional space in the inseparable problem of lower dimensional space, and the diffusion velocity of radial basis function
The value of spread directly affects the effect of RBF network, and usual spread leans on test method(s) value, needs repeatedly to attempt, and take
When it is laborious.Genetic algorithm (Genetic Algorithm, GA) due to having distribution, parallel, quick global search ability, gram
The defect of real optimal solution cannot be converged to by having taken previous Dynamic Programming nibbling method and nonlinear planning solution, be widely used in more
A field dynamic optimization problem solving.Therefore, the present invention is based on characteristic wave using the RBF neural building of genetic algorithm optimization
The potential maximum photosynthetic capacity prediction model of long eggplant.
Measure 293 groups of data are randomly divided into modeling collection and forecast set in 4:1 ratio, to model 234 groups of light of collection
The reflectivity of 6 characteristic wavelengths of modal data is as input, and the Fv/Fm value of modeling collection sample is output, using genetic algorithm optimization
Radial basis function neural network (GA-RBF) carry out the potential maximum photosynthetic capacity prediction model building of crop, GA-RBF algorithm is built
Mold process is as shown in the figure.Modeling procedure such as Fig. 4:
1. by initialization of population, spread value variation range is set for [1,30], evolutionary generation 50, population scale is
30, crossover probability 0.5, mutation probability 0.03 generates initial population at random, and each individual in population corresponds to one
Spread value generates individual chromosome using binary coding.Encode formula are as follows:
Decoding formula are as follows:
In formula, b --- the binary string after coding；M --- the taken binary string character number of chromosome；
A --- required coded decimal number；amin--- space encoder minimum decimal number；amax--- space encoder is maximum
Decimal number
2. calculating individual adaptation degree in initial population, fitness function is RBF neural, and fitness value is average for model
Error amount.I.e. using spread parameter of the individual as RBF network in population, Fv/Fm prediction model is constructed, and calculate prediction
It is worth the absolute value of mean error as the ideal adaptation angle value.Wherein, the learning procedure of RBF is, it is first determined hidden layer nerve
The center of first radial function.If training set sample input matrix is P, output matrix T, Q are training set sample number, then Q is a hidden
The corresponding radial basis function center of neuron containing layer is C=P ', determines hidden layer and output interlayer weight and threshold value is b1=
[b11,b12,...,b1Q] ', whereinWhen the radial basis function center of hidden layer neuron and threshold
After value determines, the output of hidden layer neuron are as follows:
ai=exp (- | | C-pi||2bi), i=1,2 ..., Q (4)
Wherein, pi=[pi1,pi2,...,piM] ', is i-th of training sample vector.And it is denoted as D=[d1,d2,...,dQ]。
If the connection weight w of hidden layer and output layer is
Wherein, wijIndicate the connection weight between j-th of hidden layer neuron and i-th of output layer neuron.If N number of output
The threshold value b of layer neuron2For
b2=[b21,b22,...,b2N]′ (6)
And meet:
[w,b2].[D；I]=T (7)
Wherein, [1,1 ..., 1] I=1×Q。
3. selection operation carries out population at individual by stake turntable and eliminates and select, lower its of individual adaptation degree is selected
Probability is bigger, and the selected probability of each individual is Si, calculation formula are as follows:
F in formulaiThe inverse of the fitness of individual i in-mono- population
A N-Population Size
4. crossover operation selects two individuals from population, randomly chooses some chromosome locations and swap, to generate
New individual.
5. mutation operation, an optional individual in population, a little making a variation in selective staining body be more excellent to generate
Individual.Choose j-th of gene a of i-th of individualijIt makes a variation, mutation operation method is as follows:
In formula, amaxFor gene aijThe upper bound；aminFor gene aijLower bound；F (g)=O2(1-g/Gmax)2； O2It is one
Random number, g are current evolutionary generation；GmaxFor maximum evolution number；Random number of the r between [0,1].
6. cross and variation, genetic manipulation obtains population of new generation by selection, the fitness of each generation population at individual is calculated,
Record its minimum fitness and its corresponding spread value.
7. after iterative evolution is multiple, fitness converges to minimum.Circulation is exited, the corresponding spread of minimum fitness is obtained
Value.
8. the diffusion velocity using optimal spread value as RBF network radial basis function, establishes Fv/Fm prediction model.
As shown in figure 5, increasing with evolutionary generation, respectively gradually it is lower for adaptive optimal control angle value, genetic algorithm iteration to 48 generations
Minimum is converged to, and algebra fitness thereafter no longer changes, therefore the corresponding spread value of the fitness value is optimal, is
4.256。
Using the value as optimal spread value, RBF neural is established with 234 groups of data of training set, wherein inputting
For 234 training set sample spectrums, 6 characteristic wavelength reflectivity, export as the Fv/Fm of corresponding sample.Obtain the model training collection
The coefficient of determination is 0.85675, root-mean-square error 0.013386, prediction result such as Fig. 6.Using 59 groups of data inspection in test set
Test model accuracy.As a result as shown in fig. 7, its test set coefficient of determination is 0.85379, root-mean-square error 0.013943, with instruction
It is close to practice collection result.Should be the result shows that the model accuracy be high, generalization ability is strong.The model is compared to the Fv/ using vegetation index
Fm prediction model, precision significantly improve.
The present invention provides new approaches for the potential maximum photosynthetic capacity Fast nondestructive evaluation of plant, can be used for plant physiology shape
State accurately detects.Testing cost can be reduced using the detection mode of spectral signature wavelength, detection speed is fast, is more able to satisfy practical inspection
Survey demand has good practicability.
Claims (10)
1. the potential maximum photosynthetic capacity prediction technique of the plant based on characteristic wavelength characterized by comprising
Step 1, mutually synthermal, humidity, CO under multiple and different lighting gradients are set2The culture environment of concentration, to different lighting processes
Under plant generate difference, randomly select plant leaf blade as experiment sample, measure the potential maximum light of plant leaf blade and energy respectively
Power and Visible-to-Near InfaRed reflectance spectrum, as sample data；
Step 2, random division training set and test set；
Step 3, characteristic wavelength is extracted；
It step 4, is input with the corresponding reflectivity of training set sample spectrum characteristic wavelength, the potential maximum photosynthetic capacity of plant is defeated
Out, the potential maximum photosynthetic capacity prediction model of plant is established；
Step 5, using the model, maximum photosynthetic capacity potential to plant is predicted.
2. the potential maximum photosynthetic capacity prediction technique of plant according to claim 1 based on characteristic wavelength, which is characterized in that
In the step 1, six lighting gradients, respectively 50,90,140,220,280,340 μm of olm are set-2s-1, the step 2
In, using Monte Carlo method rejecting abnormalities sample, by 4:1 random division training set and test set.
3. the potential maximum photosynthetic capacity prediction technique of plant according to claim 1 based on characteristic wavelength, which is characterized in that
In the step 3, unrelated wavelength is first removed using correlation coefficient process, is then selected using successive projection method wherein most representative
Wavelength as characteristic wavelength.
4. the potential maximum photosynthetic capacity prediction technique of plant according to claim 3 based on characteristic wavelength, which is characterized in that
Each wavelength reflection of spectrum is calculated with the related coefficient of potential maximum photosynthetic capacity using Pearson correlation coefficient calculation formula,
Under calculation formula:
Wherein, riIt is i-th of wave band and the related coefficient of Fv/Fm, n is number of samples, xijFor i-th of wave band of j-th of sample
Reflectivity, yjFor the Fv/Fm value of j-th of sample；
The unrelated wavelength refers to that related coefficient absolute value is less than the wavelength of setting value, will remove unrelated wavelength using successive projection method
Remaining wavelength further compress, extract characteristic wavelength.
5. the potential maximum photosynthetic capacity prediction technique of plant according to claim 1 based on characteristic wavelength, which is characterized in that
In the step 4, the potential maximum photosynthetic capacity of plant is established using the radial basis function neural network of genetic algorithm optimization and is predicted
Model.
6. the potential maximum photosynthetic capacity prediction technique of plant according to claim 5 based on characteristic wavelength, which is characterized in that
The establishment step of the potential maximum photosynthetic capacity prediction model of the plant is as follows:
1), by initialization of population, stochastic parameter is set and generates initial population, each individual in population corresponds to radial basis function
A spread value, using binary coding generate individual chromosome；
2) individual adaptation degree in initial population, is calculated, fitness function is radial basis function neural network, and fitness value is potential
The average error value of maximum photosynthetic capacity predicted value and measured value, i.e., using the corresponding spread value building of each individual of initial population
Potential maximum photosynthetic capacity prediction model, and the average error value of each model is calculated as its fitness value.
3), selection operation carries out initial population individual by stake turntable and eliminates and select, and lower its of individual adaptation degree is selected
Probability is bigger；
4), crossover operation selects two individuals from initial population, randomly chooses some chromosome locations and swap, to produce
Raw new individual；
5), mutation operation, an optional individual from initial population, a little making a variation in selective staining body are more excellent to generate
Elegant individual；
6), by selection, intersection, variation, heredity, population of new generation is obtained, the fitness of each generation population at individual is calculated, is recorded most
Small fitness and its corresponding spread value；
7) step 3) -6, is repeated) after iterative evolution is multiple, fitness converges to minimum, exits circulation, obtains minimum fitness
Corresponding spread value, i.e., optimal spread value；
8), the radial base diffusion velocity using optimal spread value as radial basis function neural network establishes potential maximum light
Conjunction ability prediction model.
7. the potential maximum photosynthetic capacity prediction technique of plant according to claim 6 based on characteristic wavelength, which is characterized in that
In the step 1), according to genetic algorithm optimization radial basis function neural network spread parameter, the index of spread value is set
Range is [1,30], is translated into individual by binary coding, number of individuals is set as 30, forms a population, that is, plants
Group's scale is 30, and Evolution of Population algebra is set as 50 generations, crossover probability 0.5, and mutation probability 0.03 is produced by this stochastic parameter
Raw initial population.
8. the potential maximum photosynthetic capacity prediction technique of plant according to claim 7 based on characteristic wavelength, which is characterized in that
Formula is encoded in the step 1) are as follows:
Decoding formula are as follows:
In formula, b is the binary string after coding；M takes binary string character number by chromosome；A is required coded decimal
Number；aminFor space encoder minimum decimal number；amaxFor space encoder maximum decimal number.
9. the potential maximum photosynthetic capacity prediction technique of plant according to claim 7 based on characteristic wavelength, which is characterized in that
In the step 2), the learning procedure of radial basis function is, it is first determined the center of hidden layer neuron radial function, if training
Integrate sample input matrix as P, output matrix T, Q are training set sample number, then the corresponding radial base letter of Q hidden layer neuron
Number center is C=P ', determines hidden layer and output interlayer weight and threshold value is b1=[b11,b12,...,b1Q] ', whereinAfter the radial basis function center of hidden layer neuron and threshold value determine, hidden layer neuron
Output are as follows:
Di=exp (- | | C-pi||2bi), i=1,2,, Q
Wherein, pi=[pi1,pi2,...,piM] ', is i-th of training sample vector, and is denoted as D=[d1,d2,...,dQ], it will be each
Hidden neuron output is always denoted as vector D, if the connection weight W of hidden layer and output layer is
Wherein, wijThe connection weight between j-th of hidden layer neuron and i-th of output layer neuron is indicated, if N number of output layer is refreshing
Threshold value b through member2=[b21,b22,...,b2N] ', and satisfaction: [w, b2].[D；I]=T, wherein I=[1,1 ...,11×]Q。
10. the potential maximum photosynthetic capacity prediction technique of plant according to claim 7 based on characteristic wavelength, feature exist
In in the step 3), the selected probability of each individual is Si,F in formulaiFor the adaptation of individual i in population
The inverse of degree, N indicate Population Size；
In the step 5), j-th of gene a of i-th of individual is chosenijIt makes a variation, mutation operation method is as follows:
In formula, amaxFor gene aijThe upper bound；aminFor gene aijLower bound；F (g)=o2(1-g/Gmax)2；O2It is random for one
Number, g are current evolutionary generation；GmaxFor maximum evolution number；Random number of the O between [0,1].
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910655457.9A CN110414729A (en) | 2019-07-19 | 2019-07-19 | The potential maximum photosynthetic capacity prediction technique of plant based on characteristic wavelength |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910655457.9A CN110414729A (en) | 2019-07-19 | 2019-07-19 | The potential maximum photosynthetic capacity prediction technique of plant based on characteristic wavelength |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110414729A true CN110414729A (en) | 2019-11-05 |
Family
ID=68360352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910655457.9A Pending CN110414729A (en) | 2019-07-19 | 2019-07-19 | The potential maximum photosynthetic capacity prediction technique of plant based on characteristic wavelength |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110414729A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111398200A (en) * | 2020-05-18 | 2020-07-10 | 安徽理工大学 | Method for determining optimal inversion waveband of incoherent broadband cavity enhanced absorption spectrum |
| CN111487211A (en) * | 2020-05-11 | 2020-08-04 | 安徽理工大学 | Incoherent broadband cavity enhanced absorption spectrum fitting waveband selection method |
| CN111860987A (en) * | 2020-07-08 | 2020-10-30 | 江苏科慧半导体研究院有限公司 | Mixed fluorescent material emission spectrum prediction method and device |
| CN112613391A (en) * | 2020-12-18 | 2021-04-06 | 湖北工业大学 | Hyperspectral image band selection method based on reverse learning binary rice breeding algorithm |
| CN112666328A (en) * | 2020-11-25 | 2021-04-16 | 广东省科学院广州地理研究所 | Chlorophyll a remote sensing inversion method and device based on genetic neural network model |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100068750A1 (en) * | 2006-09-13 | 2010-03-18 | Sergey Iosifovich Pogosjan | Method for fluorometrically determining photosynthesis parameters of photoautotropic organisms, device for carrying out said method and a measurement chamber |
| US20150000384A1 (en) * | 2012-02-03 | 2015-01-01 | The Trustees of Colubia University in the City of New York | Optical analysis of emissions from stimulated liquids |
| RU2013130388A (en) * | 2013-07-02 | 2015-01-10 | ГНУ Всероссийский научно-исследовательский институт имени И.В. Мичурина Россельхозакадемии (ГНУ ВНИИС) | OPTICAL METHOD FOR ESTIMATING THE STABILITY OF PHOTOSYNTHESIS PLANT TISSUES TO PHOTO INHIBITING AND DEVICE FOR ITS IMPLEMENTATION |
| CN104914053A (en) * | 2015-05-15 | 2015-09-16 | 中国农业大学 | Tree canopy photosynthetic character prediction model construction and photosynthetic character detection |
| CN105678405A (en) * | 2015-12-31 | 2016-06-15 | 西北农林科技大学 | Modeling method for cucumber GA-RBF photosynthetic rate prediction model integrated with stomatal conductance |
| CN105913150A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | BP neural network photovoltaic power station generating capacity prediction method based on genetic algorithm |
-
2019
- 2019-07-19 CN CN201910655457.9A patent/CN110414729A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100068750A1 (en) * | 2006-09-13 | 2010-03-18 | Sergey Iosifovich Pogosjan | Method for fluorometrically determining photosynthesis parameters of photoautotropic organisms, device for carrying out said method and a measurement chamber |
| US20150000384A1 (en) * | 2012-02-03 | 2015-01-01 | The Trustees of Colubia University in the City of New York | Optical analysis of emissions from stimulated liquids |
| RU2013130388A (en) * | 2013-07-02 | 2015-01-10 | ГНУ Всероссийский научно-исследовательский институт имени И.В. Мичурина Россельхозакадемии (ГНУ ВНИИС) | OPTICAL METHOD FOR ESTIMATING THE STABILITY OF PHOTOSYNTHESIS PLANT TISSUES TO PHOTO INHIBITING AND DEVICE FOR ITS IMPLEMENTATION |
| CN104914053A (en) * | 2015-05-15 | 2015-09-16 | 中国农业大学 | Tree canopy photosynthetic character prediction model construction and photosynthetic character detection |
| CN105678405A (en) * | 2015-12-31 | 2016-06-15 | 西北农林科技大学 | Modeling method for cucumber GA-RBF photosynthetic rate prediction model integrated with stomatal conductance |
| CN105913150A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | BP neural network photovoltaic power station generating capacity prediction method based on genetic algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 朱艳;田永超;马吉锋;姚霞;刘小军;曹卫星;: "小麦叶片叶绿素荧光参数与反射光谱特征的关系", 作物学报, no. 08, pages 1287 - 1292 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111487211A (en) * | 2020-05-11 | 2020-08-04 | 安徽理工大学 | Incoherent broadband cavity enhanced absorption spectrum fitting waveband selection method |
| CN111487211B (en) * | 2020-05-11 | 2022-09-30 | 安徽理工大学 | Incoherent broadband cavity enhanced absorption spectrum fitting waveband selection method |
| CN111398200A (en) * | 2020-05-18 | 2020-07-10 | 安徽理工大学 | Method for determining optimal inversion waveband of incoherent broadband cavity enhanced absorption spectrum |
| CN111398200B (en) * | 2020-05-18 | 2022-09-30 | 安徽理工大学 | Method for determining optimal inversion waveband of incoherent broadband cavity enhanced absorption spectrum |
| CN111860987A (en) * | 2020-07-08 | 2020-10-30 | 江苏科慧半导体研究院有限公司 | Mixed fluorescent material emission spectrum prediction method and device |
| CN112666328A (en) * | 2020-11-25 | 2021-04-16 | 广东省科学院广州地理研究所 | Chlorophyll a remote sensing inversion method and device based on genetic neural network model |
| CN112613391A (en) * | 2020-12-18 | 2021-04-06 | 湖北工业大学 | Hyperspectral image band selection method based on reverse learning binary rice breeding algorithm |
| CN112613391B (en) * | 2020-12-18 | 2022-06-07 | 湖北工业大学 | Hyperspectral image waveband selection method based on reverse learning binary rice breeding algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110414729A (en) | The potential maximum photosynthetic capacity prediction technique of plant based on characteristic wavelength | |
| CN102789579B (en) | Identification method for stressed state of water fertilizer of greenhouse crop on basis of computer vision technology | |
| CN108875913A (en) | A kind of matsutake Fast nondestructive evaluation system and method based on convolutional neural networks | |
| CN107271382A (en) | A kind of different growing rape leaf SPAD value remote sensing estimation methods | |
| CN106874621B (en) | A kind of aciculignosa canopy reflectance spectrum computational methods | |
| CN105678405B (en) | Modeling method for cucumber GA-RBF photosynthetic rate prediction model integrated with stomatal conductance | |
| CN104778349B (en) | One kind is used for rice table soil nitrogen application Classified Protection | |
| CN109540836A (en) | Near infrared spectrum pol detection method and system based on BP artificial neural network | |
| CN110569605B (en) | NSGA 2-ELM-based non-glutinous rice leaf nitrogen content inversion model method | |
| CN107271372A (en) | A kind of Apple Leaves chlorophyll remote sensing estimation method | |
| CN110082309B (en) | Method for establishing SPAD value comprehensive spectrum monitoring model of winter wheat canopy | |
| Huang et al. | Detection of scab in wheat ears using in situ hyperspectral data and support vector machine optimized by genetic algorithm. | |
| CN105608296B (en) | A kind of blade potassium concn inversion method based on lichee canopy spectra | |
| CN111965117A (en) | Winter wheat moisture monitoring method and system based on PROSPECT model | |
| CN103278467A (en) | Rapid nondestructive high-accuracy method with for identifying abundance degree of nitrogen element in plant leaf | |
| CN112735511A (en) | Cold damage cucumber PSII potential activity prediction method based on QGA-SVR | |
| Liu et al. | Diagnosis of citrus greening using raman spectroscopy-based pattern recognition | |
| Yu et al. | Combining the critical nitrogen concentration and machine learning algorithms to estimate nitrogen deficiency in rice from UAV hyperspectral data | |
| CN112381133A (en) | Method for determining reasonable sample number in forest ground biomass remote sensing estimation | |
| CN112362812A (en) | Remote sensing inversion model and method for rice leaf chlorophyll carotenoid content ratio based on Lars algorithm | |
| Weng et al. | Drought stress identification of tomato plant using multi-features of hyperspectral imaging and subsample fusion | |
| CN115684107A (en) | Rice salt stress early-stage quantitative monitoring method based on sunlight-induced chlorophyll fluorescence index | |
| CN115080905A (en) | Remote sensing inversion method for chlorophyll a concentration of plateau lake | |
| Li et al. | HSI combined with CNN model detection of heavy metal Cu stress levels in apple rootstocks | |
| Xie et al. | Estimating stomatal conductance of citrus under water stress based on multispectral imagery and machine learning methods |
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 |