CN108872128B - Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering - Google Patents

Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering Download PDF

Info

Publication number
CN108872128B
CN108872128B CN201810468730.2A CN201810468730A CN108872128B CN 108872128 B CN108872128 B CN 108872128B CN 201810468730 A CN201810468730 A CN 201810468730A CN 108872128 B CN108872128 B CN 108872128B
Authority
CN
China
Prior art keywords
tea
fuzzy
sample
infrared spectrum
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.)
Active
Application number
CN201810468730.2A
Other languages
Chinese (zh)
Other versions
CN108872128A (en
Inventor
武小红
傅海军
陈勇
武斌
孙俊
戴春霞
翟艳丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Enshi Zhushaxi Ecological Agriculture Co Ltd
Original Assignee
Jiangsu University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiangsu University filed Critical Jiangsu University
Priority to CN201810468730.2A priority Critical patent/CN108872128B/en
Publication of CN108872128A publication Critical patent/CN108872128A/en
Application granted granted Critical
Publication of CN108872128B publication Critical patent/CN108872128B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • 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

  • Physics & Mathematics (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

The invention discloses a tea infrared spectrum classification method based on fuzzy non-relevant C-means clustering. Firstly, collecting the infrared spectrum of a tea sample by using a Fourier infrared spectrum analyzer; then, performing multivariate scattering correction pretreatment on the infrared spectrum; then, reducing the dimension of the spectral data to 20 dimensions by a principal component analysis method; then extracting identification information in the spectral data by utilizing linear discriminant analysis; and finally, classifying the tea varieties by using a fuzzy non-relevant C mean value clustering method. The invention designs a fuzzy non-relevant C mean value clustering method on the basis of the fuzzy C mean value clustering method, has the advantages of high detection speed, high classification accuracy and the like, and can realize the correct classification of tea varieties.

Description

Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering
Technical Field
The invention relates to a tea leaf classification method, in particular to a tea leaf infrared spectrum classification method based on fuzzy non-correlated C-means clustering.
Background
Tea drinking is the traditional food culture of Chinese. The tea contains tea polyphenols, tea polysaccharide, theanine and other substances beneficial to human health. At present, the tea varieties are numerous in the market, and the importance of the tea quality is gradually paid attention by people. However, the tea leaves on the market are not good and uniform, and have a plurality of varieties, so that the quality of the tea leaves is difficult to distinguish. Therefore, it is very important to develop a method for rapidly and effectively identifying the variety of tea.
Infrared spectroscopy is used primarily for qualitative and quantitative analysis of organic compounds. As a nondestructive detection technology, the infrared spectrum technology is widely applied to the fields of agricultural product and food safety detection and the like in recent years. For example: the Yangxihe and the like carry out identification research on the dark tea by utilizing a Fourier infrared spectrum method. Ayvaz and the like collect the mid-infrared spectrum of the potato juice by using a portable mid-infrared system, perform spectrum pretreatment by using normalization and a Savitzky-Golay second-order polynomial filter, and establish a correction model by using partial least squares regression to predict the contents of anthocyanin, phenolic substances and sugar of the potatoes with seven different colors.
The fuzzy clustering based on the objective function in a plurality of fuzzy clustering algorithms becomes the most widely applied fuzzy clustering algorithm at present because the fuzzy clustering algorithm has the advantages of simple design, wide problem solving range, capability of being finally classified as an optimization problem and the like. The fuzzy C-means clustering is the most representative clustering algorithm in the clustering algorithm based on the objective function. However, the fuzzy C-means clustering FCM cannot dynamically extract the identification information of the sample in the fuzzy clustering process. In order to solve the problem, the invention designs a fuzzy non-correlation C-means clustering method. The method can realize the extraction of the fuzzy non-relevant identification information of the data in the fuzzy C-means clustering process, and can achieve higher clustering accuracy.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides the tea infrared spectrum classification method based on fuzzy non-relevant C-means clustering, which has the advantages of high detection speed, high classification accuracy and high classification efficiency. Firstly, collecting the infrared spectrum of a tea sample by using a Fourier infrared spectrum analyzer; then, performing multivariate scattering correction pretreatment on the infrared spectrum; then, reducing the dimension of the spectral data by a principal component analysis method; then extracting identification information in the spectral data by utilizing linear discriminant analysis; and finally, classifying the tea varieties by using a fuzzy non-relevant C mean value clustering method.
The principle on which the invention is based: researches show that the infrared diffuse reflection spectrum of the tea contains internal variety information such as tea polyphenol, caffeine and soluble solid matters in the tea, and the infrared diffuse reflection spectra corresponding to different varieties of tea are different.
A tea infrared spectrum classification method based on fuzzy non-correlation C-means clustering specifically comprises the following steps:
firstly, collecting infrared spectrums of tea samples and preprocessing the spectrums;
and step two, performing dimensionality reduction treatment on the infrared spectrum of the tea sample by adopting a principal component analysis method and extracting identification information of the infrared spectrum of the tea training sample by linear discriminant analysis.
Step three, setting a weight index m of fuzzy C-mean clustering and the maximum iteration number rmaxAnd an error upper limit value. Using a cluster center obtained by fuzzy C-means clustering on the test sample number in the step two as an initial cluster center V of fuzzy non-relevant C-means clustering(0)。
Step four: judging the tea variety by adopting a fuzzy non-relevant C mean value clustering method:
1) initialization: setting a weight index m, a category number c and a test sample number n; setting an initial value r and a maximum iteration number r of the iteration numbermax(ii) a Setting an iteration error upper limit value as;
2) calculating an inter-ambiguity scattering matrix SfB
Figure GDA0002496763530000021
Wherein,
Figure GDA0002496763530000022
for the kth sample x at the r iterationkFuzzy membership degree belonging to the ith class, wherein m represents weight; c is the number of the categories,
Figure GDA0002496763530000023
for the class center value of the ith class at the r-th iteration,
Figure GDA0002496763530000024
in order to test the mean value of the samples,
Figure GDA0002496763530000025
n is the number of test samples, xjFor the jth test sample, the superscript T represents the matrix transpose operation.
3) Calculating a fuzzy total scattering matrix SfT
Figure GDA0002496763530000026
Wherein x iskIs the kth test sample.
4) Computing feature vectors
Figure GDA0002496763530000027
Wherein,
Figure GDA0002496763530000028
as the inverse of the fuzzy dispersion matrix, SfBIn order to blur the inter-class hash matrix, λ is the eigenvalue corresponding to the eigenvector ψ. The maximum eigenvalue lambda obtained by calculation1Corresponding eigenvector psi1As the 1 st vector of the fuzzy uncorrelated differential transformation vectors, if p fuzzy uncorrelated differential transformation vectors are psi1，ψ2，...，ψpThen, the p +1 th fuzzy uncorrelated discriminative transfer vector is calculated as follows:
QSfBψp+1＝βSfTψp+1
Q＝I-SfT TψT(ψSfTψT)-1ψ，
ψ＝[ψ1ψ2…ψP]
β is the eigenvector psip+1And the corresponding characteristic value, I, is an identity matrix.
5) X is to bek∈RqConversion into a feature space (by psi)1，ψ2，...，ψpComposition of
yk＝xk T[ψ1，ψ2，...，ψp](yk∈Rp)
Where p and q are both the dimensions of the sample,. psipIs the p-th feature vector.
6) Also will be
Figure GDA0002496763530000031
Conversion to a feature space
Figure GDA0002496763530000032
Wherein,
Figure GDA0002496763530000033
for class center value, ψ, in fuzzy C-means clustering in step threepIs the p-th feature vector.
7) Computing fuzzy membership function values in feature space
Figure GDA0002496763530000034
Wherein, ykFor the kth sample in the feature space,
Figure GDA0002496763530000035
is the sample y at the r +1 th iterationkFuzzy membership value, u, belonging to class ii k(r+1) Is the fuzzy membership value of the (r + 1) th iterative computation; v. ofi′(r)And vj′(r)The class center values of the ith class and the jth class of the ith iterative computation are respectively; c is the number of categories, and m is the weight value.
8) Computing class-centered values for classes i in feature space
Figure GDA0002496763530000036
Figure GDA0002496763530000037
Wherein,
Figure GDA0002496763530000038
class center of class i of the r +1 th iterative computation
Figure GDA0002496763530000039
The value of (c).
9) Increasing the number r of iterations, i.e. r +1, until
Figure GDA00024967635300000310
Or r > rmaxThe computation terminates, otherwise it will
Figure GDA00024967635300000311
Is given to a variable
Figure GDA00024967635300000312
Figure GDA00024967635300000313
Is given to a variable
Figure GDA00024967635300000314
And continuing to recalculate from 2).
The invention has the beneficial effects that:
1. according to the fuzzy non-relevant C mean value clustering method, the fuzzy non-relevant identification conversion vector is calculated in the fuzzy C mean value clustering process to extract the identification information of the near infrared spectrum of the tea sample, the extraction of the data non-relevant identification information in the fuzzy C mean value clustering process can be realized, and the higher clustering accuracy is achieved. Can realize the correct classification of different tea varieties.
2. The method can dynamically extract the fuzzy non-relevant identification information of the tea infrared spectrum data in the fuzzy C mean value clustering process, and can improve the accuracy of tea variety identification.
Drawings
FIG. 1 is a flow chart of a fuzzy non-correlated C-means clustering tea infrared spectrum classification method.
Figure 2 is an infrared spectrum of a sample of tea.
FIG. 3 is an infrared spectrum of tea after multiple scattering correction
FIG. 4 is a two-dimensional test sample obtained after linear discriminant analysis processing
FIG. 5 is a fuzzy membership value obtained by fuzzy uncorrelated C-means clustering.
Detailed Description
The invention is further illustrated by the following examples and figures.
As shown in fig. 1, the method of the present invention comprises the steps of:
step one, infrared spectrum collection and spectrum pretreatment of a tea sample: the FTIR-7600 type Fourier infrared spectrum analyzer is preheated for 1 hour after being started. The number of scanning times is 32, and the wave number range of spectral scanning is 7800cm-1～350cm-1With a scanning interval of 1.928cm-1Resolution of 4cm-1. Taking three kinds of tea leaves of high-quality phyllostachys chinensis, low-quality phyllostachys chinensis and Emei mountain Maofeng as research objects, grinding and crushing the three kinds of tea leaves in a proper amount, filtering the three kinds of tea leaves by using a 40-mesh sieve, and uniformly mixing 0.5g of the three kinds of tea leaves with potassium bromide in a ratio of 1: 100. For each sample, 1g of the mixture was pressed and then scanned 3 times by a spectrometer, and the average of the 3 times was taken as sample spectral data. The collection environment temperature is 25 ℃ and the relative humidity is 50%. 32 samples were taken for each tea leaf, for a total of 96 samples. Each sample is a 1868-dimensional data with wave number range of 4001.569cm-1～401.1211cm-1. Selecting 22 samples from each sample as test samples, and obtaining 66 test samples; the remaining 30 samples were used as training samples. The infrared spectrum of the tea sample is shown in figure 2.
And (3) performing spectrum pretreatment on the infrared spectrum of the tea by adopting Multivariate Scattering Correction (MSC). The infrared spectrum of the tea leaves after the spectrum pretreatment is shown in figure 3.
And step two, performing dimensionality reduction treatment on the infrared spectrum of the tea sample by adopting a principal component analysis method and extracting identification information of the infrared spectrum of the tea training sample by linear discriminant analysis.
Performing dimensionality reduction treatment on the infrared spectrum of the tea sample by adopting a principal component analysis method: carrying out characteristic decomposition on the infrared spectrum of the tea leaf sample in figure 3 by adopting a principal component analysis method to obtain the first 20 characteristic vectors v1，v2...v20And corresponding 20 eigenvalues λ1，λ2...λ20. Each feature vector is 1868-dimensional data, and the feature values are as follows
λ1＝293.9148，λ2＝129.0279，λ3＝19.0010，λ4＝14.8802，
λ5＝6.4349，λ6＝3.8189，λ7＝2.0033，λ8＝1.4310，
λ9＝1.0661，λ10＝0.6298，λ11＝0.4020，λ12＝0.3169，
λ13＝0.2706，λ14＝0.2294，λ15＝0.1928，λ16＝0.1281，λ17＝0.1066，λ18＝0.0878，λ19＝0.0727，λ20＝0.0613。
10 samples are taken from each of the three tea samples to form a tea sample training set, so that the number of the samples in the training set is 30, and the total number of the samples in the testing set is 66 when the rest samples form a tea sample testing set. The linear discriminant analysis is performed to calculate the discrimination vectors of the 20-dimensional training set samples, and the 20-dimensional test set samples are projected onto the first 2 discrimination vectors, and the LDA score chart is shown in FIG. 4. In fig. 4, the dot symbols represent "the" Emei mountain Maofeng ", and the asterisk and the circle represent" the "Gaoyang leaf Qing, Gaoyang leaf Qing", respectively. As can be seen from fig. 4, the data overlap for the three tea test samples is very small. And the data overlapping is less, so that the clustering accuracy is improved.
Step three, setting weight index of fuzzy C-means clustering (FCM)m is 2, maximum number of iterations rmaxThe upper limit of the error is 0.00001 when the value is 100. Performing fuzzy C-means clustering (FCM) on the two-dimensional test sample number shown in the second step and obtained cluster center as an initial cluster center V of the fuzzy non-correlation C-means clustering method(0)：
Figure GDA0002496763530000051
Step four: judging the tea variety by adopting a fuzzy non-relevant C mean value clustering method:
1) initialization: setting the weight index m to be 2, the category number c to be 3, and the test sample number n to be 66; setting an initial iteration number value r to be 0 and a maximum iteration number rmaxIs 100; setting the upper limit value of the iteration error to be 0.0001;
2) calculating an inter-ambiguity scattering matrix SfB
Figure GDA0002496763530000052
Wherein,
Figure GDA0002496763530000053
for the kth sample x at the r iterationkFuzzy membership degree belonging to the ith class, wherein m represents a weight index; c is the number of the categories,
Figure GDA0002496763530000054
for the class center value of the ith class at the r-th iteration,
Figure GDA0002496763530000055
in order to test the mean value of the samples,
Figure GDA0002496763530000061
n is the number of test samples, xjFor the jth test sample, the superscript T represents the matrix transpose operation.
3) Calculating a fuzzy total scattering matrix SfT
Figure GDA0002496763530000062
Wherein x iskIs the kth test sample.
4) Computing feature vectors
Figure GDA0002496763530000063
Wherein,
Figure GDA0002496763530000064
as the inverse of the fuzzy dispersion matrix, SfBIn order to blur the inter-class hash matrix, λ is the eigenvalue corresponding to the eigenvector ψ. The maximum eigenvalue lambda obtained by calculation1Corresponding eigenvector psi1As the 1 st vector of the fuzzy uncorrelated differential transformation vectors, if p fuzzy uncorrelated differential transformation vectors are psi1，ψ2，...，ψpThen, the p +1 th fuzzy uncorrelated discriminative transfer vector is calculated as follows:
QSfBψp+1＝βSfTψp+1
Q＝I-SfT TψT(ψSfTψT)-1ψ，
ψ＝[ψ1ψ2…ψP]
5) x is to bek∈RqConversion into a feature space (by psi)1，ψ2，...，ψpComposition of
yk＝xk T[ψ1，ψ2，...，ψp](yk∈Rp)
Where p and q are both the dimensions of the sample,. psipIs the p-th feature vector.
6) Also will be
Figure GDA0002496763530000065
Conversion to a feature space
Figure GDA0002496763530000066
Wherein,
Figure GDA0002496763530000067
for class center value, ψ, in fuzzy C-means clustering in step threepIs the p-th feature vector.
7) Computing fuzzy membership function values in feature space
Figure GDA0002496763530000071
Wherein, ykFor the kth sample in the feature space,
Figure GDA0002496763530000072
is the sample y at the r +1 th iterationkFuzzy membership value, u, belonging to class iik (r+1)Is the fuzzy membership value of the (r + 1) th iterative computation; v. ofi′(r)And vj′(r)The class center values of the ith class and the jth class of the ith iterative computation are respectively; c is the number of categories and m is the weight index.
8) Computing class-centered values for classes i in feature space
Figure GDA0002496763530000073
Figure GDA0002496763530000074
Wherein,
Figure GDA0002496763530000075
class center of class i of the r +1 th iterative computation
Figure GDA0002496763530000076
The value of (c).
9) Increasing the number r of iterations, i.e. r +1, until
Figure GDA0002496763530000077
Or r > rmaxThe computation terminates, otherwise it will
Figure GDA0002496763530000078
Is given to a variable
Figure GDA0002496763530000079
Figure GDA00024967635300000710
Is given to a variable
Figure GDA00024967635300000711
And continuing to recalculate from 2).
The experimental results are as follows: p is 2, q is 2, r is 6 times at the end of the iteration, and the class center matrix is
Figure GDA00024967635300000712
The fuzzy membership obtained after the iteration is terminated is shown in fig. 5.
The above-listed detailed description is only a specific description of a possible embodiment of the present invention, and they are not intended to limit the scope of the present invention, and equivalent embodiments or modifications made without departing from the technical spirit of the present invention should be included in the scope of the present invention.

Claims (7)

1. A tea infrared spectrum classification method based on fuzzy non-correlation C-means clustering is characterized by comprising the following steps:
firstly, collecting infrared spectrums of tea samples and preprocessing the spectrums;
step two, adopting a principal component analysis method to perform dimensionality reduction processing on the infrared spectrum of the tea sample and extracting identification information of the infrared spectrum of the tea training sample by linear discriminant analysis;
step three, setting a weight index m of fuzzy C-mean clustering and the maximum iteration number rmaxThe upper limit value of the error; for the test sample of step twoTaking a cluster center obtained by fuzzy C-means clustering as an initial cluster center V of fuzzy non-relevant C-means clustering(0)；
Step four: judging the tea variety by adopting a fuzzy non-relevant C mean value clustering method;
the concrete implementation of the fourth step comprises the following steps:
1) initialization: setting a weight index m, a category number c and a test sample number n; setting an initial value r and a maximum iteration number r of the iteration numbermax(ii) a Setting the upper limit value of the iterative error as;
2) calculating an inter-ambiguity scattering matrix SfB
Figure FDA0002496763520000011
Wherein,
Figure FDA0002496763520000012
for the kth sample x at the r iterationkFuzzy membership degree belonging to the ith class, wherein m represents weight; c is the number of the categories,
Figure FDA0002496763520000013
for the class center value of the ith class at the r-th iteration,
Figure FDA0002496763520000014
in order to test the mean value of the samples,
Figure FDA0002496763520000015
n is the number of test samples, xjFor the jth test sample, superscript T represents matrix transposition operation;
3) calculating a fuzzy total scattering matrix SfT
Figure FDA0002496763520000016
Wherein x iskIs the kth test sample;
4) computing feature vectors
Figure FDA0002496763520000017
Wherein,
Figure FDA0002496763520000018
as the inverse of the fuzzy dispersion matrix, SfBIn order to obtain fuzzy inter-class hash matrix, λ is the eigenvalue corresponding to the eigenvector ψ, and the maximum eigenvalue λ obtained by calculation is1Corresponding eigenvector psi1As the 1 st vector of the fuzzy uncorrelated differential transformation vectors, if p fuzzy uncorrelated differential transformation vectors are psi1，ψ2，...,ψpThen, the p +1 th fuzzy uncorrelated discriminative transfer vector is calculated as follows:
Figure FDA0002496763520000021
5) x is to bek∈RqConversion to a feature space
yk＝xk T[ψ1，ψ2，...,ψp](yk∈Rp)
Where p and q are both the dimensions of the sample,. psipFor the p-th feature vector, the feature space is defined by1,ψ2，...，ψpComposition is carried out;
6) also will be
Figure FDA0002496763520000022
Conversion to a feature space
Figure FDA0002496763520000023
Wherein,
Figure FDA0002496763520000024
clustering for fuzzy C-means in three stepsClass center value ofpIs the p-th feature vector;
7) computing fuzzy membership function values in feature space
Figure FDA0002496763520000025
Wherein, ykFor the kth sample in the feature space,
Figure FDA0002496763520000026
is the sample y at the r +1 th iterationkFuzzy membership value, u, belonging to class iik (r+1)Is the fuzzy membership value of the (r + 1) th iterative computation; v. ofi′(r)And vj′(r)The class center values of the ith class and the jth class of the ith iterative computation are respectively; c is the number of categories, and m is the weight value;
8) computing class-centered values for classes i in feature space
Figure FDA0002496763520000027
Figure FDA0002496763520000028
Wherein,
Figure FDA0002496763520000029
class center of class i of the r +1 th iterative computation
Figure FDA00024967635200000210
A value of (d);
9) increasing the number r of iterations, i.e. r +1, until
Figure FDA00024967635200000211
Or r>rmaxThe computation terminates, otherwise it will
Figure FDA0002496763520000031
Is given to a variable
Figure FDA0002496763520000032
Is given to a variable
Figure FDA0002496763520000033
And continuing to recalculate from 2).
2. The tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering as claimed in claim 1, wherein the specific implementation of the first step of tea sample infrared spectrum collection comprises:
grinding and crushing N kinds of tea leaves with a proper amount, filtering the tea leaves by using a 40-mesh sieve, and uniformly mixing 0.5g of tea leaves with 1:100 of potassium bromide; taking 1g of the mixture for film pressing of each sample, then scanning for 3 times by a spectrometer, and taking the average value of 3 times as sample spectrum data; the collection environment temperature is set to be 25 ℃, and the relative humidity is 50%; collecting 32 samples of each tea leaf, and obtaining N x 32 samples, wherein each sample is 1868-dimensional data, and the wave number range is 4001.569cm-1～401.1211cm-1(ii) a And selecting 22 samples as the test samples from each sample, wherein the total number of the test samples is N x 22, and the rest samples are used as training samples.
3. The tea infrared spectrum classification method based on fuzzy uncorrelated C-means clustering as claimed in claim 2, wherein the spectrometer is a fourier infrared spectrum analyzer of FTIR-7600 type.
4. The tea infrared spectrum classification method based on fuzzy non-correlation C-means clustering as claimed in claim 2, wherein the spectrum preprocessing in the first step is: and (3) performing spectrum pretreatment on the infrared spectrum of the tea by adopting a multi-element scattering correction MSC.
5. The tea infrared spectrum classification method based on fuzzy non-correlation C-means clustering as claimed in claim 1, wherein the second step is realized by:
performing dimensionality reduction treatment on the infrared spectrum of the tea sample by adopting a principal component analysis method: carrying out characteristic decomposition on the infrared spectrum of the tea sample by adopting a principal component analysis method to obtain the first 20 characteristic vectors v1，v2…v20And corresponding 20 eigenvalues λ1，λ2…λ20；
10 samples are taken from each of the N tea samples to form a tea sample training set, the number of the samples in the training set is N x 10, and the rest samples form a tea sample testing set; and (3) operating linear discriminant analysis to calculate the discrimination vectors of the 20-dimensional training set samples, and projecting the 20-dimensional testing set samples onto the first 2 discrimination vectors to obtain the linear discriminant analysis LDA score.
6. The tea infrared spectrum classification method of fuzzy non-correlation C-means clustering as claimed in claim 1, wherein in the third step, the weight index m is set to 2, and the maximum iteration number r is set asmaxThe upper limit of the error is 0.00001 when the value is 100.
7. The tea infrared spectrum classification method based on fuzzy non-correlation C-means clustering as claimed in claim 1, wherein the weight m in step 1) is 2, the class number C is 3, the test sample number n is 66; setting an initial iteration number value r to be 0 and a maximum iteration number rmaxIs 100; and setting the iteration maximum error parameter to be 0.0001.
CN201810468730.2A 2018-05-16 2018-05-16 Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering Active CN108872128B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810468730.2A CN108872128B (en) 2018-05-16 2018-05-16 Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810468730.2A CN108872128B (en) 2018-05-16 2018-05-16 Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering

Publications (2)

Publication Number Publication Date
CN108872128A CN108872128A (en) 2018-11-23
CN108872128B true CN108872128B (en) 2020-08-28

Family

ID=64334484

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810468730.2A Active CN108872128B (en) 2018-05-16 2018-05-16 Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering

Country Status (1)

Country Link
CN (1) CN108872128B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111815760A (en) * 2020-06-23 2020-10-23 中国地质大学（武汉） Three-dimensional stratum drillable field space modeling method
CN111898690B (en) * 2020-08-05 2022-11-18 山东大学 Power transformer fault classification method and system
CN114384042B (en) * 2021-12-30 2024-06-07 江苏大学 Huangshan Mao peak grade identification method for extracting fuzzy weighted identification information

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103425994A (en) * 2013-07-19 2013-12-04 淮阴工学院 Feature selecting method for pattern classification
CN104700108A (en) * 2015-02-13 2015-06-10 中国农业大学 Remote diagnosis method and system for celery disease

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070064983A1 (en) * 2005-09-16 2007-03-22 Wen-Chen Huang Method for automatically detecting nasal tumor

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103425994A (en) * 2013-07-19 2013-12-04 淮阴工学院 Feature selecting method for pattern classification
CN104700108A (en) * 2015-02-13 2015-06-10 中国农业大学 Remote diagnosis method and system for celery disease

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
Classification of Some Active HIV-1 Protease Inhibitors and Their Inactive Analogues Using Some Uncorrelated Three-Dimensional Molecular Descriptors and a Fuzzy c-Means Algorithm;Thy-Hou Lin,Ging-Ming Wang et al.;《journal of chemical information and computer science》;20021004;1490-1504 *
Face recognition based on the uncorrelated discriminant transformation;Zhong Jin,Jing-Yu Yang et al.;《the journal of pattern recognition society》;20011231;1405-1416 *
FUDT在苹果近红外光谱分类中的应用;武斌，马桂香等;《计算机工程与应用》;20161231;193-196,223 *
Intelligent Driver Drowsiness Detection System Using Uncorrelated Fuzzy Locality Preserving Analysis;Rami N. Khushaba，Sarath Kodagoda et al.;《2011 IEEE/RSJ international conference on intelligent robots and systems》;20110930;1-7 *
模糊非相关判别转换及其应用;武小红,武斌等;《中国图象图形学报》;20090930;1832-1836 *

Also Published As

Publication number Publication date
CN108872128A (en) 2018-11-23

Similar Documents

Publication Publication Date Title
Zhao et al. Qualitative identification of tea categories by near infrared spectroscopy and support vector machine
CN106408012A (en) Tea infrared spectrum classification method of fuzzy discrimination clustering
Cai et al. Using FTIR spectra and pattern recognition for discrimination of tea varieties
CN108872128B (en) Tea infrared spectrum classification method based on fuzzy non-correlated C-means clustering
CN110378374B (en) Tea near infrared spectrum classification method for extracting fuzzy identification information
Wu et al. Discrimination of tea varieties using FTIR spectroscopy and allied Gustafson-Kessel clustering
CN104374738B (en) A kind of method for qualitative analysis improving identification result based on near-infrared
Hu et al. Assessing the authenticity of black pepper using diffuse reflectance mid-infrared Fourier transform spectroscopy coupled with chemometrics
CN105181650B (en) A method of quickly differentiating local tea variety using near-infrared spectrum technique
Toher et al. A comparison of model-based and regression classification techniques applied to near infrared spectroscopic data in food authentication studies
CN109685098B (en) Tea variety classification method for fuzzy inter-cluster separation and clustering
CN107192686B (en) Method for identifying possible fuzzy clustering tea varieties by fuzzy covariance matrix
CN104374739A (en) Identification method for authenticity of varieties of seeds on basis of near-infrared quantitative analysis
Chai et al. Improved 1D convolutional neural network adapted to near-infrared spectroscopy for rapid discrimination of Anoectochilus roxburghii and its counterfeits
CN107271394A (en) A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network
CN108491894B (en) Tea leaf classification method capable of fuzzy identification of C-means clustering
CN110378373B (en) Tea variety classification method for fuzzy non-relevant linear discriminant analysis
Chen et al. Combination of the manifold dimensionality reduction methods with least squares support vector machines for classifying the species of sorghum seeds
CN108764288A (en) A kind of GK differentiates the local tea variety sorting technique of cluster
Yu et al. Prediction of enological parameters and discrimination of rice wine age using least-squares support vector machines and near infrared spectroscopy
CN109886296A (en) A kind of authentication information extracts the local tea variety classification method of formula noise cluster
Chen et al. Application of UV-Vis and infrared spectroscopy on wild edible bolete mushrooms discrimination and evaluation: a review
CN114112983B (en) Tibetan medicine all-edge leaf green chorine artemisia origin discrimination method based on Python data fusion
Li et al. Machine learning technique combined with data fusion strategies: A tea grade discrimination platform
CN114331474A (en) Milk producing area tracing method based on fuzzy linear discriminant analysis

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
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20210121

Address after: No.10 Jinshan Road, West Industrial Park, Ji'an County, Ji'an City, Jiangxi Province 343100

Patentee after: Ji'an Jirui Technology Co.,Ltd.

Address before: No. 605, Jianshe Road, Sanmao street, Yangzhong City, Zhenjiang City, Jiangsu Province, 212200

Patentee before: Jiangsu Jiayi Technology Information Service Co.,Ltd.

Effective date of registration: 20210121

Address after: No. 605, Jianshe Road, Sanmao street, Yangzhong City, Zhenjiang City, Jiangsu Province, 212200

Patentee after: Jiangsu Jiayi Technology Information Service Co.,Ltd.

Address before: Zhenjiang City, Jiangsu Province, 212013 Jingkou District Road No. 301

Patentee before: JIANGSU University

TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20220818

Address after: Longjiagou Formation, Zhusha Creek, Bajiao Dong Township, Enshi City, Enshi Tujia and Miao Autonomous Prefecture, Hubei Province 445033

Patentee after: Enshi Zhushaxi Ecological Agriculture Co., Ltd.

Address before: No.10 Jinshan Road, West Industrial Park, Ji'an County, Ji'an City, Jiangxi Province 343100

Patentee before: Ji'an Jirui Technology Co.,Ltd.

TR01 Transfer of patent right