CN105223140A - The method for quickly identifying of homology material - Google Patents
The method for quickly identifying of homology material Download PDFInfo
- Publication number
- CN105223140A CN105223140A CN201510654599.5A CN201510654599A CN105223140A CN 105223140 A CN105223140 A CN 105223140A CN 201510654599 A CN201510654599 A CN 201510654599A CN 105223140 A CN105223140 A CN 105223140A
- Authority
- CN
- China
- Prior art keywords
- plsda
- predicted value
- sample
- homology
- model
- 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
Abstract
The invention discloses a kind of method for quickly identifying of homology material, comprising: step 1) gather the spectral signal belonging to several samples of homology material a respectively; Step 2) pre-service is carried out to the described spectral signal of each sample; Step 3) in qualitative recognition model, set up PLSDA analytical model, obtain each sample at predicted value A corresponding to PLSDA analytical model; Step 4) determine that the preliminary threshold of this homology material a in PLSDA analytical model is interval; Step 5) scope in preliminary threshold interval described in successive optimization, until obtain the highest predicted value threshold interval of recognition correct rate; Step 6) arbitrary substance b is identified.The invention solves that homology material is not easy to identify and the technical matters such as discrimination is low, recognition speed is slow, solve the adulterated identification of material simultaneously, thus the resolution of model and convenient service can be improved in production application.
Description
Technical field
The present invention relates to Chemical Measurement Data Processing in Experiment technical field, particularly a kind of method for quickly identifying of homology material.
Background technology
Partial least square method identification is the Multivariate Correction analytical approach based on factorial analysis, in the method that a kind of linear identification grown up on the basis of traditional multiple linear regression returns, having very strong jamproof ability, is the model of most widely used qualitative recognition.At present, spectroscopy is more popular in conjunction with the research of PLSDA to homology Object Classification and adulterated identification, but relatively very few about the research of PLSDA threshold value.ChenYi etc., when setting up PLSDA Model Identification Ganodenna Lucidum P.E Source Tracing, propose using ± 0.25 as boundary value, and with SNV in conjunction with 2 order derivative process, discrimination reaches 100%.Shao Ping, Wang Jun etc., when setting up PLSDA Model Identification Ganodenna Lucidum P.E and Coriolus Versicolor P.E., propose using ± 0.5 as boundary value, although its discrimination also reaches 100%, when large number of samples, the accuracy of the identification of signal will reduce.
Carry out homology material due to Chemical Measurement in conjunction with spectral analysis to trace to the source and adulterated discriminance analysis is the probability analysis of mass data, data volume number have vital factor for the degree of accuracy of result of experiment, being therefore badly in need of one can identify homology material and adulterated knowledge method for distinguishing fast.
Summary of the invention
For above-mentioned technical matters, a kind of method for quickly identifying of homology material is proposed in the present invention, the threshold value of PLSDA can be determined by method of the present invention, the present invention carrys out optimum option PLSDA threshold interval by normal distribution 3 σ principle, normal distribution is also known as Gaussian distribution, be one in all very important probability distribution in field such as mathematics, physics and engineerings, many aspects statistically have great significance.Normal distribution is probability distribution, under normal curve, according to 3 σ principle: P (μ-σ <X≤μ+σ)=68.3%, P (μ-2 σ <X≤μ+2 σ)=95.4%, P (μ-3 σ <X≤μ+3 σ)=99.7%, chooses by normal distribution 3 σ principle the discrimination that PLSDA threshold interval has superelevation as can be seen here.
Key of the present invention is by the further process to PLSDA data, proposes a kind of method of new determination PLSDA threshold value, for practical application provides more reliable data.
The invention solves that homology material is not easy to identify and the technical matters such as discrimination is low, recognition speed is slow, solve the adulterated identification of material simultaneously, thus the resolution of model and convenient service can be improved in production application.
In order to realize, according to these objects of the present invention and other advantage, providing a kind of method for quickly identifying of homology material, comprise the following steps:
Step 1) gather the spectral signal belonging to several samples of homology material a respectively;
Step 2) pre-service is carried out to the described spectral signal of each sample, obtain one group of corresponding spectroscopic data;
Step 3) according to described spectroscopic data, in qualitative recognition model, set up PLSDA analytical model, obtain each sample at predicted value A corresponding to PLSDA analytical model;
Step 4) solve average value mu and the meansquaredeviationσ of all described predicted value A, determine that the preliminary threshold of this homology material a in PLSDA analytical model is interval by normal distribution 3 σ principle;
Step 5) progressively change major component selected by PLSDA analytical model because of subnumber i, optimize the scope in described preliminary threshold interval, until obtain the highest predicted value threshold interval of recognition correct rate, wherein, i=1 ... 10;
Step 6) arbitrary substance b carry out step 1) and 2) process after, analyze the predicted value B that this arbitrary substance b is corresponding in PLSDA analytical model, if predicted value B is in described predicted value threshold interval, then this arbitrary substance b and described material a is homology material, otherwise does not belong to homology material.
Preferably, described spectral signal can be the one near infrared spectrum and Raman spectrum.
Preferably, described step 2) in, before setting up PLSDA model, described preprocess method comprises: carry out vector normalized and multiplicative scatter correction process to spectral signal.
Preferably, described step 4) in, described preliminary threshold interval is (μ-3 σ, μ+3 σ), and wherein, μ is the mean value of all described predicted value A, and σ is the mean square deviation of all described predicted value A.
Preferably, in the process setting up PLSDA model, " picking one " cross-validation method for preventing model Expired Drugs from occurring is used to verify PLSDA model.
Preferably, described step 5) in, average value mu i and the meansquaredeviationσ i of all described predicted value Ai is solved respectively in the PLSDA analytical model that each described major component is corresponding because of subnumber i, obtain this major component because preliminary threshold interval corresponding to homology sample under subnumber is for (μ i-3 σ i, μ i+3 σ i), wherein, Ai is sample in the major component factor is predicted value corresponding to the PLSDA analytical model of i, μ i is the mean value of all described predicted value Ai, and σ i is the mean square deviation of all described predicted value Ai.
Preferably, the sample of one group of homology material a is chosen at described sample exterior, respectively by this group sample to each described major component because the preliminary threshold interval that subnumber i is corresponding is (μ i-3 σ i, μ i+3 σ i) carry out the checking of recognition correct rate, choose the highest preliminary threshold interval of recognition correct rate (μ i-3 σ i, μ i+3 σ i) corresponding major component because of subnumber i as optimum major component because of subnumber, simultaneously, (μ i-3 σ i, μ i+3 σ i) is the highest predicted value threshold interval of recognition correct rate in this preliminary threshold interval.
The present invention at least comprises following beneficial effect:
1, because PLSDA is data discrete probabilistic analysis in conjunction with spectroscopy discriminance analysis, the data volume of sample is more, and the degree of accuracy of determined threshold value is higher.The present invention focuses on a kind of method proposing new determination PLSDA threshold value, compensate for the deficiency of sample size in experiment to a certain extent, can directly apply to the on-line checkingi of homology Object Classification and adulterated screening, have vast potential for future development.
2, first the present invention uses model of cognition between PLSDA method establishment human blood and animal blood, realizes quick, correct discriminating human blood and animal blood.Although traditional analysis is comparatively accurate, often loaded down with trivial details consuming time, such as HPLS, TLC etc.By comparison, the present invention uses Raman spectrum identification human blood and animal blood to have significantly harmless quick advantage, and contributes to reaching real time on-line monitoring.
3, the present invention's model of cognition between PLSDA method establishment homology material, the discriminating sample of material that realize fast, can't harm.
4, the method for quickly identifying recognition speed of homology material of the present invention is faster, recognition correct rate is higher and be applicable to the identification of various material, and identification range is wider.
Part is embodied by explanation below by other advantage of the present invention, target and feature, part also will by research and practice of the present invention by those skilled in the art is understood.
Accompanying drawing explanation
Fig. 1 is glossy ganoderma and rainbow conk PLSDA discriminance analysis figure;
Fig. 2 is human blood and animal blood PLSDA discriminance analysis figure;
Fig. 3 is the schematic flow sheet of the method for quickly identifying of homology material of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail, can implement according to this with reference to instructions word to make those skilled in the art.
Should be appreciated that used in the present inventionly such as " to have ", other element one or more do not allotted in " comprising " and " comprising " term or the existence of its combination or interpolation.
As shown in figs. 1 and 3, the invention provides a kind of method for quickly identifying of homology material:
Embodiment one:
Select from the glossy ganoderma of Different sources and each 72 samples of the Coriolus Versicolor P.E. collected specimens as spectral signal, the near infrared light spectrum signal adopting German Brooker near infrared spectrometer to gather each sample is respectively analyzed, and spectrometer selected parameter is: wave-number range: 12500-4000cm
-1, resolution: 8cm
-1, the number of times of each Sample Scan: average for 16 times.
Pretreated spectra is carried out to each spectral signal gathered, the noise that can bring containing environment and instrument in original spectral signal, for removing the impact of spectral noise and baseline wander etc., this the present embodiment has carried out multiplicative scatter correction process to spectral signal, thus obtains two groups of spectroscopic datas.
Concrete, according to described spectroscopic data, in qualitative recognition model, set up PLSDA analytical model, according to PLSDA method of discrimination, Ganodenna Lucidum P.E Y=1 is set; Coriolus Versicolor P.E. Y=2, obtain each sample at predicted value A corresponding to PLSDA analytical model, as shown in Figure 1, it is interval that variety classes material has different distribution of forecasting values in same PLSDA analytical model, main innovation of the present invention is to determine fast and accurately that the predicted value of often kind of material in PLSDA analytical model is interval, thus judges the material which kind of unknown sample belongs to and determine.
PLSDA modeling analysis Application comparison in mass data is extensive, can go out the useful information relevant to dependent variable, and set up regression model from bulk redundancy, collinearity extracting data.In the process setting up PLSDA model, adopt the method for validation-cross can prevent the phenomenon of over-fitting to modelling verification.
Delimitation for boundary value: at present, usual way has in 2: (1) on average divides: be exactly the predicted value of the sample of Ganodenna Lucidum P.E between 0.5-1.5, sample belongs to correct distribution; The predicted value of Coriolus Versicolor P.E. sample is between 1.5 to 2.5, and Coriolus Versicolor P.E. sample belongs to correct distribution; (2) rainbow conk and Ganodenna Lucidum P.E spectral signal are mixed, determine a boundary line, as the critical value of classification;
Defect for these 2 kinds of methods: the environment of extraneous growth and interior itself in nature, different to the contribution of the difference between glossy ganoderma and Coriolus Versicolor P.E., also just say that the distribution violent rage of the scatter diagram at the spectral signal through PLSDA model is the same, as shown above, clearly, Coriolus Versicolor P.E. relatively disperses, and Ganodenna Lucidum P.E relatively tightens, if on average divide or determine a boundary, the contribution for glossy ganoderma and Coriolus Versicolor P.E. will be unfair;
In order to solve above-mentioned difficult point: analyze separately the data of sample respectively, in conjunction with normal distribution 3 σ principle of probability, P (μ-3 σ <X≤μ+3 σ)=99.7%, that is exist, through pretreated spectral signal, reach 0.997 by the probability dropped on after PLSDA model in (μ-3 σ, μ+3 σ) interval, just can reach 99.7% in theory to the discrimination of sample, achieve and identify fast and accurately.
For this reason, solve average value mu and the meansquaredeviationσ of all described predicted value A, determine that the preliminary threshold of this homology material a in PLSDA analytical model is interval by normal distribution 3 σ principle, the preliminary threshold namely calculating glossy ganoderma and rainbow conk is respectively interval.
Furtherly, the preliminary threshold interval obtained not is optimum threshold interval, namely the discrimination in this preliminary threshold interval is not the highest, major component now selected by change PLSDA analytical model is because of subnumber i, i=1 ... 10, the scope in preliminary threshold interval described in successive optimization, until obtain the highest predicted value threshold interval of recognition correct rate;
Based on this thought, glossy ganoderma and Coriolus Versicolor P.E. signal are processed respectively and major component because of the selection of subnumber, its accuracy identified is the checking according to external sample, thus obtains best major component because of subnumber:
By the major component chosen for model shown in table one because of accumulation contribution rate corresponding to subnumber, major component because of subnumber be 3 and above time, accumulation contribution rate is 97%, can represent the information content of more than 97% of whole spectrum, therefore, choosing major component because of subnumber is not less than 3.
Table one
| Because of subnumber | Accumulation contribution rate % |
| 1 | 68.51 |
| 2 | 89.01 |
| 3 | 97.33 |
| 4 | 98.26 |
| 5 | 98.96 |
| 6 | 99.19 |
| 7 | 99.36 |
| 8 | 99.47 |
| 9 | 99.55 |
| 10 | 99.63 |
If major component because of subnumber less time, then can omit the quantity of information of spectrum, cause the threshold interval determined because of subnumber according to this major component to reduce the discrimination of material; And number of principal components higher time, then there will be Expired Drugs, the threshold interval determined because of subnumber according to this major component can be caused equally to reduce the discrimination of material, therefore should choose the threshold interval that suitable major component is optimum because subnumber just can calculate, thus improve the recognition success rate to material.Concrete, by major component because of subnumber i be increased to 10 gradually from 1 time, average value mu i and the meansquaredeviationσ i of all described predicted value Ai is solved respectively in the PLSDA analytical model that each described major component is corresponding because of subnumber i, obtain this major component because preliminary threshold interval corresponding to subnumber is for (μ i-3 σ i, μ i+3 σ i), and, the sample of one group of homology material a is chosen at described sample exterior, respectively by this group sample to each described major component because the preliminary threshold interval that subnumber i is corresponding is (μ i-3 σ i, μ i+3 σ i) carry out the checking of recognition correct rate, choose the highest preliminary threshold interval of recognition correct rate (μ i-3 σ i, μ i+3 σ i) corresponding major component because of subnumber i as optimum major component because of subnumber, simultaneously, this preliminary threshold interval (μ i-3 σ i, μ i+3 σ i) be the highest predicted value threshold interval of recognition correct rate, as shown in Table 2.
Table two
| Factor | The threshold interval of glossy ganoderma | The threshold interval of rainbow conk | Accuracy % | Error number |
| 1 | 1.26-1.54 | 1.59-1.70 | 81.67 | 11 |
| 2 | 0.55-1.28 | 1.87-2.35 | 41.67 | 35 |
| 3 | 0.49-1.34 | 1.89-2.31 | 33.33 | 40 |
| 4 | 0.51-1.47 | 1.66-2.37 | 86.67 | 8 |
| 5 | 0.56-1.54 | 1.57-2.34 | 98.33 | 1 |
| 6 | 0.56-1.52 | 1.71-2.22 | 95 | 3 |
| 7 | 0.65-1.42 | 1.64-2.99 | 96.67 | 2 |
| 8 | 0.67-1.38 | 1.71-2.24 | 96.67 | 2 |
| 9 | 0.75-1.29 | 1.73-2.24 | 96.67 | 2 |
| 10 | 0.75-1.28 | 1.77-2.20 | 95 | 3 |
Have table two to draw: when major component is 5 time, accumulation contribution rate is now 98.963%, can represent the information content of 98.963% of whole spectrum.The threshold value of the PLSDA model using the method to set up, the recognition correct rate of its external samples is 98.33%, and identification error quantity is 1, and recognition correct rate is the highest.When enough large of the amount of sample, use the interval threshold of said method defined more will meet actual requirement, the value of more realistic application.Therefore major component elects 5 as because of subnumber, and the predicted value threshold interval that glossy ganoderma is corresponding is (0.56,1.54), and predicted value threshold interval corresponding to rainbow conk is (1.57,2.34).
Spectra collection is carried out to arbitrary substance b and to after spectral signal pre-service, analyze the predicted value B that this arbitrary substance b is corresponding in PLSDA analytical model, if predicted value B is in described predicted value threshold interval, then this arbitrary substance b and described material a is homology material, otherwise does not belong to homology material.That is Ganodenna Lucidum P.E near infrared light spectrum signal is through pre-service and PLSDA model, finally inner in interval (0.56,1.54), and identify correct, this material is glossy ganoderma, on the contrary identification error, and this material is not glossy ganoderma; Coriolus Versicolor P.E. near infrared light spectrum signal is through pre-service and PLSDA model, finally inner in interval (1.57,2.34), and identify correct, this material is rainbow conk, otherwise this material of identification error is not rainbow conk.Further, after establishing predicted value threshold interval corresponding to abundant material, the predicted value threshold interval which kind of material is predicted value corresponding to arbitrary substance b belong to can be analyzed, the concrete species of this material b can be analyzed, complete identifying.
Finally the PLSDA threshold value determined is verified: the determination 48 samples (Ganodenna Lucidum P.E and each 24 of Coriolus Versicolor P.E.) not participating in PLSDA model being come verification model and PLSDA threshold value; Its result is as shown in table three and table four:
Table three
Table four
Shown in table three is Ganodenna Lucidum P.E predicted value in a model, and from table, it is inner that predicted value all drops on interval (0.56,1.54), identifies entirely true.Shown in table four is Coriolus Versicolor P.E. predicted value in a model, and from table, it is inner that predicted value drops on interval (1.57,2.34) completely, identifies correct.By the checking of external data, clearly its spectral signal is final all inner at threshold interval, and namely discrimination reaches 100%.
When enough large of the amount of sample, use the interval threshold of said method defined more will meet actual requirement, the value of more realistic application.
Embodiment two:
As shown in Figures 2 and 3, choose the sample of human blood and animal blood, concrete chooses 40 human bloods (EDTA anti-coagulants) sample, 30 animal blood (EDTA anti-coagulants) sample (dogs 10, rabbit 10, rat 30), 70 samples are for subsequent use altogether.
Use confocal spectroscopic scatterometer, gather the Raman spectrogram of human blood and animal blood; With microslide of aluminizing for substrate, detection Raman shift range is 300 ~ 1700cm
-1, interval 2cm
-1, the time shutter is 1s, and each scanning is averaged for 7 times, by the computer recording absorption intensity Intensity be connected.
Because the sliminess of the blood of animal blood and human blood is different, and both difference is spectrally only showing as the fine difference in some principal ingredients, such as haemoglobin etc.And vector normalized may be used for eliminating the change such as the change of light path or the dilution of sample to the impact of spectrum generation.So before setting up PLSDA model, vector normalized is carried out to spectrum.
After above-mentioned steps, in qualitative recognition model, set up PLSDA analytical model, with animal blood sample for 1, human blood sample is 2, as shown in Figure 2.According to its discrimination, and by the method for embodiment one determine its major component because of subnumber be 3.
After above-mentioned steps, obtain data by PLSDA model, according to normal distribution 3 σ principle, by the threshold range judged of the method determination animal blood of embodiment one as (0.65,1.38), the threshold range judged of human blood is as (1.82,2.13);
16 animal blood samples and 15 human blood samples are carried out the checking of model, find the estimation range (0.91 of animal blood sample, 1.244), the estimation range of human blood is (1.781,2.128), the predicted data of 16 animal blood sample all in correct threshold range, 15 people's blood sample predicted data, there is a data (1.781) not in described threshold range, the Forecasting recognition rate of its model reaches 96.77%.
Can find out in Fig. 2, according to the PLSDA model that spectral signal is set up, obtain human blood different with the distribution of the loose point of animal blood, be evenly distributed and look for a boundary to be all irrational, use above-mentioned theoretical method, in like manner can analyze the recognition threshold interval (0.65,1.38) of the PLSDA determining animal blood sample; The recognition threshold (1.82,2.13) of human blood sample;
Finally verify the PLSDA threshold value determined: will not participate in the human blood (15) of modeling, the predicted value of animal blood sample (rat 10, mouse 3, rabbit 3) is verified, the result as shown in Table 5.
Clearly, the checking of animal blood sample is all inner in threshold interval scope, and human blood sample has 1 sample not inner at interval range, but does not appear in the interval range of the identification of animal blood sample, achieves good recognition effect.
In technique scheme, before gathering blood sample spectrum, sample is shaken up at every turn, gets quantitative sample drop and be added in and aluminize on microslide, each measure before all to aluminize microslide with 75% ethanol purge, thus avoid the cross pollution of sample room.
Use " picking one " cross-validation method to verify PLSDA model, " picking one " cross-validation method refers to only has one group of sample for modeling and this system of inspection to representing multicomponent system to be measured simultaneously; To organize sample from this before starting modeling and remove a sample; This sample is used as testing model; All the other samples are used as the modeling of system.
Table five
The present invention utilizes Reinshaw company inVia confocal micro Raman spectrum to gather raman scattering spectrum, uses Unscrambler9.7 version to carry out the foundation of PLSDA model.
First the present invention uses model of cognition between PLSDA method establishment human blood and animal blood, realizes quick, correct discriminating human blood and animal blood.Although traditional analysis is comparatively accurate, often loaded down with trivial details consuming time, such as HPLS, TLC etc.By comparison, the present invention uses Raman spectrum identification human blood and animal blood to have significantly harmless quick advantage, and contributes to reaching real time on-line monitoring.
The invention has the advantages that: because PLSDA is data discrete probabilistic analysis in conjunction with spectroscopy discriminance analysis, the data volume of sample is more, and the degree of accuracy of determined threshold value is higher.The present invention focuses on a kind of method proposing new determination PLSDA threshold value, compensate for the deficiency of sample size in experiment to a certain extent, simultaneously, the adulterated product not belonging to homology can be detected in the product, directly applying to the on-line checkingi of homology Object Classification and adulterated screening, there is vast potential for future development.
Although embodiment of the present invention are open as above, but it is not restricted to listed in instructions and embodiment utilization, it can be applied to various applicable the field of the invention completely, for those skilled in the art, can easily realize other amendment, therefore do not deviating under the universal that claim and equivalency range limit, the present invention is not limited to specific details and illustrates here and the legend described.
Claims (7)
1. a method for quickly identifying for homology material, is characterized in that, comprises the following steps:
Step 1) gather the spectral signal belonging to several samples of homology material a respectively;
Step 2) pre-service is carried out to the described spectral signal of each sample, obtain one group of corresponding spectroscopic data;
Step 3) according to described spectroscopic data, in qualitative recognition model, set up PLSDA analytical model, obtain each sample at predicted value A corresponding to PLSDA analytical model;
Step 4) solve average value mu and the meansquaredeviationσ of all described predicted value A, determine that the preliminary threshold of this homology material a in PLSDA analytical model is interval by normal distribution 3 σ principle;
Step 5) progressively change major component selected by PLSDA analytical model because of subnumber i, optimize the scope in described preliminary threshold interval, until obtain the highest predicted value threshold interval of recognition correct rate, wherein, i=1 ... 10;
Step 6) arbitrary substance b carry out step 1) and 2) process after, analyze the predicted value B that this arbitrary substance b is corresponding in PLSDA analytical model, if predicted value B is in described predicted value threshold interval, then this arbitrary substance b and described material a is homology material, otherwise does not belong to homology material.
2. the method for quickly identifying of homology material as claimed in claim 1, is characterized in that, described spectral signal can be the one near infrared spectrum and Raman spectrum.
3. the method for quickly identifying of homology material as claimed in claim 2, is characterized in that, described step 2) in, before setting up PLSDA model, described preprocess method comprises: carry out vector normalized and multiplicative scatter correction process to spectral signal.
4. the method for quickly identifying of homology material as claimed in claim 3, is characterized in that, described step 4) in, described preliminary threshold interval is (μ-3 σ, μ+3 σ), wherein, μ is the mean value of all described predicted value A, and σ is the mean square deviation of all described predicted value A.
5. the method for quickly identifying of homology material as claimed in claim 4, is characterized in that, in the process setting up PLSDA model, using " picking one " cross-validation method for preventing model Expired Drugs from occurring to verify PLSDA model.
6. the method for quickly identifying of homology material as claimed in claim 5, it is characterized in that, described step 5) in, average value mu i and the meansquaredeviationσ i of all described predicted value Ai is solved respectively in the PLSDA analytical model that each described major component is corresponding because of subnumber i, obtain this major component because preliminary threshold interval corresponding to homology sample under subnumber is for (μ i-3 σ i, μ i+3 σ i), wherein, Ai is sample in the major component factor is predicted value corresponding to the PLSDA analytical model of i, μ i is the mean value of all described predicted value Ai, σ i is the mean square deviation of all described predicted value Ai.
7. the method for quickly identifying of homology material as claimed in claim 6, it is characterized in that, the sample of one group of homology material a is chosen at described sample exterior, respectively by this group sample to each described major component because the preliminary threshold interval that subnumber i is corresponding is (μ i-3 σ i, μ i+3 σ i) carry out the checking of recognition correct rate, choose the highest preliminary threshold interval of recognition correct rate (μ i-3 σ i, μ i+3 σ i) corresponding major component because of subnumber i as optimum major component because of subnumber, simultaneously, this preliminary threshold interval (μ i-3 σ i, μ i+3 σ i) be the highest predicted value threshold interval of recognition correct rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510654599.5A CN105223140A (en) | 2015-10-10 | 2015-10-10 | The method for quickly identifying of homology material |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510654599.5A CN105223140A (en) | 2015-10-10 | 2015-10-10 | The method for quickly identifying of homology material |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105223140A true CN105223140A (en) | 2016-01-06 |
Family
ID=54992215
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510654599.5A Pending CN105223140A (en) | 2015-10-10 | 2015-10-10 | The method for quickly identifying of homology material |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105223140A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108065911A (en) * | 2016-11-10 | 2018-05-25 | 北京工商大学 | Damp-heat constitution appraisal procedure and assessment system based on skin measurement |
| CN108065910A (en) * | 2016-11-10 | 2018-05-25 | 北京工商大学 | Yin-deficiency constitution appraisal procedure and assessment system based on skin measurement |
| CN108896490A (en) * | 2018-06-06 | 2018-11-27 | 众安信息技术服务有限公司 | Meat piece affinity verification method and device |
| CN110647915A (en) * | 2019-08-23 | 2020-01-03 | 米津锐 | Dynamic mode judgment method for consistency analysis of high-dimensional data |
| CN111624190A (en) * | 2020-06-11 | 2020-09-04 | 复旦大学附属华山医院 | Method for rapidly identifying bacteria and fungi by using Raman spectrum |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937079A (en) * | 2010-06-29 | 2011-01-05 | 中国农业大学 | Remote sensing image variation detection method based on region similarity |
| CN102103132A (en) * | 2009-12-18 | 2011-06-22 | 中国科学院大连化学物理研究所 | Method for screening diabetes markers from body fluid metabonome profile |
| CN104132896A (en) * | 2014-07-15 | 2014-11-05 | 山东国家农产品现代物流工程技术研究中心 | Method for rapidly identifying adulterated meat |
| CN104316489A (en) * | 2014-08-18 | 2015-01-28 | 浙江百山祖生物科技有限公司 | Method of detecting adulteration of ganoderma lucidum extract product by near infrared spectroscopy |
-
2015
- 2015-10-10 CN CN201510654599.5A patent/CN105223140A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103132A (en) * | 2009-12-18 | 2011-06-22 | 中国科学院大连化学物理研究所 | Method for screening diabetes markers from body fluid metabonome profile |
| CN101937079A (en) * | 2010-06-29 | 2011-01-05 | 中国农业大学 | Remote sensing image variation detection method based on region similarity |
| CN104132896A (en) * | 2014-07-15 | 2014-11-05 | 山东国家农产品现代物流工程技术研究中心 | Method for rapidly identifying adulterated meat |
| CN104316489A (en) * | 2014-08-18 | 2015-01-28 | 浙江百山祖生物科技有限公司 | Method of detecting adulteration of ganoderma lucidum extract product by near infrared spectroscopy |
Non-Patent Citations (3)
| Title |
|---|
| 刘绍勇等: "基于近红外光谱技术的连翘药材质量控制方法研究", 《药物分析杂志》 * |
| 邵平等: "近红外漫反射光谱技术快速无损识别灵芝和云芝提取物研究", 《核农学报》 * |
| 韩海帆等: "NIRS 法快速测定复方阿胶浆中总黄酮、总皂苷和可溶性固形物", 《中草药》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108065911A (en) * | 2016-11-10 | 2018-05-25 | 北京工商大学 | Damp-heat constitution appraisal procedure and assessment system based on skin measurement |
| CN108065910A (en) * | 2016-11-10 | 2018-05-25 | 北京工商大学 | Yin-deficiency constitution appraisal procedure and assessment system based on skin measurement |
| CN108896490A (en) * | 2018-06-06 | 2018-11-27 | 众安信息技术服务有限公司 | Meat piece affinity verification method and device |
| CN110647915A (en) * | 2019-08-23 | 2020-01-03 | 米津锐 | Dynamic mode judgment method for consistency analysis of high-dimensional data |
| CN111624190A (en) * | 2020-06-11 | 2020-09-04 | 复旦大学附属华山医院 | Method for rapidly identifying bacteria and fungi by using Raman spectrum |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105223140A (en) | The method for quickly identifying of homology material | |
| CN102590129B (en) | Method for detecting content of amino acid in peanuts by near infrared method | |
| CN102879353B (en) | The method of content of protein components near infrared detection peanut | |
| CN104897607A (en) | Food modeling and rapid detecting integration method and system adopting portable NIRS (near infrared spectroscopy) | |
| CN104062258B (en) | Method for rapid determination of soluble solids in compound ass-hide glue pulp by near infrared spectroscopy | |
| CN105630743A (en) | Spectrum wave number selection method | |
| CN108844941B (en) | Method for identifying and classifying different-grade phosphate ores based on Raman spectrum and PCA-HCA | |
| CN104062262A (en) | Crop seed variety authenticity identification method based on near infrared spectrum | |
| TW202223750A (en) | Device and method for avoiding false positive identification during spectroscopic quantification and relevant non-transitory computer-readable medium | |
| CN103674884A (en) | Random forest classification method for tobacco leaf style characteristics based on near infrared spectral information | |
| CN109211829A (en) | A method of moisture content in the near infrared spectroscopy measurement rice based on SiPLS | |
| CN105717066A (en) | Near-infrared spectrum recognition model based on weighting association coefficients | |
| CN103411906A (en) | Near infrared spectrum qualitative identification method of pearl powder and shell powder | |
| Basile et al. | NIR analysis of intact grape berries: Chemical and physical properties prediction using multivariate analysis | |
| WO2020248961A1 (en) | Method for selecting spectral wavenumber without reference value | |
| Jiang et al. | Qualitative and quantitative analysis in solid-state fermentation of protein feed by FT-NIR spectroscopy integrated with multivariate data analysis | |
| CN102937575B (en) | Watermelon sugar degree rapid modeling method based on secondary spectrum recombination | |
| CN104596975A (en) | Method for measuring lignin of reconstituted tobacco by paper-making process by virtue of near infrared reflectance spectroscopy technique | |
| CN103743705A (en) | Rapid detection method for sorghum halepense and similar species | |
| CN105738311A (en) | Apple sweetness non-damage quick detection method based on near-infrared spectrum technology | |
| CN104316492A (en) | Method for near-infrared spectrum measurement of protein content in potato tuber | |
| CN112651173B (en) | Agricultural product quality nondestructive testing method based on cross-domain spectral information and generalizable system | |
| CN108663334B (en) | Method for searching spectral characteristic wavelength of soil nutrient based on multi-classifier fusion | |
| CN114611582A (en) | Method and system for analyzing substance concentration based on near infrared spectrum technology | |
| CN108827909B (en) | Rapid soil classification method based on visible near infrared spectrum and multi-target fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160106 |