CN109507123A - A kind of discrimination method of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its similar product - Google Patents
A kind of discrimination method of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its similar product Download PDFInfo
- Publication number
- CN109507123A CN109507123A CN201811396730.2A CN201811396730A CN109507123A CN 109507123 A CN109507123 A CN 109507123A CN 201811396730 A CN201811396730 A CN 201811396730A CN 109507123 A CN109507123 A CN 109507123A
- Authority
- CN
- China
- Prior art keywords
- node
- hidden layer
- pattern recognition
- radix angelicae
- angelicae sinensis
- 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
- 239000000126 substance Substances 0.000 title claims abstract description 28
- 238000001228 spectrum Methods 0.000 title claims abstract description 16
- 238000003909 pattern recognition Methods 0.000 title claims abstract description 13
- 238000012850 discrimination method Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 31
- 230000005284 excitation Effects 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000012567 pattern recognition method Methods 0.000 claims abstract description 9
- 238000002371 ultraviolet--visible spectrum Methods 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 21
- 238000005457 optimization Methods 0.000 claims description 14
- 239000000463 material Substances 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 3
- 229920003023 plastic Polymers 0.000 claims description 3
- 239000004033 plastic Substances 0.000 claims description 3
- 239000000843 powder Substances 0.000 claims description 3
- 238000002310 reflectometry Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract 1
- 239000003814 drug Substances 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 6
- 241000125175 Angelica Species 0.000 description 4
- 235000001287 Guettarda speciosa Nutrition 0.000 description 3
- 229940079593 drug Drugs 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000010453 quartz Substances 0.000 description 2
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 2
- 241000208173 Apiaceae Species 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 238000001157 Fourier transform infrared spectrum Methods 0.000 description 1
- 235000003143 Panax notoginseng Nutrition 0.000 description 1
- 241000180649 Panax notoginseng Species 0.000 description 1
- 241000121220 Tricholoma matsutake Species 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000000149 argon plasma sintering Methods 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000796 flavoring agent Substances 0.000 description 1
- 235000019634 flavors Nutrition 0.000 description 1
- 210000004392 genitalia Anatomy 0.000 description 1
- 235000008216 herbs Nutrition 0.000 description 1
- 230000036039 immunity Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000004089 microcirculation Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 235000019615 sensations Nutrition 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000012306 spectroscopic technique Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 238000000870 ultraviolet spectroscopy Methods 0.000 description 1
- 238000001429 visible spectrum 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
- G01N21/33—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using ultraviolet 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
- G01N2201/1296—Using chemometrical methods using neural networks
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 present invention relates to a kind of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its discrimination methods of similar product.Specific steps are as follows: the Radix Angelicae Sinensis and its similar product for collecting certain amount acquire the uv-vis spectra of sample；Sample is divided into training set and forecast set；Respectively in the class of HCA and between class distance, PLS-DA are optimized because of subnumber, the sized method of ANN and node in hidden layer, the excitation function of ELM and node in hidden layer；Under optimal parameter, four kinds of Chemical Pattern Recognition methods are investigated to the prediction accuracy of sample in forecast set, obtain optimal Chemical Pattern Recognition method.The present invention is based on UV-vis DRS spectrum and Chemical Pattern Recognition, rapidly, green is lossless for detection.The present invention is suitable for the identification of Radix Angelicae Sinensis and its similar product.
Description
Technical field
The invention belongs to the drug quality control technologies in Analysis of Chinese Traditional Medicine field, and in particular to one kind is unrestrained anti-based on UV, visible light
Penetrate the Radix Angelicae Sinensis of spectrum and Chemical Pattern Recognition and its discrimination method of similar product.
Background technique
Radix Angelicae Sinensis is the root by umbelliferae angelica by sunning, dry gained, and the western Radix Angelicae Sinensis of alias, cloud are returned, main to plant
Train the cold sensation of the genitalia humid region high in height above sea level 1800-2500 meters of China, Minxian County, Gansu and Yunnan Province.Radix Angelicae Sinensis has warm-natured, acrid-sweet flavor
Characteristic, there are the therapeutic efficiencies such as activating microcirculation and removing stasis medicinal, strengthen immunity in clinic.In recent years, the demand of Radix Angelicae Sinensis medicinal material existed
It ramps.Due to the rise at full speed of the market price, it is all umbelliferae that some unprincipled fellows, which are driven by interests and are often used,
The less expensive drug of prices such as Radix Angelicae Pubescentis, Rhizoma Chuanxiong mix the spurious with the genuine and peddled.If adulterant is used in the treatment, it is likely that meeting
Delay best occasion for the treatment even results in aggravation or deterioration, it can be seen that has for the identification of the Chinese medicine true and false important
Meaning.
The discrimination method of Chinese medicine is divided into tradition identification, physical and chemical identification, chromatography and spectroscopic methodology etc..Traditional discrimination method packet
Character identification and microscopical characters are included, this, which requires operator, practical experience abundant and deep theoretical knowledge.Physics and chemistry identifies
It is the true and false, purity and quality for identifying Chinese medicine using specific physics, chemistry and instrument analytical method.Ultraviolet visible spectrometry
According to the number, intensity of absorption peak in spectrogram to the difference between the molecular formula ingredient and content with degree of unsaturation in medicinal material
Carry out quality discrimination.From ultraviolet spectrogram the difference of the number and location of absorption peak can analyze Identification chinese herbs medicine material (Shi Youming,
Liu Gang, FTIR spectrum combination hierarchial-cluster analysis identify the research of matsutake and Agricus blazei, light scattering journal, 2010).But due to purple
Outer visible spectrum overlap of peaks is serious, can not directly be identified according to the position of spectral peak, it is therefore desirable to by Chemical Pattern Recognition
Technology (Bian Xihui, Pang Xuelu, Zhang Caixia, Guo Yugao, Zhou Yanmei, it is a kind of based on uv-vis spectra and Chemical Pattern Recognition
Radix Notoginseng and its adulterant discrimination method, Chinese invention patent, 2017, CN201710569315.1).
Chemical Pattern Recognition obtains existing inner link between sample by the common trait analyzed between different samples
And useful information carries out Classification and Identification, can improve accuracy (Gu Huiwen, the high-order instrument combination chemistry of analysis most possibly
Local method is used for complex system Quantitative Study [D], Hunan University, 2016).Chemical simulation identification includes Hierarchical Clustering point
Analyse (HCA), offset minimum binary-linear discriminant analysis (PLS-DA), artificial neural network (ANN), extreme learning machine (ELM) etc..
Summary of the invention
The purpose of the present invention is in view of the above problems, using UV-vis DRS spectrum as means of testing, first
The data set measured is used these four chemical models of HCA, PLS-DA, ANN and ELM by the spectrum for scanning Radix Angelicae Sinensis and its similar product
Recognition methods is identified.
The technical scheme comprises the following steps provided by realize the present invention:
1) sample is collected
Radix Angelicae Sinensis is bought from more pharmacies and its similar product are several, processing is numbered in medicinal material first, by sample through powder
After broken machine is polished into powder, 120 meshes are crossed, are put into the tool lid dry plastics bottle of the 50mL for the number of finishing with to be detected.
2) uv-vis spectra of sample is acquired
Acquiring wave-length coverage required for sample is 200-800nm, and sampling interval 0.5nm, scanning speed is high speed, is swept
The mode of retouching is single.Mensuration mode is reflectivity, and slit width is 5.0nm, and the time of integration is 0.1 second, and light source Wavelength-converting is
323.0nm, detector cell are external single detector, and S/R is converted to standard, close ladder correction.After instrument self checking is errorless,
Quartz c ell is put into progress reference line sweep measuring in the card slot of ultraviolet-uisible spectrophotometer.It later will be to be measured with small medicine spoon
Sample is put into sample cell, and sample, which is smeared covering, uniformly causes the bottom of sample cell not interspace, and sample cell is put into
Sample Scan measurement is carried out among the card slot of ultraviolet-uisible spectrophotometer.Sample cell is turned an angle to be subsequently placed into card slot
Second of scanning optical spectrum.Using the average value measured three times as the final spectrum of the sample.
3) sample is divided into training set and forecast set
KS method is used to divide respectively in every class sample, 3/4 sample is used as training, and 1/4 sample is used as prediction.By all classes
Other training data merges into total training set, and prediction data merges into total forecast set.
4) parameter optimization of Chemical Pattern Recognition method
HCA parameter optimization: inter- object distance successively use euclidean, seuclidean, cityblock, minkowski,
Chebychev, mahalanobis, cosine, correlation, spearman, between class distance successively use single,
Complete, average, weighted, centroid, median, ward, calculate in inhomogeneity and HCA is poly- under between class distance
The spearman related coefficient of class, the corresponding inter- object distance of spearman related coefficient maximum value and between class distance are optimal class
Interior and between class distance.
PLS-DA parameter optimization: optimum factor number is exactly that will be divided into 1 because subnumber is from 1 to 25, is calculated different because under subnumber
Prediction accuracy, prediction accuracy for the first time occur corresponding to peak because subnumber be it is optimal because of subnumber.
ANN parameter optimization: centralization, standardization, minimax normalization, hidden layer node is respectively adopted in sized method
Number is divided into 1 from 1-100, calculates the RMSECV that ANN is modeled under different scale and node in hidden layer, RMSECV minimum value
Corresponding sized method and node in hidden layer are best scale method and node in hidden layer.
ELM parameter optimization: excitation function takes sig, sin, hardlim, tribas, radbas, node in hidden layer respectively
From 1-1000, being divided into 1 is changed, and calculates the prediction accuracy that ELM is modeled under different excitation functions and node in hidden layer,
It repeats 50 times, obtains different excitation functions and node in hidden layer ELM models the average value of 50 prediction accuracy.Prediction is correct
The corresponding excitation function of the maximum value that rate average value reaches at first and node in hidden layer are the optimal parameter of ELM modeling.
5) comparison of different Chemical Pattern Recognition method identification results
To forecast set sample, using in optimal class and between class distance, HCA clustering is carried out；Using optimum factor number,
Establish PLS-DA model；ANN model is established using best scale method and node in hidden layer；Using optimal excitation function
And node in hidden layer establishes ELM model.Compare the prediction accuracy of four kinds of modeling methods, prediction accuracy maximum value is corresponding
Method is optimum chemical mode identification method.
Advantage of the invention is that UV-Vis spectroscopic techniques are easy to operate, analysis speed is fast, sample nondestructive, chemical model
Recognition methods is quick, intelligent.
Detailed description of the invention
Fig. 1 is the ultraviolet-visible spectrogram of Radix Angelicae Sinensis and its similar product
Fig. 2 be Radix Angelicae Sinensis and its similar product data set PLS-DA modeling prediction accuracy with the variation diagram because of subnumber
Fig. 3 is the prediction accuracy average value of 50 ELM of Radix Angelicae Sinensis and its similar product data set modeling with hidden layer node
Several and excitation function variation diagram
Specific embodiment
To be best understood from the present invention, the present invention will be described in further detail with reference to the following examples, but of the invention
Claimed range is not limited to range represented by embodiment.
Embodiment
1) sample is collected
39 parts of 35 parts of 38 parts of 39 parts of 40 parts of Radix Angelicae Sinensis, Radix Angelicae Pubescentis, Rhizoma Chuanxiong, Rhizoma Atractylodis Macrocephalae, the root of Dahurain angelica bought from the pharmacy of 40, Tianjin
Totally 191 parts of samples.Processing is numbered in sample first, the number of Radix Angelicae Sinensis is 1-40, and the number of Radix Angelicae Pubescentis is 41-79, Rhizoma Chuanxiong
Number is 80-117, and the number of Rhizoma Atractylodis Macrocephalae is 118-152, and the number of the root of Dahurain angelica is 153-191.191 parts of samples are polished through pulverizer
Cheng Fenhou crosses 120 meshes, is put into the tool lid dry plastics bottle of the 50mL for the number of finishing with to be detected.
2) uv-vis spectra of sample is acquired
Before experiment, first opens UV-2700 ultraviolet-uisible spectrophotometer (Japanese Shimadzu Corporation) and carry out preheating 25 minutes.In advance
The UV-probe2.34 software opened in computer after heat is configured acquisition sample institute to instrument parameter required for measuring
The wave-length coverage needed is 200-800nm, and scanning speed is high speed, and sampling interval 0.5nm, scan pattern is single.Point is opened
Instrument attribute project column, mensuration mode are reflectivity, and slit width is 5.0nm, and the time of integration is 0.1 second, and light source Wavelength-converting is
323.0nm, detector cell are external single detector, and S/R is converted to standard, close ladder correction.After instrument self checking is errorless,
Quartz c ell is put into progress reference line sweep measuring in the card slot of ultraviolet-uisible spectrophotometer.It later will be to be measured with small medicine spoon
Sample is put into sample cell, and sample, which is smeared covering, uniformly causes the bottom of sample cell not interspace, and sample cell is put into
Sample Scan measurement is carried out among the card slot of ultraviolet-uisible spectrophotometer.Sample cell is turned an angle to be subsequently placed into card slot
Second of scanning optical spectrum.Using the average value measured three times as the final spectrum of the sample.Fig. 1 is the purple of Radix Angelicae Sinensis and its similar product
Outer visible light spectrogram.
3) sample is divided into training set and forecast set
KS method is used to divide respectively in every class sample, wherein 30 samples of Radix Angelicae Sinensis are used as training set, and 10 samples are used as prediction
Collection；29 Radix Angelicae Pubescentis, root of Dahurain angelica samples are used as training set, and 10 samples are used as forecast set；28 samples of Rhizoma Chuanxiong be used as training set, 10
Sample is used as forecast set；26 samples of Rhizoma Atractylodis Macrocephalae are used as training set, and 9 samples are used as forecast set.By the training of five class traditional Chinese medicine samples
Data merge into total training set totally 142 samples, and prediction data merges into total forecast set totally 49 samples.
4) parameter optimization
The optimization of HCA parameter: inter- object distance successively use euclidean, seuclidean, cityblock,
Minkowski, chebychev, mahalanobis, cosine, correlation, spearman, between class distance successively use
Single, complete, average, weighted, centroid, median, ward, calculate inhomogeneity in and between class distance
The spearman related coefficient of lower HCA cluster, the corresponding inter- object distance of spearman related coefficient maximum value and between class distance are
In optimal class and between class distance.Table 1 shows the present embodiment difference inter- object distance and the obtained spearman of between class distance
Related coefficient, it can be seen that inter- object distance uses chebychev, and between class distance uses the spearman phase relation of centroid
Number can achieve 0.8519, be maximum value.Therefore using chebychev and centroid as best inter- object distance and class spacing
From.
The different inter- object distances of table 1 and the obtained spearman correlation coefficient charts of between class distance
PLS-DA parameter optimization: optimum factor number is exactly will be because subnumber be from 1 to 25, and being divided into 1 is changed, and calculates different
Because of the prediction accuracy under subnumber, predict that accuracy occurs corresponding to peak for the first time because subnumber is optimal because of subnumber.
Fig. 2 is shown because prediction accuracy reaches maximum value 100% at first when subnumber is 7, therefore is chosen because subnumber 7 is used as optimum factor
Number.
ANN parameter optimization: centralization, standardization, minimax normalization, hidden layer node is respectively adopted in sized method
From 1-100, be divided into 1 is changed number, calculates the RMSECV that ANN is modeled under different scale and node in hidden layer,
The corresponding sized method of RMSECV minimum value and node in hidden layer are best scale method and node in hidden layer.This reality
Optimal node number and optimum prediction accuracy are applied when example uses centralization as 4 and 26.53%；Using optimal node number when standardization
It is 3 and 81.63% with optimum prediction accuracy；When minimax being used to normalize optimal node number and optimum prediction accuracy for
5 and 46.94%.Therefore best scale method and best node in hidden layer are used as using standardization and 3.
The optimization of ELM parameter: excitation function takes sig, sin, hardlim, tribas, radbas, hidden layer node respectively
From 1-1000, be divided into 1 is changed number, and it is correct to calculate the prediction that ELM is modeled under different excitation functions and node in hidden layer
Rate repeats 50 times, obtains different excitation functions and node in hidden layer ELM models the average value of 50 prediction accuracy.Prediction
The corresponding excitation function of the maximum value that accuracy average value reaches at first and node in hidden layer are the optimal parameter of ELM modeling.
Fig. 3 shows the prediction accuracy average value of 50 ELM modeling with the variation of activation primitive and node in hidden layer.From figure
As can be seen that the corresponding excitation function of maximum value and node in hidden layer that prediction accuracy average value reaches at first are respectively
Sin function and 125, therefore Optimum Excitation function and best node in hidden layer are used as using sin function and 125.
5) comparison of different Chemical Pattern Recognition method identification results
Using optimal inter- object distance chebychev, between class distance centroid, HCA clustering is carried out；Using best
Because of subnumber 7, PLS-DA model is established；ANN model is established using best scale methodological standardization and node in hidden layer 3；It adopts
ELM model is established with Optimum Excitation function sin and node in hidden layer 125.Table 2 shows four kinds of Chemical Pattern Recognition methods
Predict accuracy, the prediction corresponding method PLS-DA of accuracy maximum value 100% is optimum chemical mode identification method.
Prediction accuracy table of the different Chemical Pattern Recognition methods of table 2 to Radix Angelicae Sinensis and its similar product
Claims (5)
1. the discrimination method of a kind of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its similar product, special
Sign is: collecting the Radix Angelicae Sinensis and its similar product of certain amount, by medicinal material after pulverizer is polished into powder, crosses 120 meshes, be put into volume
In the tool lid dry plastics bottle of good number 50mL；The wave-length coverage of ultraviolet-visual spectrometer device is 200-800nm, and scanning speed is
At a high speed, sampling interval 0.5nm, scan pattern be it is single, mensuration mode is reflectivity, acquires the uv-vis spectra of sample；
KS method is used to divide respectively in every class sample, 3/4 sample is used as training, and 1/4 sample is used as prediction, by the training of all categories
Data merge into total training set, and prediction data merges into total forecast set；Respectively in the class of HCA and between class distance, PLS-
DA's optimizes because of subnumber, the sized method of ANN and node in hidden layer, the excitation function of ELM and node in hidden layer；
Under optimal parameter, the prediction accuracy of four kinds of Chemical Pattern Recognition methods is investigated, obtains optimal Chemical Pattern Recognition method.
2. a kind of Radix Angelicae Sinensis and its phase based on UV-vis DRS spectrum and Chemical Pattern Recognition according to claim 1
Like the discrimination method of product, it is characterised in that: in the HCA class and between class distance optimization method are as follows: inter- object distance successively uses
euclidean、seuclidean、cityblock、minkowski、chebychev、mahalanobis、cosine、
Correlation, spearman, between class distance successively use single, complete, average, weighted,
Centroid, median, ward calculate the spearman related coefficient that in inhomogeneity and HCA is clustered under between class distance,
The corresponding inter- object distance of spearman related coefficient maximum value and between class distance are in optimal class and between class distance.
3. a kind of Radix Angelicae Sinensis and its phase based on UV-vis DRS spectrum and Chemical Pattern Recognition according to claim 1
Like the discrimination method of product, it is characterised in that: the PLS-DA is because of subnumber optimization method are as follows: optimum factor number is exactly will be because of subnumber
From 1 to 25, being divided into 1 is changed, and calculates the different prediction accuracy because under subnumber, and highest occurs for the first time in prediction accuracy
Because subnumber is optimal because of subnumber corresponding to value.
4. a kind of Radix Angelicae Sinensis and its phase based on UV-vis DRS spectrum and Chemical Pattern Recognition according to claim 1
Like the discrimination method of product, it is characterised in that: the sized method of the ANN and node in hidden layer optimization method are as follows: sized
Centralization, standardization, minimax normalization is respectively adopted in method, and from 1-100, be divided into 1 is changed node in hidden layer,
Calculate the RMSECV that ANN is modeled under different scale and node in hidden layer, the corresponding sized method of RMSECV minimum value and
Node in hidden layer is best scale method and node in hidden layer.
5. a kind of Radix Angelicae Sinensis and its phase based on UV-vis DRS spectrum and Chemical Pattern Recognition according to claim 1
Like the discrimination method of product, it is characterised in that: the ELM excitation function and node in hidden layer optimization method are as follows: excitation function
Take sig, sin, hardlim, tribas, radbas respectively, from 1-1000, be divided into 1 is changed node in hidden layer, calculates
The prediction accuracy that ELM is modeled under different excitation functions and node in hidden layer repeats 50 times, obtains different excitation functions and hidden
The ELM of number containing node layer models the average value of 50 prediction accuracy.The maximum value that prediction accuracy average value reaches at first is corresponding
Excitation function and node in hidden layer be ELM modeling optimal parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811396730.2A CN109507123A (en) | 2018-11-22 | 2018-11-22 | A kind of discrimination method of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its similar product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811396730.2A CN109507123A (en) | 2018-11-22 | 2018-11-22 | A kind of discrimination method of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its similar product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109507123A true CN109507123A (en) | 2019-03-22 |
Family
ID=65749556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811396730.2A Pending CN109507123A (en) | 2018-11-22 | 2018-11-22 | A kind of discrimination method of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its similar product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109507123A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110514611A (en) * | 2019-09-25 | 2019-11-29 | 深圳市药品检验研究院（深圳市医疗器械检测中心） | A kind of Chemical Pattern Recognition method for establishing evaluation traditional Chinese medicine quality based on drug effect information |
| CN111257263A (en) * | 2020-03-04 | 2020-06-09 | 浙江方圆检测集团股份有限公司 | Method for identifying chemical composition and structural morphology consistency of pearl surface layer by using ultraviolet-visible diffuse reflection spectrum |
| CN116010861A (en) * | 2022-12-13 | 2023-04-25 | 淮阴工学院 | Classification method of shape-like traditional Chinese medicines |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104345045A (en) * | 2014-11-04 | 2015-02-11 | 天津工业大学 | Chemical pattern recognition and near infrared spectrum-based similar medicinal material identification method |
| CN105158202A (en) * | 2015-09-07 | 2015-12-16 | 天津工业大学 | Quick and nondestructive sesame oil brand identification method |
| CN107132194A (en) * | 2017-07-10 | 2017-09-05 | 天津工业大学 | A kind of pseudo-ginseng and its adulterant discrimination method based on uv-vis spectra and Chemical Pattern Recognition |
| CN107290304A (en) * | 2017-07-10 | 2017-10-24 | 天津工业大学 | It is a kind of to pseudo-ginseng and its method for quick identification of adulterant |
| CN107703095A (en) * | 2017-09-27 | 2018-02-16 | 天津工业大学 | A kind of binary based near infrared spectrum and Chemical Measurement mixes pseudo- pseudo-ginseng quantitative analysis method |
| CN107727609A (en) * | 2017-09-27 | 2018-02-23 | 天津工业大学 | A kind of polynary based near infrared spectrum mixes pseudo- pseudo-ginseng quantitative analysis method |
-
2018
- 2018-11-22 CN CN201811396730.2A patent/CN109507123A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104345045A (en) * | 2014-11-04 | 2015-02-11 | 天津工业大学 | Chemical pattern recognition and near infrared spectrum-based similar medicinal material identification method |
| CN105158202A (en) * | 2015-09-07 | 2015-12-16 | 天津工业大学 | Quick and nondestructive sesame oil brand identification method |
| CN107132194A (en) * | 2017-07-10 | 2017-09-05 | 天津工业大学 | A kind of pseudo-ginseng and its adulterant discrimination method based on uv-vis spectra and Chemical Pattern Recognition |
| CN107290304A (en) * | 2017-07-10 | 2017-10-24 | 天津工业大学 | It is a kind of to pseudo-ginseng and its method for quick identification of adulterant |
| CN107703095A (en) * | 2017-09-27 | 2018-02-16 | 天津工业大学 | A kind of binary based near infrared spectrum and Chemical Measurement mixes pseudo- pseudo-ginseng quantitative analysis method |
| CN107727609A (en) * | 2017-09-27 | 2018-02-23 | 天津工业大学 | A kind of polynary based near infrared spectrum mixes pseudo- pseudo-ginseng quantitative analysis method |
Non-Patent Citations (3)
| Title |
|---|
| WENBIN ZHENG ET AL.: "Spectroscopy-based food classification with extreme learning machine", 《CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS》 * |
| 吴柄桦 等: "紫外-可见-短波近红外光谱技术快速鉴别中药材的研究", 《2014年中国药学大会暨第十四届中国药师周论文集》 * |
| 陶伟 等: "近红外漫反射方法在中药研究领域中的应用", 《实用医学杂志》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110514611A (en) * | 2019-09-25 | 2019-11-29 | 深圳市药品检验研究院（深圳市医疗器械检测中心） | A kind of Chemical Pattern Recognition method for establishing evaluation traditional Chinese medicine quality based on drug effect information |
| US11710541B2 (en) | 2019-09-25 | 2023-07-25 | Shenzhen Institute For Drug Control (Shenzhen Testing Center Of Medical Devices) | Chemical pattern recognition method for evaluating quality of traditional Chinese medicine based on medicine effect information |
| CN111257263A (en) * | 2020-03-04 | 2020-06-09 | 浙江方圆检测集团股份有限公司 | Method for identifying chemical composition and structural morphology consistency of pearl surface layer by using ultraviolet-visible diffuse reflection spectrum |
| CN116010861A (en) * | 2022-12-13 | 2023-04-25 | 淮阴工学院 | Classification method of shape-like traditional Chinese medicines |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107677647B (en) | Method for identifying origin of traditional Chinese medicinal materials based on principal component analysis and BP neural network | |
| Goodarzi et al. | Similarity analyses of chromatographic herbal fingerprints: a review | |
| CN101532954B (en) | Method for identifying traditional Chinese medicinal materials by combining infra-red spectra with cluster analysis | |
| CN109507123A (en) | A kind of discrimination method of Radix Angelicae Sinensis based on UV-vis DRS spectrum and Chemical Pattern Recognition and its similar product | |
| Alaerts et al. | Recent developments in chromatographic fingerprints from herbal products: set-up and data analysis | |
| CN110514611A (en) | A kind of Chemical Pattern Recognition method for establishing evaluation traditional Chinese medicine quality based on drug effect information | |
| CN106841083A (en) | Sesame oil quality detecting method based on near-infrared spectrum technique | |
| CN105181650B (en) | A method of quickly differentiating local tea variety using near-infrared spectrum technique | |
| CN107478595B (en) | Method for rapidly identifying authenticity of pearl powder and quantitatively predicting content of adulterated shell powder | |
| CN106408012A (en) | Tea infrared spectrum classification method of fuzzy discrimination clustering | |
| CN104062262A (en) | Crop seed variety authenticity identification method based on near infrared spectrum | |
| CN102087212A (en) | Pueraria lobata starch adulteration identification method based on principal component analysis | |
| Dong et al. | Deep learning for geographical discrimination of Panax notoginseng with directly near-infrared spectra image | |
| Chen et al. | Rapid and automatic chemical identification of the medicinal flower buds of Lonicera plants by the benchtop and hand-held Fourier transform infrared spectroscopy | |
| CN106092990A (en) | A kind of three-dimensional fluorescence spectrum discrimination method of lycium barbarum | |
| CN102243170A (en) | Method for identifying production place of dwarf lilyturf root by using near infrared spectrum technology | |
| CN107132194A (en) | A kind of pseudo-ginseng and its adulterant discrimination method based on uv-vis spectra and Chemical Pattern Recognition | |
| CN108760677A (en) | A kind of rhizoma pinellinae praeparata based on near-infrared spectrum technique mixes pseudo- discrimination method | |
| Dong et al. | A chemometric modeling-free near infrared barcode strategy for smart authentication and geographical origin discrimination of Chinese ginseng | |
| CN103411895B (en) | Pseudo-near infrared spectrum identification method mixed by pearl powder | |
| Liu et al. | Authentication of herbal medicines based on modern analytical technology combined with chemometrics approach: A review | |
| Chen et al. | Rapid identification of three varieties of Chrysanthemum with near infrared spectroscopy | |
| Zhang et al. | Recent trends of machine learning applied to multi-source data of medicinal plants | |
| CN107290304A (en) | It is a kind of to pseudo-ginseng and its method for quick identification of adulterant | |
| Wulandari et al. | Determination of total flavonoid content in medicinal plant leaves powder using infrared spectroscopy and chemometrics |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190322 |