CN111595805A - Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method - Google Patents

Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method Download PDF

Info

Publication number
CN111595805A
CN111595805A CN202010387633.8A CN202010387633A CN111595805A CN 111595805 A CN111595805 A CN 111595805A CN 202010387633 A CN202010387633 A CN 202010387633A CN 111595805 A CN111595805 A CN 111595805A
Authority
CN
China
Prior art keywords
cabbage
chinese cabbage
possibility
sample
iteration
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
CN202010387633.8A
Other languages
Chinese (zh)
Inventor
武斌
周浩祥
武小红
贾红雯
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
Chuzhou Vocational and Technical College
Original Assignee
Jiangsu University
Chuzhou Vocational and Technical College
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, Chuzhou Vocational and Technical College filed Critical Jiangsu University
Priority to CN202010387633.8A priority Critical patent/CN111595805A/en
Publication of CN111595805A publication Critical patent/CN111595805A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/25Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
    • G01N21/31Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
    • G01N21/35Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
    • G01N21/3563Investigating 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
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/23Clustering techniques
    • G06F18/232Non-hierarchical techniques
    • G06F18/2321Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Probability & Statistics with Applications (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

The invention discloses a cabbage pesticide residue qualitative analysis method based on possibility clustering, and S1, the method comprises the following steps of collecting Chinese cabbage sample mid-infrared spectrum data with different pesticide residue levels: acquiring infrared spectrum data in Chinese cabbage samples with different pesticide residue levels by applying an attenuated total reflection mode of an Agilent Cary630FTIR spectrometer; s2, performing dimensionality reduction on the mid-infrared spectrum data of the sample: the acquired mid-infrared spectral data of the sample is compressed by Principal Component Analysis (PCA). According to the method, the fuzzy maximum value possibility-based evaluation standard distance measure is adopted, so that infrared spectrum data in the Chinese cabbage samples in various forms, sizes and densities can be clustered more accurately; the fuzzy membership value and the typical value can be generated simultaneously to analyze the Chinese cabbage mid-infrared spectrum data; the method has the advantages of low detection cost, high identification speed, high classification accuracy and the like, and is suitable for qualitative analysis of pesticide residues in the Chinese cabbage and other vegetables.

Description

Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method
Technical Field
The invention relates to the technical field of cabbage pesticide residue qualitative analysis, in particular to a cabbage pesticide residue qualitative analysis method based on probability clustering.
Background
The Chinese cabbage is rich in various nutritional ingredients, has high medicinal value and plays an important role in the vegetable market. Various pesticides are widely applied to the planting process of Chinese cabbage due to the high-efficiency, reliable insect-resistant and yield-increasing effects of the pesticides. However, the Chinese cabbage with high pesticide residue taken for a long time can cause chronic diseases and even death. Therefore, there is an urgent need to find a rapid, effective and nondestructive method for qualitatively analyzing pesticide residues in Chinese cabbage. Although the chemical analysis technology can obtain the cabbage pesticide residue result with high accuracy and high reliability, the problems of high cost, low efficiency, complex steps, time and labor consumption, large product damage, serious pollution and the like generally exist. Therefore, the chemical analysis technology is severely limited in qualitative analysis of the pesticide residue in the Chinese cabbage.
The mid-infrared spectrum technology has the advantages of convenience, rapidness, high efficiency, no damage, low cost and the like, and is widely applied to various fields as a detection tool. The mid-infrared spectrum has a wave number in the range of 4000cm-1 to 400cm-1, and the fundamental frequency of chemical bond vibration of most inorganic compounds and organic compounds is in this region. The mid-infrared absorption spectra are different for different functional groups in the molecule, classes of compounds and stereo structures of compounds. In fact, the content of organic matters in the cabbage samples with different pesticide residue levels is different, so that the mid-infrared spectrum is different, and the possibility of applying the mid-infrared spectrum technology to qualitatively analyze the cabbage pesticide residue is provided.
Improved possible C mean value (IPCM) clustering method
(Zhang, J. -S., Leung, Y. -W., Improdpsisibilistic-means statistical algorithms, IEEETranss. fuzzy systems,2004,12(2): 209-; because the target function of the IPCM uses the Euclidean distance measure, the IPCM is difficult to obtain ideal clustering effect when processing sample data with different shapes, sizes and densities.
The method comprises the steps of collecting Chinese cabbage mid-infrared spectrum data with different pesticide residue levels by using a spectrometer, sequentially carrying out dimensionality reduction processing and characteristic information extraction on the collected sample spectrum data through Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA), wherein the boundary of the Chinese cabbage spectrum data is often irregular, and the classification accuracy is low if the IPCM based on the Euclidean distance is applied to cluster analysis of the Chinese cabbage mid-infrared spectrum data. Therefore, a method for qualitatively analyzing pesticide residues of Chinese cabbage with possibility of clustering is needed to solve the problems.
Disclosure of Invention
Based on the technical problems existing in the background art, the invention aims at the defects existing in the traditional IPCM clustering method when clustering Chinese cabbage spectral data with different pesticide residue levels, the IPCM clustering method is scientifically combined with the Gath-Geva clustering method, and a Chinese cabbage pesticide residue qualitative analysis method of possibility clustering is designed. Meanwhile, the invention also has the advantages of low detection cost, high identification speed, high classification accuracy and the like.
The invention provides a cabbage pesticide residue qualitative analysis method capable of clustering possibilities, which comprises the following steps:
s1, collecting infrared spectrum data of Chinese cabbage samples with different pesticide residue levels: acquiring infrared spectrum data in Chinese cabbage samples with different pesticide residue levels by applying an attenuated total reflection mode of an Agilent Cary630FTIR spectrometer;
s2, performing dimensionality reduction on the mid-infrared spectrum data of the sample: compressing the collected infrared spectrum data in the sample through Principal Component Analysis (PCA);
s3, extracting characteristic information of the mid-infrared spectrum data subjected to dimensionality reduction in the S2: extracting characteristic information of infrared spectrum data in the reduced-dimension sample through Linear Discriminant Analysis (LDA);
s4, identifying the varieties of the Chinese cabbage samples with different pesticide residue levels: and (4) performing cluster analysis on the cabbage test set data extracted by the characteristic information in the S3 by using a possibility-clustering cabbage pesticide residue qualitative analysis method, so as to classify cabbage samples with different pesticide residue levels.
Further, in step S1, during the whole spectrum collection process, the temperature, humidity and other external environments of the laboratory should be kept constant as much as possible; meanwhile, the obtained mid-infrared spectrum data of the cabbage sample is 1236-dimensional data with the wave number range of 4000cm < -1 > -400cm < -1 >.
Further, S4.1, initializing, setting a threshold > 0, fuzzy weighting parameter m, w ∈ (1, + ∞), class number C, and determining the maximum iteration number rmaxAnd an initial iteration counter r0(ii) a Setting the number of training sets n _ training and test sets n _ t est; the dimension of the Chinese cabbage sample test set is d; taking the mean value of the Chinese cabbage sample test set data extracted by the characteristic information in the S3 as the initial clustering center value of the fuzzy C-means clustering
Figure BDA0002484382380000031
The final fuzzy membership value and the clustering center value obtained by running fuzzy C-means clustering are respectively used as the initial fuzzy membership value of a possibility clustering method
Figure BDA0002484382380000041
And initial cluster center
Figure BDA0002484382380000042
Further, S4.2, the r (r ═ 1, 2.., r) is calculatedmax) Distance norm at sub-iteration;
Figure BDA0002484382380000043
Dikis a sample xkTo the center of the cluster viThe distance norm of (d); sfiIs a fuzzy covariance matrix of the ith sample class, and
Figure BDA0002484382380000044
piin order to be a priori at all,
Figure BDA0002484382380000045
n and C represent the number and the category number of the sample data respectively; v. ofi(r-1) is the cluster center value of the (r-1) th iteration calculation; fuzzy membership value uik(r-1) denotes the r-1 st iteration sample xkFuzzy membership values belonging to class i.
Further, S4.3, the r (r ═ 1, 2.., r) is calculatedmax) Typical values at the time of the second iteration:
Figure BDA0002484382380000046
tik(r) is the fuzzy membership value calculated for the r-th iteration;
s4.4, the r (r ═ 1, 2., r) is calculatedmax) Fuzzy membership value at sub-iteration:
Figure BDA0002484382380000051
uik(r) is the fuzzy membership value calculated for the r-th iteration.
Further, S4.4, the r (r ═ 1, 2.., r) is calculatedmax) Fuzzy membership value at sub-iteration:
Figure BDA0002484382380000052
uik(r) is the fuzzy membership value calculated for the r-th iteration.
Further, S4.5, the r (r ═ 1, 2.., r) is calculatedmax) Cluster center value of ith class at time of sub-iteration
Figure BDA0002484382380000053
Wherein
Figure BDA0002484382380000054
Is the r th timeIteratively calculated clustering center viA value of (d); the cluster center matrix V (r) is composed of C cluster center values, and
Figure BDA0002484382380000055
further, S4.6, the cycle count is incremented, i.e. r ═ r + 1; if the condition is satisfied: i V(r)-V(r-1)| | | < less or r > rmaxAnd (4) stopping the calculation, otherwise continuing S4.2, and classifying the Chinese cabbage samples with different pesticide residue levels according to the fuzzy membership value and the typical value obtained by calculation.
The invention has the following beneficial effects:
according to the cabbage pesticide residue qualitative analysis method based on probability clustering, the fuzzy maximum probability-based standard distance measure is adopted, so that infrared spectrum data in cabbage samples of various shapes, sizes and densities can be clustered more accurately. The method can simultaneously generate the fuzzy membership value and the typical value to analyze the Chinese cabbage mid-infrared spectrum data. The method has the advantages of low detection cost, high identification speed, high classification accuracy and the like, and is suitable for qualitative analysis of pesticide residues in the Chinese cabbage and other vegetables.
Drawings
FIG. 1 is a flow chart of a method for qualitatively analyzing pesticide residues of Chinese cabbage with possibility of clustering;
FIG. 2 is a chart of mid-infrared spectra of cabbage samples at 4 pesticide residue levels for a qualitative analysis method of cabbage pesticide residue with possibility clustering;
FIG. 3 is a score chart of the first three principal component vectors (PC1, PC2 and PC3) of a possibility clustering Chinese cabbage pesticide residue qualitative analysis method;
FIG. 4 is a score chart of three best discrimination vectors of a Chinese cabbage pesticide residue qualitative analysis method of possibility clustering;
FIG. 5 is an initial fuzzy membership value of a qualitative analysis method of pesticide residues of Chinese cabbage with possibility clustering;
FIG. 6 is fuzzy membership degree generated by a qualitative analysis method of pesticide residue of Chinese cabbage with possibility clustering;
FIG. 7 is typical values generated by a qualitative analysis method of pesticide residue in Chinese cabbage with possibility clustering.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
As shown in fig. 1, a method for qualitatively analyzing pesticide residues of Chinese cabbage with possibility clustering comprises the following steps:
s1, collecting the mid-infrared spectrum data of the Chinese cabbage sample: aiming at the Chinese cabbage samples with different pesticide residue levels, scanning the Chinese cabbage samples by a spectrometer to obtain infrared spectrum information in the Chinese cabbage samples, and storing the spectrum information in a computer.
Fresh Chinese cabbage is purchased from supermarkets, and the grades of the Chinese cabbage samples are all first grade. The cabbage samples were thoroughly washed with 45 ℃ warm water to ensure that the surface was free of pesticides. The method comprises the steps of mixing high-efficiency cyhalothrin (produced by Waerda chemical Co., Ltd. Zhejiang) with clean water to obtain four solutions (no pesticide, the ratio of the high-efficiency cyhalothrin to water is 1: 500, 1: 100 and 1:20 respectively) with different concentrations, and uniformly and comprehensively spraying the four solutions on the surface of a Chinese cabbage sample. The number of samples of each Chinese cabbage is 40, and the total number of the samples is 160 Chinese cabbage samples. In the process of collecting infrared spectrum data in a Chinese cabbage sample, the temperature of a laboratory is ensured to be 25 +/-1 ℃ and the relative humidity is ensured to be 50-60% as far as possible. The detailed spectrum acquisition procedure is as follows: first, an AgilentCary630FTIR spectrometer (Agilent technologies, USA) is preheated for 1 hour after being started up; secondly, the spectrometer scans each cabbage sample for 64 times in an attenuated total reflection mode. And the resolution of the spectrometer was 8cm-1 and the background scan was set to 64 times. The wavenumber range of the collected spectra was 4000cm-1 to 400 cm-1. And finally, the obtained cabbage sample spectral data is 1236-dimensional high-dimensional data. Meanwhile, each cabbage sample is sampled for 3 times, and the mean value of the 3 samplings is stored in a computer, so that experimental data can be provided for the establishment of a subsequent model. And setting the class number C to be 4. FIG. 2 shows the IR spectrum of 4 samples of cabbage with pesticide residue levels.
S2, performing dimensionality reduction on the infrared spectrum data in the white cabbage sample: and (3) performing dimensionality reduction on the mid-infrared spectrum data of the Chinese cabbage sample in the S1 by using Principal Component Analysis (PCA).
And (3) compressing the mid-infrared spectrum data of the Chinese cabbage sample in the S1 by utilizing Principal Component Analysis (PCA). Because the accumulated credibility of the first 14 principal components is more than 98.1%, the Principal Component Analysis (PCA) is adopted to map the 1236-dimensional cabbage sample spectral data to the first 14-dimensional space, so that the number of dimensions is reduced from 1236 to 14; the first 14 eigenvalues are as follows: lambda [ alpha ]1=3.274、λ2=2.356、λ3=1.236、λ4=0.560、λ5=0.294、λ6=0.187、λ7=0.116、λ8=0.097、λ9=0.077、λ10=0.066、λ11=0.050、λ12=0.036、λ13=0.032、λ14=0.027、λ15=0.026、λ160.024. Furthermore, the first three principal components account for 81.3% of the total variance. The score plots for the first three principal component vectors (PC1, PC2, and PC3) are shown in FIG. 3.
S3, extracting characteristic information of the mid-infrared spectrum data of the cabbage sample: and (5) extracting characteristic information of the infrared spectrum data in the cabbage sample subjected to the dimensionality reduction treatment in the step S2 by adopting Linear Discriminant Analysis (LDA).
And (5) extracting characteristic information of the 14-dimensional spectral data subjected to the dimension reduction processing in the step S2 by using Linear Discriminant Analysis (LDA). 160 cabbage samples are divided into a training set and a testing set. The number of training sets was 120 (30 samples per pesticide residue level) and the number of test sets was 40 (10 samples per pesticide residue level). Since 4 cabbage samples of different pesticide residue levels were classified, the training set was processed to generate three best discrimination vectors. And projecting the 14-dimensional spectral data of the test set onto the generated three optimal discrimination vectors so as to obtain 3-dimensional sample data. Fig. 4 shows a score plot of the three best discrimination vectors.
And S4, using a possibility clustering method to the cabbage sample test set data extracted by the characteristic information in the S3 to identify cabbage samples with different pesticide residue levels. The method comprises the following specific steps:
s4.1, initializing: setting a fuzzy weight index m to be 3 and w to be 5; the category number C is 4; initial value r of iteration times 01 and maximum number of iterations rmaxThe error upper limit value is 0.00001; and (4) fuzzy C-means clustering (FCM) is carried out, and the clustering center and the fuzzy membership value after the FCM operation is terminated are used as the initial clustering center and the fuzzy membership value of a possible clustering method. Initial fuzzy membership value
Figure BDA0002484382380000091
As shown in fig. 4. Initial cluster center
Figure BDA0002484382380000092
Comprises the following steps:
Figure BDA0002484382380000093
s4.2, the r (r ═ 1, 2., r) is calculatedmax) Distance norm D at sub-iterationik
Figure BDA0002484382380000094
Figure BDA0002484382380000095
Figure BDA0002484382380000096
Wherein: dikIs a sample xkTo the center of the cluster viThe distance norm of (d); sfiIs the fuzzy covariance matrix of the ith sample class; p is a radical ofiIs a prior probability; :
Figure BDA0002484382380000101
represents the r-1 st iteration sample xkA fuzzy membership value belonging to class i;
Figure BDA0002484382380000102
is the clustering center value of the (r-1) th iteration calculation; n and C represent the number and the category number of the sample data respectively;
s4.3, the r ═ is calculated (r ═ 1, 2.., r)max) Typical value at sub-iteration
Figure BDA0002484382380000103
Figure BDA0002484382380000104
Wherein:
Figure BDA0002484382380000105
is a typical value calculated in the r-th iteration;
s4.4, the r (r ═ 1, 2., r) is calculatedmax) Fuzzy membership value at sub-iteration
Figure BDA0002484382380000106
Figure BDA0002484382380000107
Wherein:
Figure BDA0002484382380000108
is the fuzzy membership value of the r-th iteration calculation;
s4.5, the r (r ═ 1, 2., r) is calculatedmax) Cluster center value of ith class at time of next iteration
Figure BDA0002484382380000109
Figure BDA00024843823800001010
Wherein:
Figure BDA00024843823800001011
is the clustering center v of the r-th iteration calculationiIs composed of C cluster center values, and
Figure BDA0002484382380000111
representative values representing the r-1 th iteration;
s4.6, cycle count is increased, i.e., r + 1; if the condition is satisfied: i V(r)-V(r-1)I < or r > rmaxThe calculation is terminated, otherwise step S4.2 is continued.
Fuzzy membership value and typical value can be obtained according to the calculation result, and different apple categories can be finally realized by utilizing the fuzzy membership value and the typical value. The experimental results are as follows: at the end of the iteration r is 28, and the center of the final cluster
Figure BDA0002484382380000112
Comprises the following steps:
Figure BDA0002484382380000113
fuzzy membership degree at the end of iteration is shown in FIG. 6, and classification accuracy of apple varieties obtained according to the fuzzy membership degree is 85%; typical values at the end of the iteration are shown in fig. 7, and the classification accuracy of the apple varieties obtained according to the typical values is also 85%.
The principle on which the invention is based: research shows that the Chinese cabbage mid-infrared spectrum contains the content information of organic matters in the Chinese cabbage, and the Chinese cabbage with different pesticide residue levels has different mid-infrared spectra, so that the Chinese cabbage mid-infrared spectra with different pesticide residue levels can be classified by adopting a clustering method
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.

Claims (8)

1. A Chinese cabbage pesticide residue qualitative analysis method of possibility clustering is characterized by comprising the following steps:
s1, collecting infrared spectrum data of Chinese cabbage samples with different pesticide residue levels: acquiring infrared spectrum data in Chinese cabbage samples with different pesticide residue levels by applying an attenuated total reflection mode of an Agilent Cary630FTIR spectrometer;
s2, performing dimensionality reduction on the mid-infrared spectrum data of the sample: compressing the collected infrared spectrum data in the sample through Principal Component Analysis (PCA);
s3, extracting characteristic information of the mid-infrared spectrum data subjected to dimensionality reduction in the S2: extracting characteristic information of infrared spectrum data in the reduced-dimension sample through Linear Discriminant Analysis (LDA);
s4, identifying the varieties of the Chinese cabbage samples with different pesticide residue levels: and (4) performing cluster analysis on the cabbage test set data extracted by the characteristic information in the S3 by using a possibility-clustering cabbage pesticide residue qualitative analysis method, so as to classify cabbage samples with different pesticide residue levels.
2. The method for qualitatively analyzing pesticide residues of cabbage with possibility clusters according to claim 1, wherein in step S1, the temperature, humidity and other external environments of the laboratory are kept constant as much as possible during the whole spectrum collection process; meanwhile, the obtained mid-infrared spectrum data of the cabbage sample is 1236-dimensional data with the wave number range of 4000cm < -1 > -400cm < -1 >.
3. The method for qualitatively analyzing pesticide residues of cabbage with possibility clusters as claimed in claim 1, wherein S4.1, initializing, setting a threshold value > 0, fuzzy weighting parameter m, w ∈ (1, infinity), class number C, and determining maximum iteration number rmaxAnd an initial iteration counter r0(ii) a Setting the number of training sets n _ training and test sets n _ test; the dimension of the Chinese cabbage sample test set is d; taking the mean value of the Chinese cabbage sample test set data extracted by the characteristic information in the S3 as the initial clustering center value of the fuzzy C-means clustering
Figure FDA0002484382370000021
The final fuzzy membership value and the clustering center value obtained by running fuzzy C-means clustering are respectively used as the initial fuzzy membership value of a possibility clustering method
Figure FDA0002484382370000022
And initial cluster center
Figure FDA0002484382370000023
4. The method for qualitatively analyzing pesticide residues of cabbage with possibility clusters according to claim 1, characterized in that S4.2 is used for calculating the r (r ═ 1, 2.., r) of the Chinese cabbage with possibility clustersmax) Distance norm at sub-iteration:
Figure FDA0002484382370000024
Dikis a sample xkTo the center of the cluster viThe distance norm of (d); sfiIs a fuzzy covariance matrix of the ith sample class, and
Figure FDA0002484382370000025
piin order to be a priori at all,
Figure FDA0002484382370000026
n and C represent the number and the category number of the sample data respectively; v. ofi(r-1) is the cluster center value of the (r-1) th iteration calculation; fuzzy membership value uik(r-1) denotes the r-1 st iteration sample xkFuzzy membership values belonging to class i.
5. The method for qualitatively analyzing pesticide residues of cabbage with possibility clusters according to claim 1, characterized in that S4.3 is used for calculating the r (r ═ 1, 2.., r ═ 1, 2., r ·max) Typical values at the time of the second iteration:
Figure FDA0002484382370000031
tik (r)is the fuzzy membership value of the r-th iteration calculation;
s4.4, the r (r ═ 1, 2., r) is calculatedmax) Fuzzy membership value at sub-iteration:
Figure FDA0002484382370000032
uik (r)is the fuzzy membership value calculated for the r-th iteration.
6. The method for qualitatively analyzing pesticide residues of cabbage with possibility clusters according to claim 1, characterized in that S4.4 is used for calculating the r (r ═ 1, 2.., r ═ 1, 2., r ·max) Fuzzy membership value at sub-iteration:
Figure FDA0002484382370000033
uik (r)is the fuzzy membership value calculated for the r-th iteration.
7. The method for qualitatively analyzing pesticide residues of cabbage with possibility clusters according to claim 1, wherein S4.5 is used for calculating the r (r ═ 1, 2.., r ═ 1, 2., r ·max) Class i at sub-iterationCluster center value of
Figure FDA0002484382370000034
Figure FDA0002484382370000035
Wherein
Figure FDA0002484382370000036
Is the clustering center v of the r-th iteration calculationiA value of (d); clustering center matrix V(r)Is composed of C cluster center values, and
Figure FDA0002484382370000037
8. the method for qualitatively analyzing pesticide residues of cabbage with possibility clusters according to claim 1, wherein S4.6, the cycle count is increased, i.e. r-r + 1; if the condition is satisfied: i V(r)-V(r-1)I < or r > rmaxAnd (4) stopping the calculation, otherwise continuing S4.2, and classifying the Chinese cabbage samples with different pesticide residue levels according to the fuzzy membership value and the typical value obtained by calculation.
CN202010387633.8A 2020-05-09 2020-05-09 Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method Pending CN111595805A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010387633.8A CN111595805A (en) 2020-05-09 2020-05-09 Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010387633.8A CN111595805A (en) 2020-05-09 2020-05-09 Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method

Publications (1)

Publication Number Publication Date
CN111595805A true CN111595805A (en) 2020-08-28

Family

ID=72185367

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010387633.8A Pending CN111595805A (en) 2020-05-09 2020-05-09 Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method

Country Status (1)

Country Link
CN (1) CN111595805A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117809070A (en) * 2024-03-01 2024-04-02 唐山市食品药品综合检验检测中心(唐山市农产品质量安全检验检测中心、唐山市检验检测研究院) Spectral data intelligent processing method for detecting pesticide residues in vegetables

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117809070A (en) * 2024-03-01 2024-04-02 唐山市食品药品综合检验检测中心(唐山市农产品质量安全检验检测中心、唐山市检验检测研究院) Spectral data intelligent processing method for detecting pesticide residues in vegetables
CN117809070B (en) * 2024-03-01 2024-05-14 唐山市食品药品综合检验检测中心(唐山市农产品质量安全检验检测中心、唐山市检验检测研究院) Spectral data intelligent processing method for detecting pesticide residues in vegetables

Similar Documents

Publication Publication Date Title
CN109142317B (en) Raman spectrum substance identification method based on random forest model
US6253162B1 (en) Method of identifying features in indexed data
CN107818298B (en) General Raman spectrum feature extraction method for machine learning substance identification algorithm
CN108072626B (en) Asphalt brand identification method
CN109030407B (en) Apple variety classification method based on mixed fuzzy C-means clustering
CN103048273A (en) Fruit near infrared spectrum sorting method based on fuzzy clustering
US20220390374A1 (en) Method for extracting raman characteristic peaks employing improved principal component analysis
CN110378374B (en) Tea near infrared spectrum classification method for extracting fuzzy identification information
Tang et al. Apple bruise grading using piecewise nonlinear curve fitting for hyperspectral imaging data
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
CN104345045A (en) Chemical pattern recognition and near infrared spectrum-based similar medicinal material identification method
Deulkar et al. An automated tomato quality grading using clustering based support vector machine
Solak et al. A new method for classifying nuts using image processing and k‐means++ clustering
CN105223140A (en) The method for quickly identifying of homology material
CN110987866A (en) Gasoline property evaluation method and device
CN111595805A (en) Possibility-clustering Chinese cabbage pesticide residue qualitative analysis method
Shao et al. Determination of damaged wheat kernels with hyperspectral imaging analysis
CN108663334B (en) Method for searching spectral characteristic wavelength of soil nutrient based on multi-classifier fusion
CN114611582A (en) Method and system for analyzing substance concentration based on near infrared spectrum technology
CN113310934A (en) Method for quickly identifying milk cow milk mixed in camel milk and mixing proportion thereof
CN107290299B (en) Method for detecting sugar degree and acidity of peaches in real time in nondestructive mode
CN109685099B (en) Apple variety distinguishing method based on spectrum band optimization fuzzy clustering
CN111912823A (en) Multi-component pesticide residue fluorescence detection analysis method
CN111595804A (en) Fuzzy clustering tea near infrared spectrum classification method

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