CN112801172A - Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition - Google Patents

Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition Download PDF

Info

Publication number
CN112801172A
CN112801172A CN202110098774.2A CN202110098774A CN112801172A CN 112801172 A CN112801172 A CN 112801172A CN 202110098774 A CN202110098774 A CN 202110098774A CN 112801172 A CN112801172 A CN 112801172A
Authority
CN
China
Prior art keywords
fuzzy
matrix
class
fcm
sample
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
Application number
CN202110098774.2A
Other languages
Chinese (zh)
Inventor
刘锦茂
武小红
沈砚君
谭阳
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.)
Jiangsu University
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 CN202110098774.2A priority Critical patent/CN112801172A/en
Publication of CN112801172A publication Critical patent/CN112801172A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • 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
    • G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
    • G—PHYSICS
    • 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
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
    • G06N7/023—Learning or tuning the parameters of a fuzzy system

Abstract

The invention discloses a Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition, which comprises the steps of collecting near infrared spectrum data of a vegetable sample to be analyzed; and dividing the near infrared spectral data into training samples xiAnd a test specimen
Figure DDA0002914933770000011
Extracting identification information of near infrared spectrum data of the vegetables by adopting a fuzzy singular value decomposition method; respectively converting the test sample and the training sample by adopting a linear discriminant analysis method; and performing spectral data clustering analysis on the converted test sample and training sample by adopting a fuzzy covariance matrix clustering method. The method is advantageousThe fuzzy singular value decomposition method effectively solves the problem of small samples of the existing fuzzy linear discrimination method.

Description

Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition
Technical Field
The invention relates to machine learning and artificial intelligence neighborhood, in particular to a Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition.
Background
At present, pesticide application becomes one of important measures for preventing and treating plant insect diseases and improving the yield and quality of agricultural products, but unreasonable pesticide application brings harm to human health and environment. Therefore, exploring a method for effectively detecting the concentration of pesticide residues has research value and significance for ensuring the food safety of consumers.
The near infrared spectrum detection technology is a non-destructive detection technology for determining the content of a substance by utilizing the characteristics of the substance such as absorption, scattering, reflection, transmission and the like of the substance to light. Because it is in accordance with the characteristics of accuracy, reliability, rapidness, no damage, etc., it is widely used in the detection of agricultural and sideline products. The cabbage with different pesticide residues has difference in reflected near infrared spectrum, and by utilizing the characteristic, the pesticide residues on the cabbage can be qualitatively analyzed, so that the cabbage is classified.
Fuzzy Linear Discriminant (FLDA) is based on a fuzzy set, a Linear Discriminant Analysis (LDA) method is improved by using a fuzzy internal scattering matrix and a fuzzy overall scattering matrix, and the FLDA can effectively extract fuzzy discrimination information of a sample. However, FLDA has a "small sample problem" when dealing with high-dimensional spectral data.
Clustering algorithms are divided into two main classes, the first class of algorithms is a hard clustering algorithm such as a k-means clustering algorithm, a data set is divided into different classes, and each object only belongs to one class. The second category is fuzzy clustering algorithms, which allow an object to belong to multiple categories. Because most objects are not strictly distinguished, a fuzzy clustering algorithm is selected to replace a hard clustering algorithm. The fuzzy C-means clustering algorithm (FCM) is a clustering algorithm established on the basis of the criterion of minimum square error, so that the sum of the membership degrees of data points in all classes is 1, and the solution that all the membership degrees are 0 is effectively avoided.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition, and the fuzzy singular value decomposition method is used for effectively solving the problem of small samples of the existing fuzzy linear discrimination method.
The technical scheme adopted by the invention is as follows:
s1, collecting near infrared spectrum data of the vegetable sample to be analyzed; and integrating the near infrared spectral dataDivided into training samples xiAnd a test specimen
Figure BDA0002914933750000011
S2, extracting identification information of near infrared spectrum data of the vegetables by adopting a fuzzy singular value decomposition method;
s3, respectively converting the test sample and the training sample in the S2 by adopting a linear discriminant analysis method;
and S4, performing spectral data clustering analysis on the test sample and the training sample subjected to conversion in the S3 by adopting a fuzzy covariance matrix clustering method.
Further, the method for extracting the authentication information in S2 includes:
s2.1, calculating the fuzzy membership u of the training sampleij：
Figure BDA0002914933750000021
S2.2, fuzzy membership u based on training samplesijSeparately compute training samples xiIs the matrix of dispersion S between fuzzy classesfBAnd the dispersion matrix S in the fuzzy classfW；
S2.3, based on the dispersion matrix S between fuzzy classesfBAnd the dispersion matrix S in the fuzzy classfWRespectively construct a matrix HfWAnd HfB；
S2.4, from matrix HfWAnd HfBConstruction matrix
Figure BDA0002914933750000022
Performing singular value decomposition on the matrix M to obtain a diagonal matrix R and a unitary matrix Q;
s2.5, performing singular value decomposition on the matrix P to obtain a unitary matrix V,
s2.6, constructing a matrix based on a unitary matrix Q, a diagonal matrix R, a unitary matrix V and an identity matrix I which are obtained by singular value decomposition
Figure BDA0002914933750000023
A transformation matrix G formed by the first 3 columns of vectors of the matrix W;
s2.7, respectively aligning the test samples by using the transformation matrix G
Figure BDA0002914933750000024
And training sample xiTransforming to obtain transformed test samples
Figure BDA0002914933750000025
Transformed training sample yi＝xiG。
Further, in S3, the test sample is analyzed by linear discriminant analysis
Figure BDA0002914933750000026
And training sample yiRespectively converted into test samples
Figure BDA0002914933750000027
And training sample zi。
Further, the method for clustering and analyzing the spectral data comprises the following steps:
s4.1, testing samples converted in S3
Figure BDA0002914933750000028
Obtaining a fuzzy membership value u which is subordinate to the jth class after running fuzzy C mean value clusteringjt,FCMAnd class-center value v of class jj,FCMAnd will ujt,FCMAnd vj,FCMAs initial fuzzy membership value and initial class center value of subsequent fuzzy clustering; establishing a fuzzy clustering objective function:
Figure BDA0002914933750000031
wherein the content of the first and second substances,
Figure BDA0002914933750000032
is a test specimen
Figure BDA0002914933750000033
To class center vj,FCMA distance measure of (d); d is the dimension of the test sample; sfj，FCMIs a fuzzy covariance matrix calculated after the FCM is operated;
s4.2, calculating parameters
Figure BDA0002914933750000034
Wherein the content of the first and second substances,
Figure BDA0002914933750000035
is a test specimen
Figure BDA0002914933750000036
To class center vs,FCMA distance measure of (d);
s4.3, based on the parameters calculated in step S4.2
Figure BDA0002914933750000037
For the test sample
Figure BDA0002914933750000038
And performing iterative calculation, and classifying the Chinese cabbages according to the fuzzy membership value at the end of the iteration.
Further, the iterative process in S4.3 is:
s4.3.1, calculating test sample
Figure BDA0002914933750000039
Belonging to class centre gammajFuzzy membership of (d):
Figure BDA00029149337500000310
wherein the content of the first and second substances,
Figure BDA00029149337500000311
is a test specimen
Figure BDA00029149337500000312
To class j centre gammajA measure of the distance of (a) is,
Figure BDA00029149337500000313
is a test specimen
Figure BDA00029149337500000314
To class j centre gammajA distance measure of (d);
s4.3.2 calculate class center:
Figure BDA00029149337500000315
and after the iteration is ended, classifying the near infrared spectrum of the vegetables according to the fuzzy membership value obtained by calculation.
Further, Sfj，FCMIs a fuzzy covariance matrix calculated after the FCM is run, and is expressed as:
Figure BDA00029149337500000316
wherein the content of the first and second substances,
Figure BDA00029149337500000317
to test the sample
Figure BDA00029149337500000318
Test sample after running fuzzy C-means clustering
Figure BDA00029149337500000319
Fuzzy membership value belonging to j class, m is weight index;
further, the sample is tested
Figure BDA0002914933750000041
To class center vs,FCMMeasure of distance of
Figure BDA0002914933750000042
Expressed as:
Figure BDA0002914933750000043
wherein v iss,FCMObtaining a clustering center belonging to the s-th class after the FCM is operated; sfs,FCMIs the fuzzy covariance matrix calculated after the FCM is run.
Further, the sample is tested
Figure BDA0002914933750000044
To class center gammajMeasure of distance of
Figure BDA0002914933750000045
Expressed as:
Figure BDA0002914933750000046
Figure BDA0002914933750000047
wherein S isfjIs a fuzzy covariance matrix of class j,
Figure BDA0002914933750000048
further, the vegetable near infrared spectral data collected in S1 was preprocessed with multivariate scatter correction.
The invention has the beneficial effects that:
the analysis method provided by the invention is used for detecting four pesticide residues by utilizing a near infrared spectrum technology, solves the problem that the classification effect of the traditional hard clustering algorithm is not ideal, and has the characteristics of high clustering speed and high classification accuracy. In addition, the data is processed by adopting a fuzzy singular value decomposition method in the analysis process of the method, so that the problem of small samples of the conventional fuzzy linear discrimination method is solved.
Drawings
FIG. 1 is a general flow diagram of the process of the present invention;
FIG. 2 is a test sample set distribution plot;
FIG. 3 is a distribution plot of initial fuzzy membership values;
FIG. 4 is a fuzzy membership graph for the fuzzy covariance matrix clustering method.
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.
In the embodiment, the Chinese cabbage is taken as an example of the detection object, and the Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition is disclosed in fig. 1. The method specifically comprises the following steps:
and S1, acquiring the near infrared spectrum data of the cabbage sample to be analyzed, detecting the cabbage sample by using a Fourier near infrared spectrometer, acquiring the near infrared diffuse reflection spectrum data of the cabbage sample, and storing the spectrum data in a computer.
The Chinese cabbage samples in the embodiment are fully washed by warm water to ensure that no pesticide exists, and then the treated Chinese cabbage samples are divided into 4 groups; selecting high-efficiency cyhalothrin as a pesticide, and respectively treating 4 groups of Chinese cabbage samples by adopting pesticides with different proportions, wherein the 4 groups of Chinese cabbages sequentially comprise: group 0 is pesticide-free, and the ratio of group 1 pesticide to water is 1: 500, group 2 is 1: 100, group 3 is 1: 20. the temperature and relative temperature of the laboratory were kept constant. The Agilent Cary 630FTIR spectrometer was powered on and preheated for 1 h. And (3) acquiring the near infrared spectrum of the Chinese cabbage by adopting a reflection integrating sphere mode, and scanning each sample for 64 times by adopting the resolution of 8cm & lt-1 & gt.
And S2, preprocessing the collected near infrared spectrum data of the Chinese cabbage by using Multivariate Scattering Correction (MSC) to eliminate scattering influence and improve the signal-to-noise ratio of the data. And dividing the preprocessed near infrared spectrum data into training samples xi，i＝1,2,…,n1And a test specimen
Figure BDA0002914933750000057
t＝1,2,…,n2,n1For trainingNumber of training samples, n1＝120；n2For testing the number of samples, n2＝40。
And S3, extracting the identification information of the near infrared spectrum data of the Chinese cabbage by adopting a fuzzy singular value decomposition method.
S3.1, calculating the fuzzy membership of the training sample as follows:
Figure BDA0002914933750000051
wherein u isijFor training sample xiFuzzy degree of membership, v, belonging to class jjThe sample mean value of jth (j ═ 1,2,3,4) samples in the training sample set is obtained; c is the number of classes, c is 4, 1<c<n1，vkThe sample mean value of class k (k is 1,2,3,4) samples of the training samples is shown.
S3.2 calculating training samples x according to the following formulaiIs the matrix of dispersion S between fuzzy classesfBAnd the dispersion matrix S in the fuzzy classfWThe method specifically comprises the following steps:
Figure BDA0002914933750000052
Figure BDA0002914933750000053
wherein the content of the first and second substances,
Figure BDA0002914933750000054
to be the overall average of the training samples,
Figure BDA0002914933750000055
Figure BDA0002914933750000056
is a sample x with a weight index of miFuzzy membership belonging to class j; m is a weight index, and m is 2.
S3.3, based on training sample xiClass of fuzzyThe interval divergence matrix SfBAnd the dispersion matrix S in the fuzzy classfWRespectively construct a matrix HfWAnd HfBSatisfy the following requirements
Figure BDA0002914933750000061
This gives:
Figure BDA0002914933750000062
Figure BDA0002914933750000063
wherein the content of the first and second substances,
Figure BDA0002914933750000064
c is the set of fuzzy membership vectors for class j training samples,
Figure BDA0002914933750000065
is n of class j1Fuzzy membership degree vectors of the samples;
Figure BDA0002914933750000066
c is the set of products of the jth class of training samples with the square root of the corresponding degree of membership, e.g.
Figure BDA0002914933750000067
Is the n-th1Square root and nth of fuzzy membership vector of each sample1The product of the training samples;
Figure BDA0002914933750000068
c is the mean v of class j training samplesjThe set of products with the square root of each corresponding fuzzy membership,
Figure BDA0002914933750000069
is the mean of the training samples, e is the unit column matrix, expressed as
Figure BDA00029149337500000610
X is a matrix of training samples,
Figure BDA00029149337500000611
d is the sample dimension;
Figure BDA00029149337500000612
is a set of products of class c training samples and the square roots of the corresponding degrees of membership;
Figure BDA00029149337500000613
is the set of products of the class c training sample mean and the square root of each corresponding fuzzy membership.
S3.4, from matrix HfWAnd HfBConstruction matrix
Figure BDA00029149337500000614
And performing singular value decomposition on the matrix M to obtain
Figure BDA00029149337500000615
R is a diagonal matrix and Q is a unitary matrix obtained by singular value decomposition of the matrix M, i.e. QTQ is I, and I is an identity matrix; p is an orthogonal matrix.
S3.5, carrying out singular value decomposition on the matrix P to obtain P ═ U Σ VT. V is a unitary matrix obtained by decomposing the singular value of P; u is an orthogonal matrix; Σ is a diagonal matrix.
S3.6, constructing a matrix based on the unitary matrix Q, the diagonal matrix R, the unitary matrix V and the unit matrix I obtained by the formula
Figure BDA00029149337500000616
The matrix formed by the first 3 columns of vectors of the matrix W is G, and the obtained G is a transformation matrix.
S3.7, using transformation matrix G, for the t (t is 1,2, …, n) th2) A test specimen
Figure BDA00029149337500000617
And the ith (i ═ 1,2, …, n1) A training sample xiTo carry outThe following transformations are carried out:
Figure BDA00029149337500000618
yi＝xiG。
s4, Linear Discriminant Analysis (LDA) is used to test the sample in S3.7
Figure BDA0002914933750000071
And training sample yiRespectively converted into test samples
Figure BDA0002914933750000072
And training sample zi. Test specimen
Figure BDA0002914933750000073
The distribution is shown in fig. 2.
S5, performing spectral data clustering analysis by using a fuzzy covariance matrix clustering method, wherein the specific process is as follows:
s5.1, for the test sample converted in S4
Figure BDA0002914933750000074
Obtaining a fuzzy membership value u which is subordinate to the jth class after fuzzy C-means clustering (FCM) is operatedjt,FCMAnd class-center value v of class jj,FCMAnd fuzzy membership value u is calculatedjt,FCMAnd class center value vj,FCMAs the initial fuzzy membership value and the initial class center value of the subsequent fuzzy cluster. Fuzzy membership value ujt,FCMAs shown in fig. 3.
Establishing a fuzzy clustering objective function:
Figure BDA0002914933750000075
Figure BDA0002914933750000076
is a test specimen
Figure BDA0002914933750000077
To class center vj,FCMA distance measure of (d); d is the dimension of the test sample; sfj，FCMIs a fuzzy covariance matrix calculated after the FCM is run, and is expressed as:
Figure BDA0002914933750000078
wherein the content of the first and second substances,
Figure BDA0002914933750000079
to test the sample
Figure BDA00029149337500000710
Running fuzzy C-means clustering (FCM) and then measuring sample book
Figure BDA00029149337500000711
Fuzzy membership value belonging to j class, m is weight index;
s5.2 calculating parameters
Figure BDA00029149337500000712
Wherein m is a weight index.
Figure BDA00029149337500000713
Is a test specimen
Figure BDA00029149337500000714
To class center
Figure BDA00029149337500000718
Is expressed as:
Figure BDA00029149337500000715
vs,FCMobtaining a cluster center belonging to the s (s is 1,2, …, c) th class after FCM operation; sfs,FCMIs the fuzzy covariance matrix calculated after the FCM is run.
S5.3, based on the parameters calculated in step S5.2
Figure BDA00029149337500000716
For the test sample in S4
Figure BDA00029149337500000717
Performing iterative computation, and classifying the Chinese cabbages according to the fuzzy membership value at the end of iteration, wherein the specific iterative process is as follows:
s5.3.1, calculating fuzzy membership:
Figure BDA0002914933750000081
in the above formula, test specimens
Figure BDA0002914933750000082
To class center gammajMeasure of distance of
Figure BDA0002914933750000083
γjIs class center of class j, SfjIs a fuzzy covariance matrix of class j, expressed as:
Figure BDA0002914933750000084
μjtis a test specimen
Figure BDA0002914933750000085
Belonging to class centre gammajFuzzy membership of (d); .
S5.3.2 calculate class center:
Figure BDA0002914933750000086
γjis the class center of class j.
And after the iteration is ended, classifying the near infrared spectrum of the Chinese cabbage according to the fuzzy membership value obtained by calculation, wherein the fuzzy membership after the iteration is shown in figure 4.
The above embodiments are only used for illustrating the design idea and features of the present invention, and the purpose of the present invention is to enable those skilled in the art to understand the content of the present invention and implement the present invention accordingly, and the protection scope of the present invention is not limited to the above embodiments. Therefore, all equivalent changes and modifications made in accordance with the principles and concepts disclosed herein are intended to be included within the scope of the present invention.

Claims (9)

1. A Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition is characterized by comprising the following steps:
s1, collecting near infrared spectrum data of the vegetable sample to be analyzed; and dividing the near infrared spectral data into training samples xiAnd a test specimen
Figure FDA0002914933740000016
S2, extracting identification information of near infrared spectrum data of the vegetables by adopting a fuzzy singular value decomposition method;
s3, respectively converting the test sample and the training sample in the S2 by adopting a linear discriminant analysis method;
and S4, performing spectral data clustering analysis on the test sample and the training sample subjected to conversion in the S3 by adopting a fuzzy covariance matrix clustering method.
2. The Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition as claimed in claim 1, wherein the method for extracting identification information in S2 is as follows:
s2.1, calculating the fuzzy membership u of the training sampleij：
Figure FDA0002914933740000011
S2.2, fuzzy membership u based on training samplesijSeparately compute training samples xiIs the matrix of dispersion S between fuzzy classesfBAnd the dispersion matrix S in the fuzzy classfW；
S2.3, based on the dispersion matrix S between fuzzy classesfBAnd the dispersion matrix S in the fuzzy classfWRespectively construct a matrix HfWAnd HfB；
S2.4, from matrix HfWAnd HfBConstruction matrix
Figure FDA0002914933740000012
Performing singular value decomposition on the matrix M to obtain a diagonal matrix R and a unitary matrix Q;
s2.5, performing singular value decomposition on the matrix P to obtain a unitary matrix V,
s2.6, constructing a matrix based on a unitary matrix Q, a diagonal matrix R, a unitary matrix V and an identity matrix I which are obtained by singular value decomposition
Figure FDA0002914933740000013
A transformation matrix G formed by the first 3 columns of vectors of the matrix W;
s2.7, respectively aligning the test samples by using the transformation matrix G
Figure FDA0002914933740000014
And training sample xiTransforming to obtain transformed test samples
Figure FDA0002914933740000015
Transformed training sample yi＝xiG。
3. The Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition as claimed in claim 2, wherein in S3, a linear discriminant analysis method is adopted to analyze a test sample
Figure FDA0002914933740000021
And training sample yiRespectively converted into test samples
Figure FDA0002914933740000022
And training sample zi。
4. The Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition is characterized in that the method for performing cluster analysis on spectral data comprises the following steps:
s4.1, testing samples converted in S3
Figure FDA0002914933740000023
Obtaining a fuzzy membership value u which is subordinate to the jth class after running fuzzy C mean value clusteringjt,FCMAnd class-center value v of class jj,FCMAnd will ujt,FCMAnd vj,FCMAs initial fuzzy membership value and initial class center value of subsequent fuzzy clustering; establishing a fuzzy clustering objective function:
Figure FDA0002914933740000024
wherein the content of the first and second substances,
Figure FDA0002914933740000025
is a test specimen
Figure FDA0002914933740000026
To class center vj,FCMA distance measure of (d); d is the dimension of the test sample; sfj，FCMIs a fuzzy covariance matrix calculated after the FCM is operated;
s4.2, calculating parameters
Figure FDA0002914933740000027
Wherein the content of the first and second substances,
Figure FDA0002914933740000028
is a test specimen
Figure FDA00029149337400000217
To class center vs,FCMA distance measure of (d);
s4.3, based on the parameters calculated in step S4.2
Figure FDA0002914933740000029
For the test sample
Figure FDA00029149337400000210
And performing iterative calculation, and classifying the Chinese cabbages according to the fuzzy membership value at the end of the iteration.
5. The Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition according to claim 4, characterized in that the iteration process in S4.3 is as follows:
s4.3.1, calculating test sample
Figure FDA00029149337400000211
Belonging to class centre gammajFuzzy membership of (d):
Figure FDA00029149337400000212
wherein the content of the first and second substances,
Figure FDA00029149337400000213
is a test specimen
Figure FDA00029149337400000214
To class j centre gammajA measure of the distance of (a) is,
Figure FDA00029149337400000215
is a test specimen
Figure FDA00029149337400000216
To class j centre gammajA distance measure of (d);
s4.3.2 calculate class center:
Figure FDA0002914933740000031
and after the iteration is ended, classifying the near infrared spectrum of the vegetables according to the fuzzy membership value obtained by calculation.
6. The Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition as claimed in claim 4, wherein S isfj，FCMIs a fuzzy covariance matrix calculated after the FCM is run, and is expressed as:
Figure FDA0002914933740000032
wherein the content of the first and second substances,
Figure FDA0002914933740000033
to test the sample
Figure FDA0002914933740000034
Test sample after running fuzzy C-means clustering
Figure FDA0002914933740000035
And (4) fuzzy membership values belonging to the j-th class, wherein m is a weight index.
7. The Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition as claimed in claim 4, characterized in that a test sample
Figure FDA0002914933740000036
To class center vs,FCMMeasure of distance of
Figure FDA0002914933740000037
Expressed as:
Figure FDA0002914933740000038
wherein v iss,FCMObtaining a clustering center belonging to the s-th class after the FCM is operated; sfs,FCMIs the fuzzy covariance matrix calculated after the FCM is run.
8. The Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition as claimed in claim 5, characterized in that a test sample
Figure FDA0002914933740000039
To class center gammajMeasure of distance of
Figure FDA00029149337400000310
Expressed as:
Figure FDA00029149337400000311
Figure FDA00029149337400000312
wherein S isfjIs a fuzzy covariance matrix of class j,
Figure FDA00029149337400000313
9. the qualitative analysis method for pesticide residues in Chinese cabbage according to any one of claims 1 to 8, characterized in that the vegetable near infrared spectrum data collected in S1 is preprocessed by multivariate scattering correction.
CN202110098774.2A 2021-01-25 2021-01-25 Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition Pending CN112801172A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110098774.2A CN112801172A (en) 2021-01-25 2021-01-25 Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110098774.2A CN112801172A (en) 2021-01-25 2021-01-25 Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition

Publications (1)

Publication Number Publication Date
CN112801172A true CN112801172A (en) 2021-05-14

Family

ID=75811623

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110098774.2A Pending CN112801172A (en) 2021-01-25 2021-01-25 Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition

Country Status (1)

Country Link
CN (1) CN112801172A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114993982A (en) * 2022-06-02 2022-09-02 震坤行工业超市(上海)有限公司 Method for calculating oil performance parameters and device for monitoring lubricating oil on line

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114993982A (en) * 2022-06-02 2022-09-02 震坤行工业超市(上海)有限公司 Method for calculating oil performance parameters and device for monitoring lubricating oil on line

Similar Documents

Publication Publication Date Title
Bock et al. From visual estimates to fully automated sensor-based measurements of plant disease severity: status and challenges for improving accuracy
Zhang et al. Application of near-infrared hyperspectral imaging for variety identification of coated maize kernels with deep learning
Kimuli et al. Utilisation of visible/near-infrared hyperspectral images to classify aflatoxin B1 contaminated maize kernels
Bai et al. Accurate prediction of soluble solid content of apples from multiple geographical regions by combining deep learning with spectral fingerprint features
Zhou et al. Non-destructive discrimination of the variety of sweet maize seeds based on hyperspectral image coupled with wavelength selection algorithm
Xie et al. Classification of tomatoes with different genotypes by visible and short-wave near-infrared spectroscopy with least-squares support vector machines and other chemometrics
CN106706546A (en) Analysis method for artificial intelligence learning materials on basis of infrared and Raman spectrum data
CN102830087A (en) Method for quickly identifying food waste oils based on near infrared spectroscopy
CN110378374B (en) Tea near infrared spectrum classification method for extracting fuzzy identification information
CN109685098B (en) Tea variety classification method for fuzzy inter-cluster separation and clustering
He et al. Fast discrimination of apple varieties using Vis/NIR spectroscopy
Chen et al. Combination of the manifold dimensionality reduction methods with least squares support vector machines for classifying the species of sorghum seeds
CN113008805A (en) Radix angelicae decoction piece quality prediction method based on hyperspectral imaging depth analysis
Yin et al. Non-destructive detection of foreign contaminants in toast bread with near infrared spectroscopy and computer vision techniques
He et al. Study on the identification of resistance of rice blast based on near infrared spectroscopy
Pourdarbani et al. Using metaheuristic algorithms to improve the estimation of acidity in Fuji apples using NIR spectroscopy
CN112801172A (en) Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition
CN113310934A (en) Method for quickly identifying milk cow milk mixed in camel milk and mixing proportion thereof
Wang et al. SVM classification method of waxy corn seeds with different vitality levels based on hyperspectral imaging
Du et al. Quantitative detection of azodicarbonamide in wheat flour by near-infrared spectroscopy based on two-step feature selection
Tang et al. A study on nitrogen concentration detection model of rubber leaf based on spatial-spectral information with NIR hyperspectral data
Wu et al. Nitrogen status assessment for multiple cultivars of strawberries using portable FT-NIR spectrometers combined with cultivar recognition and multivariate analysis
Li et al. SVM-based apple classification of soluble solids content by near-infrared spectroscopy
CN112801173A (en) Lettuce near infrared spectrum classification method based on QR fuzzy discrimination analysis
Wang et al. Detection of Insect-Damaged maize seeds using hyperspectral imaging and hybrid 1D-CNN-BiLSTM models

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