CN105842190B - A kind of method for transferring near infrared model returned based on spectrum - Google Patents
A kind of method for transferring near infrared model returned based on spectrum Download PDFInfo
- Publication number
- CN105842190B CN105842190B CN201610153646.2A CN201610153646A CN105842190B CN 105842190 B CN105842190 B CN 105842190B CN 201610153646 A CN201610153646 A CN 201610153646A CN 105842190 B CN105842190 B CN 105842190B
- Authority
- CN
- China
- Prior art keywords
- near infrared
- infrared spectrum
- spectrum
- slave
- host
- 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
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000001228 spectrum Methods 0.000 title claims abstract description 45
- 238000002329 infrared spectrum Methods 0.000 claims abstract description 114
- 241000208125 Nicotiana Species 0.000 claims abstract description 53
- 235000002637 Nicotiana tabacum Nutrition 0.000 claims abstract description 53
- 238000012546 transfer Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 9
- 239000000126 substance Substances 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 30
- 230000009466 transformation Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 4
- 238000003780 insertion Methods 0.000 claims description 4
- 230000037431 insertion Effects 0.000 claims description 4
- 230000001373 regressive effect Effects 0.000 claims description 2
- 235000019504 cigarettes Nutrition 0.000 claims 1
- 238000004566 IR spectroscopy Methods 0.000 abstract description 5
- 230000003595 spectral effect Effects 0.000 description 6
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 4
- 238000002790 cross-validation Methods 0.000 description 3
- SNICXCGAKADSCV-JTQLQIEISA-N (-)-Nicotine Chemical compound CN1CCC[C@H]1C1=CC=CN=C1 SNICXCGAKADSCV-JTQLQIEISA-N 0.000 description 2
- ZAMOUSCENKQFHK-UHFFFAOYSA-N Chlorine atom Chemical compound [Cl] ZAMOUSCENKQFHK-UHFFFAOYSA-N 0.000 description 2
- ZLMJMSJWJFRBEC-UHFFFAOYSA-N Potassium Chemical compound [K] ZLMJMSJWJFRBEC-UHFFFAOYSA-N 0.000 description 2
- 239000000460 chlorine Substances 0.000 description 2
- 229910052801 chlorine Inorganic materials 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 229960002715 nicotine Drugs 0.000 description 2
- SNICXCGAKADSCV-UHFFFAOYSA-N nicotine Natural products CN1CCCC1C1=CC=CN=C1 SNICXCGAKADSCV-UHFFFAOYSA-N 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 150000008442 polyphenolic compounds Chemical class 0.000 description 2
- 235000013824 polyphenols Nutrition 0.000 description 2
- 239000011591 potassium Substances 0.000 description 2
- 229910052700 potassium Inorganic materials 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000011425 standardization method Methods 0.000 description 1
- 238000013519 translation Methods 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
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- 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
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of method for transferring near infrared model returned based on spectrum, comprising: step 1, demarcates sample for multiple tobaccos, is utilized respectively host and slave carries out near infrared spectra collection, obtain host near infrared spectrum XmWith slave near infrared spectrum Xs；Step 2, to XmAnd XsPretreated spectra is carried out respectively；Step 3, using spectrum homing method, X is calculated separatelymAnd XsIn the expression Z of low dimensionalmAnd Zs；Step 4, Z is utilizedmAnd ZsTransformation is calculated, by the host near infrared spectrum X of training set tobacco sampletrnTransfer is slave near infrared spectrum Xt；Step 5, slave near infrared spectrum X is utilizedtAnd the chemical score of tobacco sample is modeled；Step 6, near infrared spectra collection is carried out to sample to be tested using slave, the calculating of tobacco components content is then carried out using the model that step 5 is established.Method provided by the invention can disclose the internal structure of data, inhibit noise and redundancy feature in data, improve the probability of success of infrared spectroscopy Model transfer.
Description
Technical field
The present invention relates to Infrared Spectrum Technology fields, and in particular to a kind of near-infrared model transfer returned based on spectrum
Method.
Background technique
Infrared spectroscopy is quick, accurate and lossless because having the advantages that, is widely used in industrial circle.Spectrum Multivariate Correction
Technology can be efficiently used for material component content detection and online process monitoring, but multivariate calibration techniques are in practical applications
It usually will receive limitation, new environmental condition can not be effectively applied to this is mainly due to the multivariate calibration model having built up
The infrared spectrum of lower observation or the spectrogram of different instrument acquisitions.
The method of re-calibrating can overcome this limitation, but require to re-establish model every time, and not only cost is suitable
Greatly, and it is time-consuming and laborious.Model transfer is a kind of effective ways for solving the problems, such as above-mentioned model and promoting, it will be on an instrument
The qualitative or quantitative calibration model established, which is reliably transplanted on other same or similar instruments, to be used, or will be in a certain condition
The spectrogram that the model of lower foundation acquires under the conditions of another for same instrument, essence are different between elimination measuring signal
Cause property, makes it suitable for the same model.
Model transfer is a kind of method of spectrum transform, and master will be mapped to from instrument spectral response by finding a transformation matrix
Instrument, implementation model is shared, and representative method has sub-space learning, directly standardization (DirectStandard, DS) and divides
Section directly standardization (PiecewiseDirectStandard, PDS).
Common sub-space learning method has principal component analysis (PCA), locality preserving projections (LPP) and neighborhood to keep insertion
(NPE) etc., these sub-space learning methods can be brought under the Unified frame of figure insertion.Involved in sub-space learning method
To the feature decomposition of dense matrix, calculation amount and amount of storage are larger, and when data dimension be more than number of samples when algorithm pole not
Stablize.
The transformation matrix between principal and subordinate's instrument spectral is directly sought in direct standardized method.Piecewise direct standardization method
In, some wavelength of main instrument establishes transformation relation with from the corresponding local spectrum range of instrument.In existing spectrum transform method
In, PDS method is most widely used algorithm, this local regression model being primarily due in PDS method is able to reflect principal and subordinate
Variation of the instrument spectral at corresponding wavelength, while PDS method is based on spectral signal and is directly corrected, it is easy to use.
But PDS method needs to be determined in advance the size of mapping window, and transfer is easy to appear when window selection is inappropriate
Failure.
Summary of the invention
The present invention provides a kind of method for transferring near infrared model returned based on spectrum, can disclose the internal junction of data
Structure inhibits noise and redundancy feature in data, improves the probability of success of infrared spectroscopy Model transfer.
A kind of method for transferring near infrared model returned based on spectrum, comprising:
Step 1, sample is demarcated for multiple tobaccos, is utilized respectively host and slave carries out near infrared spectra collection, obtains
Host near infrared spectrum XmWith slave near infrared spectrum Xs；
Step 2, to host near infrared spectrum XmWith slave near infrared spectrum XsPretreated spectra is carried out respectively；
Step 3, using spectrum homing method, host near infrared spectrum X is calculated separatelymWith slave near infrared spectrum XsIn low-dimensional
The expression Z of degreemAnd Zs；
Step 4, Z is utilizedmAnd ZsTransformation is calculated, by the host near infrared spectrum X of training set tobacco sampletrnTransfer is slave
Near infrared spectrum Xt；
Step 5, slave near infrared spectrum X is utilizedtAnd the chemical score of tobacco sample is modeled；
Step 6, using slave to sample to be tested carry out near infrared spectra collection, then using step 5 establish model into
The calculating of row tobacco components content.
Method for transferring near infrared model provided by the invention is used to solve the problems, such as that the model between different nir instruments to share,
It carries out spectrum respectively by the near infrared spectrum to host and slave and returns insertion, then look for the relationship between transformation spectrogram.It adopts
The immanent structure that data can be disclosed with spectrum homing method, inhibits the noise and redundancy feature in data, improves Model transfer
The probability of success, the prediction error of the model after reducing transfer.
It models, is avoided using slave after using the method for the present invention that main engine modeling spectral translation is modeled spectrum for slave
It after re-starting measurement to sample, then is modeled, reduces the time cost and economic cost of analysis.
The host and slave is the near infrared spectrometer of two same models, and host is the near infrared spectrum for having model
Instrument, slave are the near infrared spectrometer of not model.In step 1 when carrying out near-infrared spectral measurement, in identical test-strips
It is carried out under part.
Near infrared spectrum is down to low-dimensional using the method that spectrum returns by the present invention, appearance when avoiding dimension greater than sample number
Unstable problem is calculated, in addition, the invention avoids existing department pattern transfer methods to need application model information, directly
Realize the transfer by host to slave.
It after converting slave near infrared spectrum for host near infrared spectrum, is modeled using chemical score, then to slave
Sample carries out the prediction of tobacco components content using the model established.
Preferably, step 3 the following steps are included:
Step 3-1 constructs host near infrared spectrum X respectivelymWith slave near infrared spectrum XsCorrelation matrix；
Step 3-2 is utilized respectively spectrum Regressive Solution host near infrared spectrum XmWith slave near infrared spectrum XsFeature to
Amount, finds h feature vector, and the maximum value of h is the quantity of transfer correction spectrum；
Step 3-3 solves host near infrared spectrum X using the resulting h feature vector of step 3 respectivelymIt is closely red with slave
External spectrum XsProjection vector collection；
Step 3-4, using following formula respectively to host near infrared spectrum XmWith slave near infrared spectrum XsIt is embedding to carry out spectrum recurrence
Enter:
In formula, AmFor the projection vector collection of host near infrared spectrum；
AsFor the projection vector collection of slave near infrared spectrum.
In step 3-1~step 3-4, identical processing side is used for host near infrared spectrum and slave near infrared spectrum
Formula is respectively processed.
The chemical score of tobacco sample in step 5 according to People's Republic of China (PRC) tobacco business standard YC/T32-1996,
YC/T161-2002, YC/T160-2002, YC/T173-2003, YC/T162-2002, YC/T202-2006 detect the total of tobacco
Sugar, reduced sugar, total nitrogen, nicotine, potassium, chlorine, polyphenol content.
Modeling method in step 5 is partial least squares algorithm (Partial Least Squares, PLS), latent variable number
Purpose is chosen for 5 folding cross validations.
Preferably, calculating transformation matrix F using following formula in step 41With transformation matrix F2:
In formula: the broad sense inverse operation of+representing matrix.
Preferably, in step 4, using following formula by the host near infrared spectrum X of training set tobacco sampletrnTransfer for from
Machine near infrared spectrum Xt:
Preferably, in step 2, to host near infrared spectrum XmWith slave near infrared spectrum XsMake identical spectrum to locate in advance
Reason, Pretreated spectra be it is smooth, seek first derivative, seek second dervative, at least one of standard normal corrects.
Preferably, in step 3-1 correlation matrix element wijCalculation formula it is as follows:
In formula: xiFor the near infrared spectrum of tobacco sample i；
xjFor the near infrared spectrum of tobacco sample j；
N(xi) indicate tobacco sample xiNeighbour domain；
N(xj) indicate tobacco sample xjNeighbour domain；
σ takes 0.1.
Preferably, solving Wy=λ Dy in step 3-2, finding maximum h feature vector；
In formula, W is correlation matrix；
λ is regularization parameter:
D is a diagonal matrix, and diagonal element is
Preferably, solving projection vector in step 3-3 using regularization least square method, projection vector being made to meet item
Part XTY=a, in formula, a is projection vector；X is the machine near infrared spectrum or slave near infrared spectrum that tobacco demarcates sample；Y is
Feature vector corresponding with X.
Preferably, calculation formula when solving projection vector using regularization least square method is as follows in step 3-3:
In formula, a is projection vector；
X is the machine near infrared spectrum or slave near infrared spectrum that tobacco demarcates sample；
Y is feature vector corresponding with X；
γ is regularization parameter.
The method for transferring near infrared model provided by the invention returned based on spectrum, can disclose the internal structure of data, press down
Noise and redundancy feature in data processed improve the probability of success of infrared spectroscopy Model transfer.
Detailed description of the invention
Fig. 1 is the flow chart of the method for transferring near infrared model returned the present invention is based on spectrum；
Fig. 2 a is the host near infrared spectrum that tobacco demarcates sample；
Fig. 2 b is the slave near infrared spectrum that tobacco demarcates sample；
Fig. 3 a is the difference of host near infrared spectrum and slave near infrared spectrum；
Fig. 3 b is the difference that aft engine near infrared spectrum and slave near infrared spectrum are handled using PDS method；
Fig. 3 c is using method for transferring near infrared model provided by the invention treated host near infrared spectrum and slave
The difference of near infrared spectrum.
Specific embodiment
With reference to the accompanying drawing, the method for transferring near infrared model returned the present invention is based on spectrum is described in detail.
As shown in Figure 1, the method for transferring near infrared model returned based on spectrum, comprising:
Step 1,33 tobacco leaf samples is selected to demarcate sample as tobacco, for be respectively utilized respectively host and slave carry out it is close
Infrared spectroscopy acquisition, obtains host near infrared spectrum XmWith slave near infrared spectrum Xs, the instrument type of host and slave and adopt
Sample parameter is consistent.Host near infrared spectrum is as shown in Figure 2 a, and slave near infrared spectrum is as shown in Figure 2 b.
Step 2, it is corrected using standard normal to host near infrared spectrum XmWith slave near infrared spectrum XsIt carries out identical pre-
Processing.
Step 3, using spectrum homing method, host near infrared spectrum X is calculated separatelymWith slave near infrared spectrum XsIn low-dimensional
The expression Z of degreemAnd Zs, concrete operations are as follows:
Step 3-1 constructs host near infrared spectrum X respectivelymWith slave near infrared spectrum XsCorrelation matrix, wherein
Host near infrared spectrum XmCorresponding correlation matrix is Wm, slave near infrared spectrum XsCorresponding correlation matrix is Ws。
If there is K tobacco calibration sample, correlation matrix is the symmetrical matrix of a K × K, correlation matrix
Element wijIndicate the correlation of tobacco sample i and tobacco sample j, the element w of correlation matrixijCalculation formula it is as follows:
In formula: xiFor the near infrared spectrum of tobacco sample i；
xjFor the near infrared spectrum of tobacco sample j；
N(xi) indicate tobacco sample xiNeighbour domain, the present embodiment takes 3 neighborhoods (i.e. apart from tobacco sample xiNearest 3
Tobacco sample)；
N(xj) indicate tobacco sample xjNeighbour domain, the present embodiment takes 3 neighborhoods；
σ takes 0.1.
Step 3-2 solves host near infrared spectrum X respectivelymWith slave near infrared spectrum XsFeature vector, find maximum
H feature vector be respectively y1y2...yh。
In the step, Wy=λ Dy is solved, finds h feature vector, h maximum value is to shift the quantity of calibration samples, this reality
It applies example and takes 30.H is too small, and dimensionality reduction has lost excessive information, and generally when sample number is little, h value is more slightly lower than sample number i.e.
It can.
In formula, W is correlation matrix；
λ is regularization parameter, and the present embodiment takes 0.001；
D is a diagonal matrix, diagonal element Dii=Σjwji。
Step 3-3 solves host near infrared spectrum X using step 3 gained feature vector respectivelymWith slave near infrared light
Compose XsProjection vector.
In this step, projection vector is solved using regularization least square method, projection vector is made to meet condition XTY=a, formula
In, a is projection vector；X is the machine near infrared spectrum or slave near infrared spectrum that tobacco demarcates sample；Y is spy corresponding with X
Levy vector.
For example, XmProjection vector be a1a2…ah, XmProjection vector collection Am=(a1,a2,…,ah), meet condition Xm Tyi
=ai, the value range of i is 1~h.
Calculation formula when solving projection vector using regularization least square method is as follows:
In formula, a is projection vector；
X is the machine near infrared spectrum or slave near infrared spectrum that tobacco demarcates sample；
Y is feature vector corresponding with X；
γ is regularization parameter, takes 0.001 in the present embodiment.
Step 3-4, using following formula respectively to host near infrared spectrum XmWith slave near infrared spectrum XsIt is embedding to carry out spectrum recurrence
Enter:
In formula, AmFor the projection vector collection of host near infrared spectrum；
AsFor the projection vector collection of slave near infrared spectrum.
Step 4, Z is utilizedmAnd ZsTransformation matrix is calculated, by the host near infrared spectrum X of training set tobacco sampletrnTransfer is
Slave near infrared spectrum Xt。
In the step, transformation matrix F is calculated using following formula1With transformation matrix F2:
In formula: the broad sense inverse operation of+representing matrix.
In the step, using following formula by the host near infrared spectrum X of training set tobacco sampletrnTransfer is slave near-infrared
Spectrum Xt:
Step 5, slave near infrared spectrum X is utilizedtAnd the chemical score of tobacco sample is modeled.
The chemical score of tobacco sample in the step according to People's Republic of China (PRC) tobacco business standard YC/T32-1996,
YC/T161-2002, YC/T160-2002, YC/T173-2003, YC/T162-2002, YC/T202-2006 detect the total of tobacco
Sugar, reduced sugar, total nitrogen, nicotine, potassium, chlorine, polyphenol content.
Modeling method in the step is partial least squares algorithm (Partial Least Squares, PLS), latent variable
Number is chosen for 5 folding cross validations.
Step 6, using slave to sample to be tested carry out near infrared spectra collection, then using step 5 establish model into
The calculating of row tobacco components content.
Fig. 3 a~Fig. 3 b respectively indicates difference spectrum of the tobacco calibration sample on host and slave, can be seen by Fig. 3 a~3b
Out, difference of the tobacco calibration sample on host and slave can be reduced using transfer method provided by the invention.
Spectrum predicts that error is as shown in table 1 in a model after shifting under different Model transfer methods.
Table 1
In table 1, RMSEC: training set root mean square error；RMSECV: training set cross validation root mean square error；RMSEP: it surveys
Examination collection root mean square error.
Method provided by the invention compares traditional Model transfer method, is returned using spectrum, in low dimensional implementation model
Transfer avoids the singularity in calculating caused by being greater than sample number because of dimension, and the model after ensure that transfer has good
Application effect.
Claims (4)
1. a kind of method for transferring near infrared model returned based on spectrum characterized by comprising
Step 1, sample is demarcated for multiple tobaccos, is utilized respectively host and slave carries out near infrared spectra collection, obtains host
Near infrared spectrum XmWith slave near infrared spectrum Xs；
Step 2, to host near infrared spectrum XmWith slave near infrared spectrum XsPretreated spectra is carried out respectively；In step 2, to master
Machine near infrared spectrum XmWith slave near infrared spectrum XsMake identical Pretreated spectra, Pretreated spectra be it is smooth, ask single order to lead
Number asks at least one of second dervative, standard normal correction；
Step 3, using spectrum homing method, host near infrared spectrum X is calculated separatelymWith slave near infrared spectrum XsIn low dimensional
Indicate ZmAnd Zs；
Step 3 the following steps are included:
Step 3-1 constructs host near infrared spectrum X respectivelymWith slave near infrared spectrum XsCorrelation matrix；Related coefficient
The element w of matrixijCalculation formula it is as follows:
In formula: xiFor the near infrared spectrum of tobacco sample i；
xjFor the near infrared spectrum of tobacco sample j；
N(xi) indicate tobacco sample xiNeighbour domain；
N(xj) indicate tobacco sample xjNeighbour domain；
σ takes 0.1；
Step 3-2 is utilized respectively spectrum Regressive Solution host near infrared spectrum XmWith slave near infrared spectrum XsFeature vector, look for
To h feature vector, the maximum value of h is the quantity of transfer correction spectrum；
Step 3-3 solves host near infrared spectrum X using the resulting h feature vector of step 3 respectivelymWith slave near infrared light
Compose XsProjection vector collection；
Step 3-4, using following formula respectively to host near infrared spectrum XmWith slave near infrared spectrum XsIt carries out spectrum and returns insertion:
In formula, AmFor the projection vector collection of host near infrared spectrum；
AsFor the projection vector collection of slave near infrared spectrum；
Step 4, Z is utilizedmAnd ZsTransformation is calculated, by the host near infrared spectrum X of training set tobacco sampletrnTransfer is that slave is closely red
External spectrum Xt；
In step 4, transformation matrix F is calculated using following formula1With transformation matrix F2:
In formula: the broad sense inverse operation of+representing matrix；
Using following formula by the host near infrared spectrum X of training set tobacco sampletrnTransfer is slave near infrared spectrum Xt:
Step 5, slave near infrared spectrum X is utilizedtAnd the chemical score of tobacco sample is modeled；
Step 6, near infrared spectra collection is carried out to sample to be tested using slave, then carries out cigarette using the model that step 5 is established
The calculating of careless chemical composition content.
2. the method for transferring near infrared model returned as described in claim 1 based on spectrum, which is characterized in that in step 3-2, ask
Wy=λ Dy is solved, maximum h feature vector is found；
In formula, W is correlation matrix；
λ is regularization parameter；
D is a diagonal matrix, diagonal element Dii=∑jwji。
3. the method for transferring near infrared model returned as claimed in claim 2 based on spectrum, which is characterized in that in step 3-3, benefit
Projection vector is solved with regularization least square method, projection vector is made to meet condition XTY=a, in formula, a is projection vector；X is
The machine near infrared spectrum or slave near infrared spectrum of tobacco calibration sample；Y is feature vector corresponding with X.
4. the method for transferring near infrared model returned as claimed in claim 3 based on spectrum, which is characterized in that in step 3-3, benefit
Calculation formula when solving projection vector with regularization least square method is as follows:
In formula, a is projection vector；
X is the machine near infrared spectrum or slave near infrared spectrum that tobacco demarcates sample；
Y is feature vector corresponding with X；
γ is regularization parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610153646.2A CN105842190B (en) | 2016-03-17 | 2016-03-17 | A kind of method for transferring near infrared model returned based on spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610153646.2A CN105842190B (en) | 2016-03-17 | 2016-03-17 | A kind of method for transferring near infrared model returned based on spectrum |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105842190A CN105842190A (en) | 2016-08-10 |
| CN105842190B true CN105842190B (en) | 2018-12-11 |
Family
ID=56587170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610153646.2A Active CN105842190B (en) | 2016-03-17 | 2016-03-17 | A kind of method for transferring near infrared model returned based on spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105842190B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106872396B (en) * | 2016-12-28 | 2019-06-07 | 南京农业大学 | A kind of method of difference nir instrument measurement grape pol model conversion |
| CN106872397A (en) * | 2016-12-29 | 2017-06-20 | 深圳市芭田生态工程股份有限公司 | A kind of method based on existing calibration model quick detection agricultural product chemical constituent |
| CN106951720B (en) * | 2017-04-12 | 2019-05-31 | 山东省科学院海洋仪器仪表研究所 | Soil nutrient Model transfer method based on canonical correlation analysis and linear interpolation |
| CN108287144B (en) * | 2017-12-05 | 2020-12-08 | 华东理工大学 | Preparation method and application of standard sample transmitted by natural plant near-infrared model |
| CN109444072B (en) * | 2018-10-12 | 2021-02-09 | 南京富岛信息工程有限公司 | Method for solving rogue peaks in near-infrared spectrum model transfer |
| CN109444066B (en) * | 2018-10-29 | 2020-04-14 | 山东大学 | Model transfer method based on spectral data |
| CN111579526B (en) * | 2020-06-08 | 2023-05-30 | 贵州中烟工业有限责任公司 | Method for representing difference and correction of near infrared instrument |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102135496A (en) * | 2010-12-23 | 2011-07-27 | 华东交通大学 | Infrared spectrum quantitative analysis method and infrared spectrum quantitative analysis device based on multi-scale regression |
| CN103018178A (en) * | 2012-12-06 | 2013-04-03 | 江苏省质量安全工程研究院 | LPP (Local Preserving Projection)-based Infrared spectrometer calibration method |
| CN103592257A (en) * | 2013-12-02 | 2014-02-19 | 广东产品质量监督检验研究院 | Method for fast determining toluene, ethylbenzene and xylene content of woodware paint diluent by near-infrared spectroscopy |
| CN104596982A (en) * | 2015-01-30 | 2015-05-06 | 云南中烟工业有限责任公司 | Method for measuring pectin of paper-making reconstituted tobacco by near-infrared diffuse reflection spectrum technology |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4054853B2 (en) * | 2000-10-17 | 2008-03-05 | 独立行政法人農業・食品産業技術総合研究機構 | Blood analysis using near infrared spectroscopy |
-
2016
- 2016-03-17 CN CN201610153646.2A patent/CN105842190B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102135496A (en) * | 2010-12-23 | 2011-07-27 | 华东交通大学 | Infrared spectrum quantitative analysis method and infrared spectrum quantitative analysis device based on multi-scale regression |
| CN103018178A (en) * | 2012-12-06 | 2013-04-03 | 江苏省质量安全工程研究院 | LPP (Local Preserving Projection)-based Infrared spectrometer calibration method |
| CN103592257A (en) * | 2013-12-02 | 2014-02-19 | 广东产品质量监督检验研究院 | Method for fast determining toluene, ethylbenzene and xylene content of woodware paint diluent by near-infrared spectroscopy |
| CN104596982A (en) * | 2015-01-30 | 2015-05-06 | 云南中烟工业有限责任公司 | Method for measuring pectin of paper-making reconstituted tobacco by near-infrared diffuse reflection spectrum technology |
Non-Patent Citations (2)
| Title |
|---|
| Near-infrared calibration transfer based on spectral regression;Jiangtao Peng等;《Spectrochimica Acta Part A》;20111231(第78期);第2节，附图1 * |
| Spectral Regression for Efficient Regularized Subspace Learning;Deng Cai等;《IDEALS》;20090422;公式（6） * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105842190A (en) | 2016-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105842190B (en) | A kind of method for transferring near infrared model returned based on spectrum | |
| Clairotte et al. | National calibration of soil organic carbon concentration using diffuse infrared reflectance spectroscopy | |
| Mishra et al. | Partial least square regression versus domain invariant partial least square regression with application to near-infrared spectroscopy of fresh fruit | |
| Shiklomanov et al. | Quantifying the influences of spectral resolution on uncertainty in leaf trait estimates through a Bayesian approach to RTM inversion | |
| Cho et al. | Towards red‐edge positions less sensitive to canopy biophysical parameters for leaf chlorophyll estimation using properties optique spectrales des feuilles (PROSPECT) and scattering by arbitrarily inclined leaves (SAILH) simulated data | |
| Peng et al. | Near-infrared calibration transfer based on spectral regression | |
| CN103854305B (en) | A kind of Model Transfer method based on multi-scale Modeling | |
| Zhang et al. | Classification modeling method for near‐infrared spectroscopy of tobacco based on multimodal convolution neural networks | |
| CN104990895B (en) | A kind of near infrared spectrum signal standards normal state bearing calibration based on regional area | |
| Shafri et al. | Detection of stressed oil palms from an airborne sensor using optimized spectral indices | |
| CN114139444A (en) | Offshore sea surface temperature inversion method based on machine learning | |
| CN111563436B (en) | Infrared spectrum measuring instrument calibration migration method based on CT-CDD | |
| Lu et al. | Quantitative determination of rice starch based on hyperspectral imaging technology | |
| CN105628645A (en) | Infrared analysis model transfer method | |
| Li et al. | Wavelength selection method for near-infrared spectroscopy based on standard-sample calibration transfer of mango and apple | |
| Lu et al. | Fluorescence hyperspectral image technique coupled with HSI method to predict solanine content of potatoes | |
| Yu et al. | Prediction of soil properties based on characteristic wavelengths with optimal spectral resolution by using Vis-NIR spectroscopy | |
| Preys et al. | Robust calibration using orthogonal projection and experimental design. Application to the correction of the light scattering effect on turbid NIR spectra | |
| Hong-hong et al. | Application of wavelength selection strategy based on SWCSS-UVE algorithm in improving model transfer for near infrared analysis of holocellulose and lignin | |
| Yoshioka et al. | An isoline-based translation technique of spectral vegetation index using EO-1 Hyperion data | |
| CN112926625B (en) | Deviation influence factor analysis method for satellite radiation data | |
| Liu et al. | Estimation of chlorophyll content in maize canopy using wavelet denoising and SVR method | |
| Shao et al. | Noise-sensitivity analysis and improvement of automatic retrieval of temperature and emissivity using spectral smoothness | |
| Zhang et al. | A fast progressive spectrum denoising combined with partial least squares algorithm and its application in online Fourier transform infrared quantitative analysis | |
| CN115984715A (en) | Hyperspectral satellite adaptive atmosphere correction method based on prior ground object spectrum matching |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |