Cinnamon identification method and equipment based on three-dimensional fluorescence spectrum, medium and product

Abstract

The invention discloses a cinnamon identification method and device based on a three-dimensional fluorescence spectrum, a medium and a product, and relates to the field of multidisciplinary interdiscipling.The method comprises the steps that three-dimensional fluorescence spectrum data of cinnamon to be identified are obtained; performing factorization on the three-dimensional fluorescence spectrum data of the cinnamon to be identified to obtain three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified; the three-dimensional fluorescence spectrum data of all the components of the cinnamon to be identified are input into a constructed cinnamon identification model, the cinnamon identification model outputs identification index parameters, and the index parameters comprise at least one of geology, authenticity and specification. The method has the advantages of high sensitivity, good reproducibility, high detection precision, high detection speed, low manpower and time cost, greenness, safety and the like.

Description

Technical Field

[0001] The present application relates to multidisciplinary cross-fields such as Chinese medicine identification, chemometrics, artificial intelligence and machine learning, and high-dimensional spectral analysis, and in particular to a cinnamon identification method, equipment, medium, and product based on three-dimensional fluorescence spectroscopy. Background Art

[0002] Cinnamon is widely used in the food and pharmaceutical industries. However, the content of active ingredients in cinnamon grown around the world varies, resulting in uneven quality. The content of cinnamaldehyde in cinnamon from some regions is generally lower than the requirements of the pharmacopoeia, resulting in a wide range of medicinal effects. Similarly, the quality and medicinal value of cinnamon of different specifications are also different. Therefore, it is necessary to accurately identify the authenticity, authenticity and specifications of cinnamon.

[0003] At present, the commonly used methods for identifying the authenticity, authenticity and specifications of cinnamon include near-infrared spectroscopy and liquid chromatography. However, the sensitivity and reproducibility of near-infrared spectroscopy are poor; the sample pretreatment of liquid chromatography is relatively complicated, resulting in slow detection speed and high labor and time costs. Summary of the invention

[0004] The purpose of the present application is to provide a cinnamon identification method, equipment, medium and product based on three-dimensional fluorescence spectroscopy to solve the problems of poor sensitivity and reproducibility, slow detection speed and high cost of existing cinnamon identification methods.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a cinnamon identification method based on three-dimensional fluorescence spectroscopy, comprising:

[0007] Acquiring three-dimensional fluorescence spectrum data of cinnamon to be identified;

[0008] Factoring the three-dimensional fluorescence spectrum data of the cinnamon to be identified to obtain three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified;

[0009] The three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is input into the constructed cinnamon identification model, and the cinnamon identification model outputs identification index parameters, and the index parameters include at least one of authenticity, truthfulness and specification.

[0010] Optionally, the SWATLD factor decomposition algorithm is used to factorize the three-dimensional fluorescence spectrum data of the cinnamon to be identified.

[0011] Optionally, the cinnamon identification method based on three-dimensional fluorescence spectroscopy further includes:

[0012] Before factoring the three-dimensional fluorescence spectrum data of the cinnamon to be identified, background and scattered noise of the three-dimensional fluorescence spectrum data of the cinnamon to be identified are removed.

[0013] Optionally, the cinnamon identification method based on three-dimensional fluorescence spectroscopy further includes:

[0014] Before inputting the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified into the constructed cinnamon identification model, the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is subjected to data preprocessing, and the data preprocessing includes natural logarithm transformation, square root transformation or scaling.

[0015] Optionally, the cinnamon identification model includes at least one of a cinnamon authenticity identification classification model, a cinnamon authenticity identification classification model, and a cinnamon specification identification classification model.

[0016] Optionally, the cinnamon authenticity identification classification model, the cinnamon authenticity identification classification model, and the cinnamon specification identification classification model respectively adopt supervised Bayesian classification models.

[0017] Optionally, the cinnamon identification model is constructed according to the following process:

[0018] Acquiring three-dimensional fluorescence spectrum data of cinnamon samples, wherein the cinnamon samples include cinnamon of different specifications, cinnamon from different origins, genuine cinnamon and fake cinnamon;

[0019] Factoring the three-dimensional fluorescence spectrum data of each cinnamon sample to obtain three-dimensional fluorescence spectrum data of each component of each cinnamon sample;

[0020] The three-dimensional fluorescence spectrum data of each component of the cinnamon sample is used as sample data, the index parameters of the cinnamon sample are used as labels, the sample data is divided into a training set and a test set, and a cinnamon identification model capable of identifying the index parameters of the cinnamon sample is trained through supervised learning. During the training process, the quality of the cinnamon identification model is evaluated, and the model parameters are adjusted according to the evaluation results.

[0021] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any one of the above-described cinnamon identification methods based on three-dimensional fluorescence spectroscopy.

[0022] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the above-mentioned methods for identifying cinnamon based on three-dimensional fluorescence spectroscopy.

[0023] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned cinnamon identification methods based on three-dimensional fluorescence spectroscopy.

[0024] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0025] The present application provides a cinnamon identification method, device, medium and product based on three-dimensional fluorescence spectroscopy. The three-dimensional fluorescence spectrum data of the cinnamon to be identified is obtained by obtaining the three-dimensional fluorescence spectrum data of the cinnamon to be identified, and the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is factored, so as to obtain the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified, and the cinnamon is identified based on the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified. Compared with the existing near-infrared spectroscopy and liquid chromatography, the method has the advantages of high sensitivity, good reproducibility, high detection accuracy, fast detection speed, low manpower and time costs, green safety, etc., and solves the problems of poor sensitivity and reproducibility, slow detection speed and high cost of the existing cinnamon identification method; by inputting the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified into the constructed cinnamon identification model, the cinnamon identification model outputs the identification index parameters, and the index parameters include at least one of authenticity, authenticity and specification, so as to further improve the speed and accuracy of identifying the authenticity, authenticity and specification of cinnamon, and realize the rapid and accurate identification of the authenticity, authenticity and specification of cinnamon. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 This is a schematic diagram of a process of identifying cinnamon based on three-dimensional fluorescence spectroscopy in one embodiment of the present application;

[0028] Figure 2 This is the three-dimensional fluorescence spectrum of domestic cinnamon obtained;

[0029] Figure 3 The three-dimensional fluorescence spectrum of domestic cinnamon after removing background noise and its component analysis diagram after factor decomposition;

[0030] Figure 4 The three-dimensional fluorescence spectrum of Indian cinnamon after removing background noise and its component analysis diagram after factor decomposition;

[0031] Figure 5The three-dimensional fluorescence spectrum of Burmese cinnamon after removing background noise and its component analysis diagram after factor decomposition;

[0032] Figure 6 The three-dimensional fluorescence spectrum of Vietnamese cinnamon after removing background noise and its component analysis diagram after factor decomposition;

[0033] Figure 7 The three-dimensional fluorescence spectrum of the Vietnamese mountain laurel (fake) after removing the background noise and the component analysis diagram after factor decomposition;

[0034] Figure 8 The three-dimensional fluorescence spectrum of cinnamon twig (counterfeit) after removing background noise and its component analysis diagram after factor decomposition;

[0035] Fig. 9 The three-dimensional fluorescence spectrum of cinnamon about 5 years old after removing the background noise and its component analysis diagram after factor decomposition;

[0036] Fig.10 The three-dimensional fluorescence spectrum of cinnamon about 10 years old after removing the background noise and the component analysis diagram after factor decomposition;

[0037] Fig.11 The three-dimensional fluorescence spectrum of cinnamon about 30 years old after removing the background noise and its component analysis diagram after factor decomposition;

[0038] Fig.12 This is the three-dimensional fluorescence factor decomposition component spectrum of domestic cinnamon;

[0039] Fig.13 A schematic diagram of functional modules of a cinnamon identification device based on three-dimensional fluorescence spectroscopy provided in another embodiment of the present application;

[0040] Fig.14 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0042] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0043] In an exemplary embodiment, Figure 1As shown, a cinnamon identification method based on three-dimensional fluorescence spectroscopy is provided, comprising the following steps 101 to 104. Among them:

[0044] Step 101, obtaining three-dimensional fluorescence spectrum data of cinnamon to be identified, wherein the three-dimensional fluorescence spectrum data includes excitation wavelength, emission wavelength, and fluorescence intensity.

[0045] In the embodiment of the present application, a three-dimensional fluorescence spectrometer is used to obtain the three-dimensional fluorescence spectrum data of the cinnamon to be identified. The three-dimensional fluorescence spectrometer used is not specifically limited and can be selected according to actual needs, and the most suitable detection concentration of the reference substance and the cinnamon to be identified is selected according to the actual detection spectrum for detection. For example, UltraFLUOR 1000 , and obtain the three-dimensional fluorescence spectrum data of the cinnamon to be identified according to the following process:

[0046] Weigh 0.2500 g of the powder of the cinnamon bark to be identified, place it in a 25 mL volumetric flask, add methanol solution (chromatographic grade) to the scale, perform ultrasonic centrifugation (such as ultrasonic extraction for 10 min, and centrifuge at 3000 r / min for 10 min at room temperature after extraction), take the supernatant, and dilute it with purified water to a test solution with a concentration of 0.1 mg / mL;

[0047] The solution to be tested is placed in a cuvette and then placed in a three-dimensional fluorescence spectrometer for measurement. The parameters of the three-dimensional fluorescence spectrometer are set as follows: excitation wavelength λex: 200-500nm, bandwidth 5nm; emission wavelength λem: 200-700nm, bandwidth 5nm; photomultiplier diode (PMT) voltage: 600V, and three-dimensional fluorescence spectrum data of the cinnamon to be identified are collected, and the three-dimensional fluorescence spectrum data include fluorescence intensity (z-axis) information of the cinnamon to be identified at different excitation wavelengths (y-axis) and emission wavelengths (x-axis).

[0048] Step 102: factorize the three-dimensional fluorescence spectrum data of the cinnamon to be identified to obtain the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified.

[0049] Step 103, inputting the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified into the constructed cinnamon identification model, and the cinnamon identification model outputs identification index parameters, wherein the index parameters include at least one of authenticity, truthfulness and specifications, and the specifications include growing years.

[0050] In the embodiments of the present application, the three-dimensional fluorescence spectra of domestic cinnamon, Indian cinnamon, Burmese cinnamon, Vietnamese cinnamon, Vietnamese mountain cinnamon (fake), cassia twig (fake), cinnamon grown for about 5 years, cinnamon grown for about 10 years, and cinnamon grown for about 30 years after removing the background noise and the component analysis diagram after factor decomposition are shown as follows Figures 3 to 11As shown, it can be seen that there are characteristic differences in the three-dimensional fluorescence spectra of cinnamon of different specifications, different origins and authenticity and the component analysis diagram after factor decomposition thereof. Based on the characteristic differences, the embodiment of the present application uses the constructed cinnamon identification model to identify cinnamon, which can accurately identify the authenticity, specifications and origin of cinnamon.

[0051] Implement the above steps 101 to 103, use a three-dimensional fluorescence spectrometer to scan and obtain the three-dimensional fluorescence spectrum data of the cinnamon to be identified, factorize the obtained three-dimensional fluorescence spectrum data of the cinnamon to be identified, and obtain the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified, and input the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified into the constructed cinnamon identification model, and the cinnamon identification model outputs the identification index parameters, and the index parameters include at least one of authenticity, authenticity and specifications, so as to achieve the rapid and accurate identification of the authenticity, authenticity and specifications of cinnamon. The three-dimensional fluorescence spectrum data is used in conjunction with the cinnamon identification model for identification, which has the advantages of high selectivity, high sensitivity, good reproducibility, high detection accuracy, fast detection speed, low manpower and time cost, green safety, etc. In addition, the infrared spectroscopy method has high requirements for the content of known components of the sample, and the present application can identify cinnamon and counterfeit products when the content of each component in cinnamon and counterfeit products is low, and the identification range is wider and more practical.

[0052] In another exemplary embodiment of the present application, in order to accurately perform factor decomposition, the SWATLD (self-weighted alternating trilinear) factor decomposition algorithm is used to factorize the three-dimensional fluorescence spectrum data of the cinnamon to be identified.

[0053] In the embodiment of the present application, the extracts of K cinnamon sample test solutions are subjected to three-dimensional fluorescence spectrum scanning under the condition of selecting an excitation wavelength number of I and an emission wavelength number of J, and a series of fluorescence spectrum response matrices are obtained and collected in a three-dimensional array of size I×J×K, that is, a three-dimensional fluorescence spectrum matrix X. The matrix is ​​composed of a relative excitation spectrum matrix A (size of I×N), a relative emission spectrum matrix B (size of J×N), a relative concentration matrix C (size of K×N), and a residual matrix E. According to the linear addition principle, each element x of the three-dimensional fluorescence spectrum matrix X ijk The representation is as follows:

[0054]

[0055] Among them, x ijk is the fluorescence intensity of position (i, j, k) in matrix X, that is, the fluorescence intensity of the kth sample when the excitation wavelength is i and the emission wavelength is j; N is the number of components; a in represents the relative excitation intensity of position (i,n) in the relative excitation spectrum matrix A, that is, the relative excitation intensity of component n when the excitation wavelength is i; bjn Represents the relative emission intensity of position (i,n) in the relative emission spectrum matrix B, that is, the relative emission intensity of component n when the emission wavelength is j; c kn represents the relative concentration of position (k,n) in the relative concentration matrix C, that is, the relative concentration of component n in the kth sample; e ijk It is the residual term at position (i, j, k) in the residual matrix E, that is, the residual part that cannot be explained by the model in the measurement of the kth sample when the excitation wavelength is i and the emission wavelength is j.

[0056] SWATLD decomposes the three-dimensional fluorescence spectrum matrix X by using the alternating least squares principle and Moore-Penrose generalized inverse calculation based on T-SVD (truncation singular value decomposition). Let a (i) represents the row vector consisting of the elements in the i-th row of matrix A, b (j) represents the row vector consisting of the j-th row elements in matrix B, c (k) represents the row vector composed of the k-th row elements in the matrix C. When the three-dimensional fluorescence spectrum matrix data X is decomposed, the iterative formulas of A, B, and C are respectively composed of the following:

[0057]

[0058] X i.. =Bdiagm(a (i) )C T +E i.. ,i=1,…,I;

[0059] X .j. =Cdiagm(b (j) )A T +E .j. ,j=1,…,J;

[0060] X ..k =Adiagm(c (k) )B T +E ..k ,k=1,…,K;

[0061] Among them, X i.. represents the i-th horizontal plane matrix (J×K) of the three-dimensional fluorescence spectrum matrix X, .j. represents the j-th side matrix (K×I) of the three-dimensional fluorescence spectrum matrix X, X ..k represents the kth positive matrix (I×J) of the three-dimensional fluorescence spectrum matrix X; diagm(a (i) )、diagm(b (j) )、diagm(c (k) ) means taking the diagonal elements of matrices A, B, and C.T , B T , C T Represents the transpose of matrices A, B, and C. A + , B + , C + represents the Moore-Penrose generalized inverse matrix of matrices A, B, and C, E i.. 、E .j. 、E ..k are the residual term (error term), which is a random variable used to represent the variability that cannot be explained by the model. (i) 、b (j) and c (k) , that is, we get a in 、b jn 、c kn , which is the three-dimensional fluorescence spectrum data of each component of cinnamon.

[0062] like Fig.12 As shown, the three-dimensional fluorescence factor decomposition component spectra of domestic cinnamon, among which the upper left figure is the excitation spectrum of each component, and the upper right figure is the emission spectrum of each component. It can be seen from the excitation spectrum and emission spectrum of each component that when the set component number deviates from the true value, the SWATLD factor decomposition algorithm can still parse and obtain the correct component excitation spectrum and emission spectrum, and present the redundant components as background noise. Fig.12 The lower left figure shows the distribution of fluorescence intensity (concentration response signal) of different components, and the lower right figure shows the factor decomposition iteration residual, indicating that the SWATLD factor decomposition algorithm has an extremely fast convergence speed, and the result usually tends to a stable optimal solution after 10 iterations.

[0063] The embodiment of the present application uses the SWATLD factor decomposition algorithm to decompose and correct the collected three-dimensional fluorescence spectrum data of cinnamon to be identified, and replaces or enhances chemical and physical separation with mathematical separation, saving time and cost.

[0064] In another exemplary embodiment of the present application, in order to accurately perform factor decomposition, the cinnamon identification method based on three-dimensional fluorescence spectroscopy further includes:

[0065] Before factoring the three-dimensional fluorescence spectrum data of the cinnamon to be identified, the background noise of the three-dimensional fluorescence spectrum data of the cinnamon to be identified is removed. The three-dimensional fluorescence spectrum of the domestic cinnamon before removing the background noise is as follows: Figure 2 As shown in the figure, the three-dimensional fluorescence spectrum of domestic cinnamon after removing background noise is as follows Figure 3 shown.

[0066] As an optional embodiment, removing the background and scattered noise of the three-dimensional fluorescence spectrum data of the cinnamon to be identified comprises:

[0067] (1) importing the three-dimensional fluorescence spectrum data of the cinnamon to be identified, and determining the dimensions of the data: excitation wavelength, emission wavelength, and fluorescence intensity;

[0068] (2) Prepare a pure water sample and obtain its three-dimensional fluorescence spectrum data as the background spectrum. It is necessary to ensure that the test conditions of the pure water sample and the cinnamon to be identified are consistent, including temperature, pH value, etc.;

[0069] (3) subtracting the background spectrum from the three-dimensional fluorescence spectrum data of the cinnamon to be identified;

[0070] (4) In order to eliminate the influence of Rayleigh scattering and Raman scattering on the three-dimensional fluorescence spectrum data, the Rayleigh scattering bandwidth is set to 80 nm, the secondary Rayleigh scattering bandwidth is set to 85 nm, and the Raman scattering bandwidth is set to 0 nm, and Rayleigh scattering and Raman scattering are deducted from the three-dimensional fluorescence spectrum data of the cinnamon to be identified;

[0071] (5) Using a cubic spline interpolation algorithm to interpolate the three-dimensional fluorescence spectrum of the cinnamon to be identified after deducting Rayleigh scattering and Raman scattering, to generate smooth and continuous three-dimensional fluorescence spectrum data of the cinnamon to be identified.

[0072] The embodiment of the present application adopts a cubic spline interpolation method to deduct the Raman and Rayleigh scattering generated during the collection process of the three-dimensional fluorescence spectrum of the cinnamon to be identified, and still obtains a complete three-dimensional fluorescence spectrum of the cinnamon to be identified, further obtains more complete cinnamon component information, and improves the accuracy of detection and identification.

[0073] In another exemplary embodiment of the present application, in order to improve the identification accuracy of cinnamon by the cinnamon identification model, before the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is input into the constructed cinnamon identification model, the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is subjected to data preprocessing, and the data preprocessing includes natural logarithm transformation, square root transformation or scaling.

[0074] In the embodiment of the present application, the data preprocessing method is selected according to the structural characteristics of the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified obtained by factor decomposition, so as to optimize the data and improve the quality, while ensuring that cinnamon samples of different groups can be effectively distinguished. Different data preprocessing methods are used to make the clustering and differentiation between cinnamon samples of different specifications, different origins, and true and false more obvious, thereby improving the classification accuracy of unknown samples.

[0075] As an optional implementation, the scaling includes any one of minimum-maximum scaling, UV scaling, UVN scaling, Pareto scaling, ParetoN scaling, homogenization, and normalization.

[0076] Due to the large individual differences (intra-group differences) between samples of the same type of cinnamon samples from different origins and different years, the model prediction results are poor. In addition, the differences in the measurement scale or measurement principle of different variables of the same cinnamon sample, such as excitation wavelength and fluorescence intensity, lead to differences in the distribution range of variables, affecting the model prediction results. In view of the above situation, the embodiment of the present application performs data preprocessing on the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified to eliminate the differences between the variables or between the variables, thereby improving the accuracy of the model prediction, so as to more accurately identify and distinguish cinnamon from different origins, different specifications, and authenticity.

[0077] In another exemplary embodiment of the present application, the cinnamon identification model is constructed according to the following process:

[0078] Acquiring three-dimensional fluorescence spectrum data of cinnamon samples, wherein the cinnamon samples include cinnamon of different specifications, cinnamon from different origins, genuine cinnamon and fake cinnamon;

[0079] Factoring the three-dimensional fluorescence spectrum data of each cinnamon sample to obtain three-dimensional fluorescence spectrum data of each component of each cinnamon sample;

[0080] The three-dimensional fluorescence spectrum data of each component of the cinnamon sample is used as sample data, the index parameters of the cinnamon sample are used as labels, the sample data is divided into a training set and a test set, and a cinnamon identification model capable of identifying the index parameters of the cinnamon sample is trained through supervised learning. During the training process, the quality of the cinnamon identification model is evaluated, and the model parameters are adjusted according to the quality of the cinnamon identification model.

[0081] In the embodiments of the present application, the cinnamon of different specifications, cinnamon of different origins, genuine cinnamon and fake cinnamon include genuine cinnamon of the same origin and different specifications, genuine cinnamon of different origins and the same specifications, genuine cinnamon and fake cinnamon of different origins, genuine cinnamon and fake cinnamon of the same origin and specifications. The genuine cinnamon of different specifications includes genuine cinnamon of different growing years, such as genuine cinnamon grown from 1 to 30 years, and genuine cinnamon grown from 1 to 40 years. The range of growing years covered by genuine cinnamon of different specifications is not specifically limited, and can be set according to the growing year requirements of the cinnamon to be identified.

[0082] The quality of the cinnamon identification model is evaluated, that is, at least one of the accuracy (the proportion of samples with correct classification to the total samples), precision (the proportion of samples classified as "true" that are actually "true"), recall (the proportion of samples that are actually "true" that are classified as "true"), F1 score (the harmonic mean of precision and recall), and mean square error of the cinnamon identification model is calculated.

[0083] In another exemplary embodiment of the present application, in order to improve the accuracy of the cinnamon identification model in identifying cinnamon, the cinnamon identification model includes at least one of a cinnamon authenticity identification classification model, a cinnamon authenticity identification classification model, and a cinnamon specification identification classification model. The cinnamon authenticity identification classification model is used to identify the authenticity of cinnamon, that is, to identify whether it is genuine cinnamon or fake cinnamon. The cinnamon authenticity identification classification model is used to identify the authenticity of cinnamon, that is, to identify the origin of cinnamon. The cinnamon specification identification classification model is used to identify the specifications of cinnamon, and the specifications of cinnamon include the growing years. The three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is input into at least one of the constructed cinnamon authenticity identification classification model, cinnamon authenticity identification classification model, and cinnamon specification identification classification model to obtain identification index parameters.

[0084] In another exemplary embodiment of the present application, in order to improve the accuracy of the cinnamon identification model for identifying cinnamon, the cinnamon authenticity identification classification model, the cinnamon authenticity identification classification model, and the cinnamon specification identification classification model are respectively trained using a supervised Bayesian classification model, and the identification results of the authenticity of cinnamon are shown in Table 1, the identification results of the origin of cinnamon are shown in Table 2, and the identification results of the specifications of cinnamon are shown in Table 3. The results show that the Bayesian classification model has a prediction accuracy of 100% for the authenticity, origin, and specifications of cinnamon. In Tables 1 to 3, the number after each sample number (for example, 249 in Thanh Hoa 249, Vietnam) is the code corresponding to each cinnamon sample, and the code varies depending on the origin, authenticity, and specifications of the cinnamon sample, so as to distinguish all cinnamon samples.

[0085] Table 1 Data on the effectiveness of the supervised Bayesian classification model in identifying the authenticity of cinnamon

[0086] Sample No. Actual classification Prediction Classification Vietnam Thanh Hoa 249 Authentic Authentic Vietnam Thanh Hoa 250 Authentic Authentic Vietnam Thanh Hoa 251 Authentic Authentic Domestic cinnamon 10 Counterfeit Counterfeit Domestic cinnamon twig 54 Counterfeit Counterfeit Domestic cinnamon twig 55 Counterfeit Counterfeit Domestic Dongxing 118 Authentic Authentic Domestic Dongxing 120 Authentic Authentic Domestic Dongxing 117 Authentic Authentic Vietnamese Mountain Gui 321 Counterfeit Counterfeit Vietnamese Mountain Gui 322 Counterfeit Counterfeit Vietnamese Mountain Gui 322 Counterfeit Counterfeit

[0087] Table 2 Data on the identification effect of supervised Bayesian classification model on cinnamon origin

[0088] Sample No. Origin Predicted Origin Domestic Dongxing 118 China China Domestic Dongxing 120 China China India 247 India India India 248 India India Myanmar 217 Myanmar Myanmar Myanmar 221 Myanmar Myanmar Vietnam Thanh Hoa 252 Vietnam Vietnam Vietnam Thanh Hoa 257 Vietnam Vietnam

[0089] Table 3 The effect of the supervised Bayesian classification model on the identification of cinnamon specifications

[0090]

[0091] The embodiment of the present application achieves automated and standardized accurate identification of cinnamon of different origins, different specifications, and authenticity by establishing different machine learning models for authenticity, truthfulness, and specification identification.

[0092] As an optional implementation, the supervised Bayesian classification model identifies various index parameters (authenticity, authenticity, specifications) of cinnamon according to the following process:

[0093]

[0094] Where x is the given input data, x=(x1,x2,...,x n ), n is the total number of feature vectors in the input data x; C k is the category label corresponding to each indicator parameter. In the authenticity identification of cinnamon, the category labels are genuine and fake, in the authenticity identification of cinnamon, the category label is the origin, and in the specification identification of cinnamon, the category label is the year; P(C k |x) is a given input data x, judging the category of a certain indicator parameter as C k The posterior probability of k ) is a category of an indicator parameter of C k The prior probability of a certain indicator parameter is C k The proportion of cinnamon in the training set; P(x) is the marginal probability of input data x. The marginal probability P(x) of all categories of each indicator parameter is the same, which is a constant used to normalize the posterior probability P(C k |x);

[0095] The following formula is used to calculate P(C k |x):

[0096]

[0097] Among them, x i is the i-th eigenvector, P(x i |C k ) is the category of each feature under a certain indicator parameter C k The conditional probability of .

[0098] Choose the posterior probability P(C k |x)The largest category is used as the prediction result of the supervised Bayesian classification model for a certain indicator parameter, namely the authenticity, authenticity and specifications of the identified cinnamon.

[0099] This application is based on three-dimensional fluorescence spectral data and uses the SWATLD factor decomposition algorithm combined with a Bayesian model to carry out machine learning of three-dimensional fluorescence spectra for a large number of cinnamon samples, forming an integrated and automated model training and prediction mechanism based on sample data, thereby achieving rapid identification of genuine cinnamon and counterfeit cinnamon of different origins and specifications.

[0100] Based on the same inventive concept, the embodiment of the present application also provides a cinnamon identification device based on three-dimensional fluorescence spectroscopy for implementing the above-mentioned cinnamon identification method based on three-dimensional fluorescence spectroscopy. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the cinnamon identification device based on three-dimensional fluorescence spectroscopy provided below can refer to the limitations of the cinnamon identification method based on three-dimensional fluorescence spectroscopy above, and will not be repeated here.

[0101] In an exemplary embodiment, Fig.13 As shown, a cinnamon identification device 20 based on three-dimensional fluorescence spectroscopy is provided, comprising:

[0102] A fluorescence spectrum acquisition module 201 is used to acquire three-dimensional fluorescence spectrum data of the cinnamon to be identified;

[0103] A factor decomposition module 202 is used to factorize the three-dimensional fluorescence spectrum data of the cinnamon to be identified to obtain the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified;

[0104] The cinnamon identification module 203 is used to input the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified into the constructed cinnamon identification model, and the cinnamon identification model outputs identification index parameters, and the index parameters include at least one of authenticity, truthfulness and specification.

[0105] As an optional implementation, the factor decomposition module 202 is further configured to:

[0106] The SWATLD factor decomposition algorithm is used to factorize the three-dimensional fluorescence spectrum data of the cinnamon to be identified. For details, please refer to the description of the embodiment of the cinnamon identification method based on three-dimensional fluorescence spectrum, which will not be repeated here.

[0107] As an optional implementation, the cinnamon identification device 20 based on three-dimensional fluorescence spectroscopy further includes:

[0108] The background removal module 204 is used to remove the background and scattered noise of the three-dimensional fluorescence spectrum data of the cinnamon to be identified before factoring the three-dimensional fluorescence spectrum data of the cinnamon to be identified. The relevant introduction of removing the background noise of the three-dimensional fluorescence spectrum data of the cinnamon to be identified is detailed in the embodiment of the cinnamon identification method based on three-dimensional fluorescence spectrum, which will not be repeated here.

[0109] As an optional implementation, the cinnamon identification device 20 based on three-dimensional fluorescence spectroscopy further includes:

[0110] The data preprocessing module 205 is used to perform data preprocessing on the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified before inputting the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified into the constructed cinnamon identification model, wherein the data preprocessing includes natural logarithm transformation, square root transformation or scaling. The relevant introduction of the scaling is detailed in the embodiment of the cinnamon identification method based on three-dimensional fluorescence spectrum, which will not be repeated here.

[0111] As an optional implementation, the cinnamon identification model includes at least one of a cinnamon authenticity identification classification model, a cinnamon authenticity identification classification model, and a cinnamon specification identification classification model; the cinnamon authenticity identification classification model is used to identify the authenticity of cinnamon, that is, to identify whether it is genuine cinnamon or fake cinnamon; the cinnamon authenticity identification classification model is used to identify the authenticity of cinnamon, that is, to identify the origin of cinnamon; the cinnamon specification identification classification model is used to identify the specifications of cinnamon, and the specifications include the growing years;

[0112] The cinnamon identification module 203 is also used for:

[0113] The three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is input into at least one of the constructed cinnamon authenticity identification classification model, cinnamon authenticity identification classification model, and cinnamon specification identification classification model to obtain identification index parameters.

[0114] As an optional implementation manner, the cinnamon authenticity identification classification model, the cinnamon authenticity identification classification model, and the cinnamon specification identification classification model are respectively trained using a supervised Bayesian classification model.

[0115] The cinnamon identification module 203 is also used for:

[0116] The three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is input into at least one of the cinnamon authenticity identification classification model, the cinnamon authenticity identification classification model, and the cinnamon specification identification classification model trained by the supervised Bayesian classification model to obtain identification index parameters. The relevant introduction of the process of identifying each index parameter (authenticity, authenticity, specification) by the supervised Bayesian classification model is detailed in the description of the embodiment of the cinnamon identification method based on three-dimensional fluorescence spectroscopy, which will not be repeated here.

[0117] As an optional implementation, the cinnamon identification device 20 based on three-dimensional fluorescence spectroscopy further includes:

[0118] The model building module 206 is used to build the cinnamon identification model according to the following process:

[0119] Acquiring three-dimensional fluorescence spectrum data of cinnamon samples, wherein the cinnamon samples include cinnamon of different specifications, cinnamon from different origins, genuine cinnamon and fake cinnamon;

[0120] Factoring the three-dimensional fluorescence spectrum data of each cinnamon sample to obtain three-dimensional fluorescence spectrum data of each component of each cinnamon sample;

[0121] The three-dimensional fluorescence spectrum data of each component of the cinnamon sample is used as sample data, the index parameters of the cinnamon sample are used as labels, the sample data is divided into a training set and a test set, and a cinnamon identification model capable of identifying the index parameters of the cinnamon sample is trained through supervised learning. During the training process, the quality of the cinnamon identification model is evaluated, and the model parameters are adjusted according to the evaluation results.

[0122] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Fig.14 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store video tag processing data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a cinnamon identification method based on three-dimensional fluorescence spectroscopy is implemented.

[0123] Those skilled in the art will understand that Fig.14 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0124] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0125] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0126] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0127] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0128] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0129] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0130] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0131] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A cinnamon identification method based on three-dimensional fluorescence spectroscopy, characterized in that: The cinnamon identification method based on three-dimensional fluorescence spectroscopy comprises: Acquiring three-dimensional fluorescence spectrum data of cinnamon to be identified; Factoring the three-dimensional fluorescence spectrum data of the cinnamon to be identified to obtain three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified; The three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is input into the constructed cinnamon identification model, and the cinnamon identification model outputs identification index parameters, and the index parameters include at least one of authenticity, truthfulness and specification.

2. The cinnamon identification method based on three-dimensional fluorescence spectroscopy according to claim 1, characterized in that: The SWATLD factor decomposition algorithm is used to factorize the three-dimensional fluorescence spectrum data of the cinnamon to be identified.

3. The cinnamon identification method based on three-dimensional fluorescence spectroscopy according to claim 1, characterized in that: Also includes: Before factoring the three-dimensional fluorescence spectrum data of the cinnamon to be identified, background and scattered noise of the three-dimensional fluorescence spectrum data of the cinnamon to be identified are removed.

4. The cinnamon identification method based on three-dimensional fluorescence spectroscopy according to claim 1, characterized in that: Also includes: Before inputting the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified into the constructed cinnamon identification model, the three-dimensional fluorescence spectrum data of each component of the cinnamon to be identified is subjected to data preprocessing, and the data preprocessing includes natural logarithm transformation, square root transformation or scaling.

5. The cinnamon identification method based on three-dimensional fluorescence spectroscopy according to claim 1, characterized in that: The cinnamon identification model includes at least one of a cinnamon authenticity identification classification model, a cinnamon authenticity identification classification model, and a cinnamon specification identification classification model.

6. The cinnamon identification method based on three-dimensional fluorescence spectroscopy according to claim 5, characterized in that: The cinnamon authenticity identification classification model, the cinnamon authenticity identification classification model, and the cinnamon specification identification classification model respectively adopt supervised Bayesian classification models.

7. The cinnamon identification method based on three-dimensional fluorescence spectroscopy according to any one of claims 1 to 6, characterized in that: The cinnamon identification model is constructed according to the following process: Acquiring three-dimensional fluorescence spectrum data of cinnamon samples, wherein the cinnamon samples include cinnamon of different specifications, cinnamon from different origins, genuine cinnamon and fake cinnamon; Factoring the three-dimensional fluorescence spectrum data of each cinnamon sample to obtain three-dimensional fluorescence spectrum data of each component of each cinnamon sample; The three-dimensional fluorescence spectrum data of each component of the cinnamon sample is used as sample data, the index parameters of the cinnamon sample are used as labels, the sample data is divided into a training set and a test set, and a cinnamon identification model capable of identifying the index parameters of the cinnamon sample is trained through supervised learning. During the training process, the quality of the cinnamon identification model is evaluated, and the model parameters are adjusted according to the evaluation results.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the cinnamon identification method based on three-dimensional fluorescence spectroscopy described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the cinnamon identification method based on three-dimensional fluorescence spectroscopy described in any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the cinnamon identification method based on three-dimensional fluorescence spectroscopy described in any one of claims 1 to 7 are implemented.

Images