CN114527088A - Folium artemisiae argyi producing area tracing method based on infrared spectrum fingerprint technology - Google Patents
Folium artemisiae argyi producing area tracing method based on infrared spectrum fingerprint technology Download PDFInfo
- Publication number
- CN114527088A CN114527088A CN202210268408.1A CN202210268408A CN114527088A CN 114527088 A CN114527088 A CN 114527088A CN 202210268408 A CN202210268408 A CN 202210268408A CN 114527088 A CN114527088 A CN 114527088A
- Authority
- CN
- China
- Prior art keywords
- folium artemisiae
- artemisiae argyi
- infrared spectrum
- infrared
- samples
- 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
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000005516 engineering process Methods 0.000 title claims abstract description 27
- 238000001228 spectrum Methods 0.000 claims abstract description 29
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 238000005033 Fourier transform infrared spectroscopy Methods 0.000 claims abstract description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 45
- 238000012360 testing method Methods 0.000 claims description 43
- 238000010521 absorption reaction Methods 0.000 claims description 36
- 230000000694 effects Effects 0.000 claims description 29
- 238000012567 pattern recognition method Methods 0.000 claims description 26
- 238000005452 bending Methods 0.000 claims description 21
- 238000004519 manufacturing process Methods 0.000 claims description 13
- 150000001335 aliphatic alkanes Chemical class 0.000 claims description 12
- 150000001298 alcohols Chemical class 0.000 claims description 9
- 150000001412 amines Chemical class 0.000 claims description 9
- 238000002156 mixing Methods 0.000 claims description 9
- 150000002989 phenols Chemical class 0.000 claims description 9
- 238000007637 random forest analysis Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 150000001299 aldehydes Chemical class 0.000 claims description 6
- 150000001336 alkenes Chemical class 0.000 claims description 6
- 150000002148 esters Chemical class 0.000 claims description 6
- 238000000227 grinding Methods 0.000 claims description 6
- UHOVQNZJYSORNB-UHFFFAOYSA-N monobenzene Natural products C1=CC=CC=C1 UHOVQNZJYSORNB-UHFFFAOYSA-N 0.000 claims description 6
- 238000001035 drying Methods 0.000 claims description 4
- 230000002068 genetic effect Effects 0.000 claims description 4
- JRZJOMJEPLMPRA-UHFFFAOYSA-N olefin Natural products CCCCCCCC=C JRZJOMJEPLMPRA-UHFFFAOYSA-N 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 230000003595 spectral effect Effects 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 4
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims description 3
- ISWSIDIOOBJBQZ-UHFFFAOYSA-N Phenol Chemical compound OC1=CC=CC=C1 ISWSIDIOOBJBQZ-UHFFFAOYSA-N 0.000 claims description 3
- 150000001408 amides Chemical class 0.000 claims description 3
- 150000001555 benzenes Chemical class 0.000 claims description 3
- 238000007635 classification algorithm Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 239000012535 impurity Substances 0.000 claims description 3
- 150000002576 ketones Chemical class 0.000 claims description 3
- 239000002245 particle Substances 0.000 claims description 3
- 238000010845 search algorithm Methods 0.000 claims description 3
- 238000007873 sieving Methods 0.000 claims description 3
- 238000005303 weighing Methods 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 2
- 238000000643 oven drying Methods 0.000 claims description 2
- 238000010298 pulverizing process Methods 0.000 claims description 2
- 238000012843 least square support vector machine Methods 0.000 claims 1
- 239000003153 chemical reaction reagent Substances 0.000 abstract description 4
- 238000003909 pattern recognition Methods 0.000 abstract description 4
- 239000000523 sample Substances 0.000 description 32
- 238000012549 training Methods 0.000 description 19
- 238000010586 diagram Methods 0.000 description 9
- 238000009826 distribution Methods 0.000 description 6
- 239000000463 material Substances 0.000 description 5
- 240000006891 Artemisia vulgaris Species 0.000 description 4
- 235000003261 Artemisia vulgaris Nutrition 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000000513 principal component analysis Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 235000009051 Ambrosia paniculata var. peruviana Nutrition 0.000 description 1
- 235000003097 Artemisia absinthium Nutrition 0.000 description 1
- 235000017731 Artemisia dracunculus ssp. dracunculus Nutrition 0.000 description 1
- 235000017519 Artemisia princeps Nutrition 0.000 description 1
- 244000065027 Artemisia princeps Species 0.000 description 1
- 238000001157 Fourier transform infrared spectrum Methods 0.000 description 1
- 238000004566 IR spectroscopy Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 239000001138 artemisia absinthium Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 231100000481 chemical toxicant Toxicity 0.000 description 1
- 239000013068 control sample Substances 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 238000007603 infrared drying Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010008 shearing Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- KDYFGRWQOYBRFD-UHFFFAOYSA-N succinic acid Chemical compound OC(=O)CCC(O)=O KDYFGRWQOYBRFD-UHFFFAOYSA-N 0.000 description 1
- 239000003440 toxic substance Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
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
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- 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
- G01N2021/3572—Preparation of samples, e.g. salt matrices
-
- 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
- G01N2021/3595—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using FTIR
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a folium artemisiae argyi producing area tracing method based on an infrared spectrum fingerprint technology, which comprises the following steps of: pretreating folium Artemisiae Argyi samples of different producing areas, and tabletting to obtain tablet; scanning the ingot slice by a Fourier transform infrared spectrometer to obtain an infrared fingerprint spectrum of the folium artemisiae argyi sample; establishing a control infrared spectrum of the folium artemisiae argyi; analyzing and contrasting the infrared spectrum, and picking up a plurality of characteristic peaks as common characteristic peaks; selecting a wave band with obvious producing area characteristics as a characteristic wave band, and carrying out similarity analysis on a common characteristic peak, the characteristic wave band and a full spectrum of infrared fingerprint spectrums of folium artemisiae argyi samples in different producing areas and a reference infrared spectrum; constructing a folium artemisiae argyi producing area tracing model through a pattern recognition technology; inputting an infrared spectrum of the folium artemisiae argyi sample to be identified, and obtaining a source tracing result of the folium artemisiae argyi sample to be identified through a source tracing model. The method solves the technical problems of low accuracy and low efficiency of producing area identification, and has the advantages of no damage, rapidness, high accuracy, no need of complex pretreatment, no need of chemical reagents and the like.
Description
Technical Field
The invention relates to the technical field of origin tracing, in particular to a folium artemisiae argyi origin tracing method based on an infrared spectrum fingerprint technology.
Background
The infrared spectrum analysis technology has the obvious advantages of no damage, rapidness, online detection and the like, particularly, a detected sample is not limited by a physical state, and gas, liquid and solid can be detected, which is beyond the reach of other analysis means. The infrared spectrometry sample has less consumption and no complex pretreatment, does not relate to toxic chemical reagents, and has simple and quick operation and high detection speed, and the measurement process can be completed only in dozens of seconds. The traditional Chinese medicinal materials from different regional sources are different in ecological environment (climate, soil, water quality and the like), the content and the type of chemical components of the traditional Chinese medicinal materials are different, and the difference of infrared spectra is caused by the vibration form caused by infrared radiation. In the prior art, the method for tracing the producing area by analyzing the difference between infrared spectrums is lacked, so that the invention provides the folium artemisiae argyi producing area tracing method based on the infrared spectrum fingerprint technology.
Disclosure of Invention
In order to solve the problems, the invention provides a folium artemisiae argyi producing area tracing method based on an infrared spectrum fingerprint technology, and the method solves the technical problems of low producing area identification accuracy and low identification efficiency.
The invention provides the following technical scheme.
A folium artemisiae argyi producing area tracing method based on an infrared spectrum fingerprint technology comprises the following steps:
pretreating folium Artemisiae Argyi samples of different production places, mixing with dried KBr, grinding, and tabletting to obtain tablet;
scanning the ingot slice by a Fourier transform infrared spectrometer to obtain average spectrograms of spectrograms in three directions as folium artemisiae argyi sample infrared fingerprint spectrums;
selecting folium artemisiae argyi samples in a genuine producing area as control folium artemisiae argyi samples, establishing a control infrared spectrum of folium artemisiae argyi in a common mode, and taking the average value of the absorption intensity of the control folium artemisiae argyi samples as the absorption intensity of the control spectrum;
analyzing and contrasting the infrared spectrum, and picking up a plurality of characteristic peaks as common characteristic peaks;
selecting a wave band with obvious producing area characteristics as a characteristic wave band, and carrying out similarity analysis on a common characteristic peak, the characteristic wave band and a full spectrum of infrared fingerprint spectrums of folium artemisiae argyi samples in different producing areas and a reference infrared spectrum;
constructing a folium artemisiae argyi producing area tracing model by a pattern recognition method;
inputting the information of the common characteristic peak, the characteristic wave band and the full spectrum of the infrared fingerprint spectrum of the folium artemisiae argyi sample to be identified, and obtaining the origin tracing result of the folium artemisiae argyi sample to be identified through an origin tracing model.
Preferably, the acquisition of the folium artemisiae argyi sample infrared fingerprint spectrum comprises the following steps:
removing impurities from folium Artemisiae Argyi samples of different producing areas, selecting, drying, cutting, mixing, oven drying at 45 deg.C until constant weight, pulverizing, and sieving with 80 mesh sieve;
weighing folium Artemisiae Argyi sample 10mg and dried KBr 150mg, mixing, grinding, tabletting into 1mm thick tablet, and testing on machine;
in the scanning process, each sample slice randomly scans 3 directions to obtain 3 spectrograms, and an average spectrogram is taken;
the temperature and humidity in the laboratory are respectively controlled at 22-27 ℃ and 30% -40%, each ingot slice is scanned for 64 times, and H is deducted during scanning2O and CO2Interference of (2);
and carrying out baseline correction and smoothing treatment on the obtained original spectral data through OPUS software to obtain the final folium artemisiae argyi sample infrared fingerprint.
Preferably, the common characteristic peaks include:
3273cm-1an absorption peak with low frequency and wide spectrum band appears nearby, is N-H stretching vibration or O-H stretching vibration, and contains alcohols, phenols or amides; 3004cm-1And 2919cm-1Nearby absorption peaks are C-H stretching vibration which contains olefin and alkane; 1656cm-1The nearby absorption peak is C ═ C or C ═ O stretching vibration, which contains olefins, aldehydes or ketones; 1485cm-1Nearby absorption peaks are C-H bending vibration which contains alkanes; 1263cm-1The nearby absorption peak is O-H bending vibration or C-O-C stretching vibration, which contains phenols, alcohols or esters; 1195cm-1Nearby absorption peaks are N-H bending vibration, C-O stretching vibration and C-C stretching vibration, and contain alkane, amine, ester, alcohol or phenol; 1053cm-1Nearby absorption peaks are C-H bending vibration, C ═ O stretching vibration, N-H bending vibration, and they contain alkanes, phenols, alcohols, or amines; 845cm-1Nearby absorption peaks are C-H out-of-plane bending vibration which contains aldehydes or substituted benzene; 664cm-1The nearby absorption peak is O-H out-of-plane bending vibration, which contains an amine component.
Preferably, the pattern recognition method adopts a KNN classification algorithm, and finally obtains the classification result of the folium artemisiae argyi by comparing the classification effect of the Euclidean distance and the cosine of the included angle and continuously optimizing the K value.
Preferably, the pattern recognition method employs a random forest algorithm.
Preferably, the pattern recognition method optimizes a support vector machine algorithm by using a genetic algorithm.
Preferably, the pattern recognition method employs a grid search algorithm.
Preferably, the pattern recognition method employs a particle swarm algorithm.
Preferably, the pattern recognition method employs a BP neural network algorithm.
Preferably, the pattern recognition method adopts a least squares support vector machine algorithm.
The invention has the beneficial effects that:
the invention provides a folium artemisiae argyi producing area tracing method based on an infrared spectrum fingerprint technology, which has the remarkable advantages of no damage, quickness, online detection and the like, particularly, a detected sample is not limited by a physical state and is expected to be applied and popularized in a medicine enterprise or a supervision department; according to the method, traceability algorithms such as SVM-pso, RF, KNN, LS-SVM, BP-NN, SVM-ga and the like are constructed, the classification effect of 3 SVM-pso, RF and KNN metrology models is relatively good, and the method is more suitable for tracing the origin of folium artemisiae argyi.
Drawings
FIG. 1 is an overall flow diagram of an embodiment of the present invention;
FIG. 2 is a schematic diagram of an folium Artemisiae Argyi infrared control spectrum and common characteristic peaks in the embodiment of the present invention;
FIG. 3 is an original cross-sectional view of an IR spectrum of mugwort leaves produced in different regions according to an embodiment of the present invention;
FIG. 4 is a selected band infrared cross-section of mugwort leaves from different origins according to the embodiment of the present invention;
FIG. 5 is a graph of the first three principal components of Artemisia princeps Pampanini according to the present invention;
FIG. 6 is a diagram showing the test results of the cosine of the Euclidean distance and the included angle in the KNN algorithm according to the embodiment of the present invention;
FIG. 7 is a diagram illustrating the distribution of predicted values and true values in a random forest algorithm according to an embodiment of the present invention;
FIG. 8 is a fitness graph of the SVM-ga algorithm of an embodiment of the present invention;
FIG. 9 is a diagram illustrating the distribution of predicted values and true values in the SVM-ga algorithm according to an embodiment of the present invention;
FIG. 10 is a fitness graph of an SVM-grid algorithm of an embodiment of the present invention;
FIG. 11 is a diagram illustrating the distribution of predicted values and true values in the SVM-grid algorithm according to an embodiment of the present invention;
FIG. 12 is a graph of fitness of an SVM-pso algorithm of an embodiment of the present invention;
FIG. 13 is a diagram of the distribution of predicted values and true values in the SVM-pso algorithm of an embodiment of the present invention;
FIG. 14 is a diagram illustrating the distribution of predicted values and true values in a BP neural network according to an embodiment of the present invention;
fig. 15 is a diagram showing the distribution of predicted values and true values in the LS-SVM according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
The materials and reagents in this example were: ALPHA II Fourier transform infrared spectrometer (Bruker, Germany), DTGS detector (8000-350 cm-1, signal-to-noise ratio of 4000:1, scan accumulation 64 times); an MS-105 type electronic balance (manufactured by Metler-Torledo, Switzerland), an FSJ-A05N6 micro-pulverizer (manufactured by Kagaku Kogyo Co., Ltd.); type FY-15 powder tableting machines (shanghai tianhe mechanical equipment ltd); model 8401-2A far infrared drying oven (changzhou nokiang instruments ltd); KBr (spectral purity, tianjinke miou chemical reagents ltd).
And (3) data processing: software such as SPSS 19.0(IBM, USA), Simca-p 11.5(Umetrics, Sweden), Matlab 2017A (Mathworks Inc., USA) and the like are used for data mining, drawing and pattern recognition.
The folium artemisiae argyi producing area tracing method based on the infrared spectrum fingerprint technology, disclosed by the invention, comprises the following steps of:
s1: pretreating folium Artemisiae Argyi samples of different production places, mixing with dried KBr, grinding, and tabletting to obtain tablet; specifically, the method comprises the following steps: the folium artemisiae argyi samples are collected in 5-6 months in 2020, south-south Yang city and Anyang city of Henan province, Huanggang city of Hubei province, Ningbo city of Zhejiang province and Baoding city of Hebei province, and the like, the folium artemisiae argyi samples in different producing areas are subjected to impurity removal, selection, drying, shearing, uniformly mixing, drying in a 45 ℃ oven at low temperature until constant weight is achieved, crushing and sieving with a 80-mesh sieve for later use. Weighing folium Artemisiae Argyi sample 10mg and dried KBr 150mg, mixing, grinding, tabletting into 1mm thick tablet, and measuring with a machine.
In the scanning, each sample slice randomly scans 3 directions to obtain 3 spectrograms, and the average spectrogram is taken as the final sample spectrogram.
S2: scanning the ingot slice by a Fourier transform infrared spectrometer to obtain average spectrograms of spectrograms in three directions as an infrared fingerprint of the folium artemisiae argyi sample. The temperature and humidity in the laboratory are respectively controlled at 22-27 ℃ and 30% -40%, each ingot slice is scanned for 64 times, and H is deducted during scanning2O and CO2The interference of (2). The obtained original spectrum data is subjected to baseline correction and smoothing processing through OPUS software, and the influence of baseline and noise is eliminated preliminarily.
S3: methodology investigation: the precision experiment result shows that the relative standard deviation of the common peak wave number is less than 0.35 percent; the stability experiment result shows that the relative standard deviation of the common peak wave number is between 0.12 and 4.45 percent, and the sample is stable within 24 hours; the result of repeated experiments shows that the relative standard deviation of the common peak wave number is between 0.16 and 5.13 percent. The result of the methodology investigation shows that the method is reliable, good in repeatability and strong in stability, and meets the requirement of the fingerprint.
S4: selecting folium Artemisiae Argyi samples in the region of genuine production as control folium Artemisiae Argyi samples, establishing folium Artemisiae Argyi control infrared spectra in common mode, and taking the average value of the absorption intensity of the control folium Artemisiae Argyi samples as the absorption intensity of the control spectra. Specifically, the method comprises the following steps:
the Chinese spring county of Hubei province is the traditional genuine area of folium artemisiae argyi and is also the main collection area for clinical preparations and folk traditional medicine. In this section, a comparison infrared spectrum of folium artemisiae argyi is established in a common mode, 15 batches of samples of 3 towns in the county of qichun county of Hubei province are selected as comparison medicinal materials, each sample is randomly scanned for 3 times, and finally, the absorption intensity average value of 45 batches of comparison medicinal materials is used as the absorption intensity of the comparison spectrum, as shown in fig. 2.
S5: and (5) analyzing and contrasting the infrared spectrum, and picking up a plurality of characteristic peaks as common characteristic peaks.
Through comparison and analysis of 45 batches of folium artemisiae argyi infrared spectra, 10 characteristic peaks are picked up and analyzed as common characteristic peaks (the difference of absorption peaks is not more than +/-4 cm)-1) Characterized as follows: 3273cm-1Near the absorption peak with low frequency and wide band, which is N-H stretching vibration or O-H stretching vibration, and contains alcohols, phenols or amides；3004cm-1And 2919cm-1Nearby absorption peaks are C-H stretching vibration which contains olefin and alkane; 1656cm-1The nearby absorption peak is C ═ C or C ═ O stretching vibration, which contains olefin, aldehyde or ketone; 1485cm-1Nearby absorption peaks are C-H bending vibration which contains alkanes; 1263cm-1The nearby absorption peak is O-H bending vibration or C-O-C stretching vibration, which contains phenols, alcohols or esters; 1195cm-1Nearby absorption peaks are N-H bending vibration, C-O stretching vibration and C-C stretching vibration, and contain alkane, amine, ester, alcohol or phenol; 1053cm-1Nearby absorption peaks are C-H bending vibration, C ═ O stretching vibration, N-H bending vibration, and they contain alkanes, phenols, alcohols, or amines; 845cm-1Nearby absorption peaks are C-H out-of-plane bending vibration which contains aldehydes or substituted benzene; 664cm-1The nearby absorption peak is O-H out-of-plane bending vibration, which contains an amine component.
S6: selecting a wave band with obvious producing area characteristics as a characteristic wave band, and carrying out similarity analysis on the common characteristic peak, the characteristic wave band and the full spectrum of the infrared fingerprint spectrums of the folium artemisiae argyi samples in different producing areas and the reference infrared spectrum.
S6.1: through analysis and comparison of the three modes, the characteristics of the common characteristic peak can be highlighted, and the infrared information can be comprehensively displayed.
Similarity analysis is carried out on the common characteristic peak, the selected waveband and the full spectrum of the folium artemisiae argyi samples in different producing areas and the control infrared spectrum, and the analysis results are shown in table 1. As can be seen from Table 1, the characteristic peaks of the folium artemisiae argyi samples in different producing areas are different from the characteristic peaks of the control sample to a certain extent, but the overall similarity is high and is more than 90%; in the comparison of the similarity of the selected wave bands, except that the folium artemisiae argyi samples in Anyang city, Henan province and the comparison chart show more remarkable differences (respectively 80.3% and 73.4%), the samples in other producing areas have higher similarity (both more than 90%), and the attributions of the producing areas cannot be effectively distinguished. On the whole, the similarity of the infrared spectrum information of the folium artemisiae argyi between producing areas is high, and the producing areas of the folium artemisiae argyi cannot be effectively predicted and identified only by comparing the similarity of common characteristic peaks, characteristic wave bands and full spectrums.
TABLE 1 similarity analysis of common characteristic peaks of FTIR spectra of folium Artemisiae Argyi in different producing areas
S6.2: band selection
Fig. 3 shows that the whole spectrum is overlapped seriously, the characteristics of the producing area are not obvious, and discrimination and classification can not be realized only by map information such as peak position, peak intensity, peak height and the like. In addition, 550-400 cm-1And 4000 to 3650cm-1The wave band has larger noise and stronger background; 2600-1600 cm-1Band existence H2O、CO2And the interference is avoided, the effective information of the wave band is less, the similarity is too high, and the factors can influence the origin identification of the folium artemisiae argyi to a certain extent. Therefore, 3650-2600 cm is selected after preliminary analysis is carried out and recommendation is carried out by combining with TQ analysis-1And 1600-550 cm-1The identification model was established and optimized as the study band, the selected band being shown in fig. 4.
S7: principal component analysis
In the process of principal component analysis, the spectral information of the fingerprint area is converted into 1022 data nodes, an 1022 × 75 high-dimensional matrix is formed, and the data is substituted into SIMCA and MATLAB for dimension reduction and calculation. Through feature extraction of 75 parts of folium artemisiae argyi samples, 6 effective main components (table 2) are obtained in total, the variance contribution rates of the effective main components are respectively 80.91%, 10.99%, 4.17%, 2.52%, 0.85% and 0.29%, the accumulated contribution rate reaches 99.73%, and the 6 main components can fully explain the whole spectrum information, the main component cross validation accuracy is high, and the constructed analysis method is effective and stable.
TABLE 2 eigenvalues and cumulative confidence of the top 6 principal components
By using the first three principal components as a map (3D-plots scattergram), as shown in FIG. 5, although the folium artemisiae argyi samples in different producing areas are overlapped, the classification trend is more obvious from the whole, and the samples in different producing areas are relatively concentrated. Therefore, the folium artemisiae argyi origin tracing based on the infrared spectrum technology has certain feasibility, but the principal component analysis only can provide clustering and distance trends and cannot quantitatively classify folium artemisiae argyi samples of different origins, and in order to obtain more accurate and intuitive results, the folium artemisiae argyi origin tracing based on the infrared spectrum technology is analyzed by means of a metrology model.
S8: constructing a folium artemisiae argyi producing area tracing model through a pattern recognition technology;
s9: inputting the information of the common characteristic peak, the characteristic wave band and the full spectrum of the infrared fingerprint spectrum of the folium artemisiae argyi sample to be identified, and obtaining the origin tracing result of the folium artemisiae argyi sample to be identified through an origin tracing model.
Example 2 Pattern recognition method Using KNN Classification Algorithm
The origin of the folium artemisiae argyi is identified by applying a KNN algorithm, classification results are finally obtained by comparing the classification effect of the Euclidean distance and the cosine of the included angle and continuously optimizing the K value, and the classification results are shown in the table 3 and the figure 6 (figure 6a is the test condition of the Euclidean distance, and figure 6b is the test condition of the cosine of the included angle).
TABLE 3 discriminating Effect of KNN Algorithm
As can be seen from table 3, the accuracy of the euclidean distance test set in the KNN algorithm is 86.67%, the accuracy of the cosine distance test set of the included angle is 86.67%, and the overall effect is good. In the Euclidean distance test set, 1 of samples in Ningbo City in Zhejiang province is misjudged as Anguo City in Hebei province, and 1 of samples in Anguo City in Hebei province is misjudged as Nanyang City in Henan province. Other samples are correctly classified, which shows that the folium artemisiae argyi origin tracing research and identification effect based on the K-nerest Neighbor algorithm is good.
Example 3 Pattern recognition method Using random forest Algorithm (RF model)
The origin of the argy wormwood leaves is identified by using a random forest algorithm, 80% of data amount is selected as a training set (the number of samples is 60), 20% of data amount is selected as a testing set (the number of samples is 15), learning and recognition are carried out by Matlab software, and the result is shown in Table 4 and figure 7 (figure 7a is the testing condition of the training set, and figure 7b is the testing condition of the testing set).
TABLE 4 discrimination Effect of random forest Algorithm
As can be seen from Table 4, the accuracy of the training set in the random forest algorithm is 100%, the accuracy of the test set is 86.67%, and the overall effect is ideal. Wherein, all samples are correctly classified in the training set, while 1 of the Ningbo City samples in Zhejiang province is misjudged as south-Yang City in Henan province, 1 of the Anguo City in Hebei province is misjudged as the Qi Chun county in Hubei province, and other samples are correctly classified in the testing set, which shows that the random forest model based on the infrared technology is suitable for the identification of the producing area of the folium artemisiae argyi, and has good identification effect.
Example 4 Pattern recognition method Using genetic Algorithm (SVM-ga model)
In the genetic algorithm optimization support vector machine algorithm, 80% of data volume is selected as a training set (the number of samples is 60), 20% of data volume is selected as a testing set (the number of samples is 15), learning and recognition are carried out through Matlab software, the result of fitness curve selection is shown in a figure 8, and the analysis result is shown in a table 5 and a figure 9 (figure 9a is the testing condition of the training set, and figure 9b is the testing condition of the testing set).
TABLE 5 authentication Effect of SVM-ga Algorithm
As can be seen from Table 5, the accuracy of the training set in the SVM-ga algorithm is 100%, the accuracy of the test set is 26.67%, and the classification effect is poor. In the test set, 5 samples in south-yang city of south-Henan province were all erroneously determined as Qi Chun county of Hehu-north province, 1 sample in Anyang city of south-Henan province was erroneously determined as Qi Chun county of Hubei province, 3 samples in Ningbo city of Zhejiang province were all erroneously determined as Qi Chun county of Hubei province, 2 samples in Anguo city of Hebei province were erroneously determined as Qi Chun county of Hubei province, and the other samples were all correctly classified. Generally, the identification effect of the SVM-ga model based on the isotope technology on the folium artemisiae argyi producing areas is poor, and further optimization is needed if the tracing of the producing areas is initially realized.
80% of the data volume was selected as the training set (sample number 60), 20% of the data volume was selected as the test set (sample number 15), learning and identification were performed by Matlab software, SVR parameters were selected as Best c-5.6569, g-0.044194, CVmse-80 (fig. 10), and the analysis results are shown in table 6 and fig. 11. (FIG. 11a is the test case for the training set, and FIG. 11b is the test case for the test set).
TABLE 6 discrimination Effect of SVM-grid algorithm
As can be seen from table 6, the accuracy of the training set in the SVM-grid algorithm is 98.33%, wherein only 1 sample in ningbo city, zhejiang province is misjudged as ann country city, hebei province; the accuracy of the test set was 86.67%, 1 sample in south yang city of Henan province was judged incorrectly as Ningbo city of Zhejiang province, 1 sample in Anguo city of Hebei province was judged incorrectly as Qi Chun county of Hubei province, and other samples were classified correctly. The SVM-grid model has good identification effect on the origin of the folium artemisiae argyi.
Embodiment 6, the pattern recognition method employs a particle swarm algorithm (SVM-pso model)
In the SVM-pso algorithm, 80% of the data volume is selected as a training set (the number of samples is 60), 20% of the data volume is selected as a testing set (the number of samples is 15), learning and recognition are performed by Matlab software, the MSE parameter selection result is c 1-2, c 2-2 termination algebra is 100, Best c-63.6717 g-0.69837 (fig. 12), and the analysis result is shown in table 7 and fig. 13. (FIG. 13a is a test case of the training set, and FIG. 13b is a test case of the test set).
TABLE 7 discrimination Effect of SVM-grid algorithm
As can be seen from table 7, the accuracy of the training set in the SVM-pso algorithm is 100%, the accuracy of the test set is 86.67%, 1 sample in ningbo city, zhejiang province is misjudged as south yang city, hannan province, 1 sample in ann country, north Hei province is misjudged as long as spring county, north Hu province, and other samples are correctly classified, so that the overall identification effect is good. The SVM-pso model has good identification effect on the production area of the folium artemisiae argyi.
Embodiment 7, the pattern recognition method employs the BP neural network algorithm (BP-NN model)
In the construction of the neural network traceability model, 80% of the data volume is selected as a training set (the number of samples is 60), 20% of the data volume is selected as a testing set (the number of samples is 15), learning and recognition are performed through Matlab software, and the analysis results are shown in table 8 and fig. 14 (fig. 14a is the testing situation of the training set, and fig. 14b is the testing situation of the testing set).
TABLE 8 discriminating Effect of BP neural network Algorithm
From table 8, it can be seen that the accuracy of the training set in the BP neural network algorithm is 100%, the accuracy of the test set is 53.33%, and the classification effect is poor. In the test set, 1 sample of 2 samples in south-south China, Henan province, Qi Yang city and Hubei province, Qi spring county and Hubei province, respectively, is misjudged as AnYang city in Henan province, 1 sample of Qi spring county in Hubei province, respectively, 1 sample of Ningbo city in Zhejiang province, respectively, is misjudged as Qi spring county in Hebei province, respectively, and 2 samples in Annational city in Hebei province, respectively, are misjudged as Ningbo city in Zhejiang province, respectively, and the whole classification effect is poor, so that the discrimination effect of the folium artemisiae argyi based on the BP neural network model is poor, and the method is not suitable for tracing the origin of the folium artemisiae argyi.
Embodiment 8, the pattern recognition method adopts LS-SVM
In the construction of the LS-SVM traceability model, 80% of the data volume is selected as a training set (the number of samples is 60), 20% of the data volume is selected as a testing set (the number of samples is 15), learning and recognition are performed by Matlab software, and the analysis result is shown in table 9 and fig. 15 (fig. 15a is the testing situation of the training set, and fig. 15b is the testing situation of the testing set).
TABLE 15 discrimination Effect of LS-SVM Algorithm
From table 9, it can be seen that the accuracy of the training set in the LS-SVM algorithm is 100%, the accuracy of the test set is 60%, and the classification effect is general. In the test set, 3 samples in south-yang city of Henan province were misjudged as 1 sample in Qi Chun county of Hubei province, 1 sample in Ningbo city of Zhejiang province was misjudged as south-yang city of Henan province, 1 sample in Anguo city of Hebei province was misjudged as Qi Chun county of Hubei province, and the other samples were correctly classified. Generally speaking, the LS-SVM model has a general identification effect on the origin of folium artemisiae argyi, and further optimization is needed when the LS-SVM model is used for tracing the origin of the folium artemisiae argyi.
Comparing the tracing effects of different metrology models:
as can be seen from Table 10, the difference of the identification effect of different pattern recognition methods on the origin and the production area of mugwort leaves is large, and the order of the accuracy of the above 6 pattern recognition is: the SVM-pso ﹦ RF ﹦ KNN is more than LS-SVM more than BP-NN more than SVM-ga, the classification effect of SVM-pso, RF and KNN 3 metrology models is relatively good, the accuracy of a test set is 86.67%, and the SVM-pso model is considered to be supervised and trained, so that the stability is relatively high, and the method is more suitable for tracing the origin of the folium artemisiae argyi. Generally, the method can be used for identifying the origin of the folium artemisiae argyi based on the combination of an infrared spectrum technology and a pattern recognition method.
TABLE 10 comparison of different pattern recognition method effects
The present invention is not limited to the above preferred embodiments, and any modifications, equivalents and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A folium artemisiae argyi producing area tracing method based on an infrared spectrum fingerprint technology is characterized by comprising the following steps:
pretreating folium Artemisiae Argyi samples of different production places, mixing with dried KBr, grinding, and tabletting to obtain tablet;
scanning the ingot slice by a Fourier transform infrared spectrometer to obtain average spectrograms of spectrograms in three directions as folium artemisiae argyi sample infrared fingerprint spectrums;
selecting folium artemisiae argyi samples in a genuine producing area as control folium artemisiae argyi samples, establishing a control infrared spectrum of folium artemisiae argyi in a common mode, and taking the average value of the absorption intensity of the control folium artemisiae argyi samples as the absorption intensity of the control spectrum;
analyzing and contrasting the infrared spectrum, and picking up a plurality of characteristic peaks as common characteristic peaks;
selecting a wave band with obvious producing area characteristics as a characteristic wave band, and carrying out similarity analysis on a common characteristic peak, the characteristic wave band and a full spectrum of infrared fingerprint spectrums of folium artemisiae argyi samples in different producing areas and a reference infrared spectrum;
constructing a folium artemisiae argyi producing area tracing model by a pattern recognition method;
and obtaining the origin tracing result of the folium artemisiae argyi sample to be identified through an origin tracing model.
2. The folium artemisiae argyi production area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the obtaining of the folium artemisiae argyi sample infrared fingerprint spectrum comprises the following steps:
removing impurities from folium Artemisiae Argyi samples of different producing areas, selecting, drying, cutting, mixing, oven drying at 45 deg.C until constant weight, pulverizing, and sieving with 80 mesh sieve;
weighing folium Artemisiae Argyi sample 10mg and dried KBr 150mg, mixing, grinding, tabletting into 1mm thick tablet, and testing on machine;
in the scanning process, each sample slice randomly scans 3 directions to obtain 3 spectrograms, and an average spectrogram is taken;
the temperature and humidity in the laboratory are respectively controlled at 22-27 ℃ and 30% -40%, each ingot slice is scanned for 64 times, and H is deducted during scanning2O and CO2Interference of (2);
and carrying out baseline correction and smoothing treatment on the obtained original spectral data through OPUS software to obtain the final folium artemisiae argyi sample infrared fingerprint.
3. The folium artemisiae argyi producing area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the common characteristic peaks comprise:
3273cm-1an absorption peak with low frequency and wide spectrum band width appears nearby, is N-H stretching vibration or O-H stretching vibration, and contains alcohols, phenols or amides; 3004cm-1And 2919cm-1Nearby absorption peaks are C-H stretching vibration which contains olefin and alkane; 1656cm-1The nearby absorption peak is C ═ C or C ═ O stretching vibration, which contains olefins, aldehydes or ketones; 1485cm-1Nearby absorption peaks are C-H bending vibration which contains alkanes; 1263cm-1The nearby absorption peak is O-H bending vibration or C-O-C stretching vibration, which contains phenols, alcohols or esters; 1195cm-1Nearby absorption peaks are N-H bending vibration, C-O stretching vibration and C-C stretching vibration, and contain alkane, amine, ester, alcohol or phenol; 1053cm-1Nearby absorption peaks are C-H bending vibration, C ═ O stretching vibration, N-H bending vibration, and they contain alkanes, phenols, alcohols, or amines; 845cm-1Nearby absorption peaks are C-H out-of-plane bending vibration which contains aldehydes or substituted benzene; 664cm-1The nearby absorption peak is O-H out-of-plane bending vibration, which contains an amine component.
4. The folium artemisiae argyi origin tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the pattern recognition method adopts a KNN classification algorithm, classification results of folium artemisiae argyi are finally obtained by comparing classification effects of Euclidean distance and cosine of included angle and continuous optimization of K value.
5. The folium artemisiae argyi production area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the pattern recognition method adopts a random forest algorithm.
6. The folium artemisiae argyi production area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the pattern recognition method adopts a genetic algorithm to optimize a support vector machine algorithm.
7. The folium artemisiae argyi production area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the pattern recognition method adopts a grid search algorithm.
8. The folium artemisiae argyi origin and production area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the pattern recognition method adopts a particle swarm algorithm.
9. The folium artemisiae argyi production area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the pattern recognition method adopts a BP neural network algorithm.
10. The folium artemisiae argyi production area tracing method based on the infrared spectrum fingerprint technology as claimed in claim 1, wherein the pattern recognition method adopts a least square support vector machine algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210268408.1A CN114527088A (en) | 2022-03-18 | 2022-03-18 | Folium artemisiae argyi producing area tracing method based on infrared spectrum fingerprint technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210268408.1A CN114527088A (en) | 2022-03-18 | 2022-03-18 | Folium artemisiae argyi producing area tracing method based on infrared spectrum fingerprint technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114527088A true CN114527088A (en) | 2022-05-24 |
Family
ID=81627189
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210268408.1A Pending CN114527088A (en) | 2022-03-18 | 2022-03-18 | Folium artemisiae argyi producing area tracing method based on infrared spectrum fingerprint technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114527088A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115046956A (en) * | 2022-08-17 | 2022-09-13 | 扬州金韵乐器御工坊有限公司 | Zheng panel material detection method based on optical information |
-
2022
- 2022-03-18 CN CN202210268408.1A patent/CN114527088A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115046956A (en) * | 2022-08-17 | 2022-09-13 | 扬州金韵乐器御工坊有限公司 | Zheng panel material detection method based on optical information |
| CN115046956B (en) * | 2022-08-17 | 2022-12-13 | 扬州金韵乐器御工坊有限公司 | Zheng panel material detection method based on optical information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105574474B (en) | A kind of biometric image recognition methods based on Information in Mass Spectra | |
| CN103150498A (en) | Hardware trojan identification method based on single classification supporting vector machine | |
| CN106950315A (en) | The method that chemical composition in sample is quickly characterized based on UPLC QTOF | |
| He et al. | High-precision identification of the actual storage periods of edible oil by FT-NIR spectroscopy combined with chemometric methods | |
| CN114527088A (en) | Folium artemisiae argyi producing area tracing method based on infrared spectrum fingerprint technology | |
| CN108489929A (en) | Ginseng, Radix Notoginseng and the legal base source Panax polysaccharide of three kinds of American Ginseng discrimination method | |
| CN104949956A (en) | Detection method of generic drugs pretending original drugs | |
| CN105139022A (en) | Plastic identification model establishment method via near-infrared hyper-spectral image technology | |
| CN113008805A (en) | Radix angelicae decoction piece quality prediction method based on hyperspectral imaging depth analysis | |
| CN104345045A (en) | Chemical pattern recognition and near infrared spectrum-based similar medicinal material identification method | |
| CN114112983B (en) | Tibetan medicine all-edge leaf green chorine artemisia origin discrimination method based on Python data fusion | |
| CN110376153B (en) | Method for tracing origin of market saffron by combining ATR-FTIR with RBF neural network | |
| CN111523587A (en) | Woody plant species spectrum identification method based on machine learning | |
| CN111428585A (en) | Metamaterial terahertz spectroscopy identification method based on deep learning | |
| CN109975236A (en) | A method of identifying the honeysuckle place of production using near-infrared spectrum technique | |
| Tian et al. | Multi-classification identification of PLS in rice spectra with different pre-treatments and K/S optimisation | |
| CN110008989A (en) | The infrared spectroscopy recognition methods of different target under a kind of spectral signature condition of similarity | |
| CN110751048B (en) | Laser probe classification method and device based on image characteristic automatic spectral line selection | |
| CN108169162B (en) | Rapid evaluation method for soil fertility level of tea garden | |
| CN103076300A (en) | Method for judging and analyzing traditional Chinese medicine resource fingerprint information by specific mode identification model | |
| CN110749575A (en) | Traditional Chinese medicine polysaccharide two-dimensional infrared spectrum identification prediction model and construction method and application thereof | |
| Liu et al. | Discrimination and feature selection of geographic origins of traditional Chinese medicine herbs with NIR spectroscopy | |
| CN112834451B (en) | Sample identification method and device based on infrared spectrum and storage medium | |
| CN108760679A (en) | A kind of gastrodia elata f. glauca discriminating side based on near-infrared spectrum technique | |
| CN116008245A (en) | Application of Sang Shela Manchurian spectral fingerprint establishment combined with machine learning algorithm in mulberry leaf origin identification |
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 |