Spectral data processing method, device and electronic equipment for tea pile

Abstract

The present invention provides a method, apparatus and electronic device for processing spectral data of a tea pile, which are applied to the technical field of rapid detection of the quality of tea buds in the field. The above method includes: inputting the mixed spectrum of the tea pile into a bilinear spectral model to obtain the front leaf spectrum, wherein the bilinear spectral model is used to correct the mixed spectrum of the tea pile; taking the derivative of the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, wherein taking the derivative is used to eliminate the tea bud stacking structure information of the tea pile; inputting the derivative set into a Gaussian process regression model using a radial basis kernel function to obtain the quality of the front leaf, wherein the radial basis kernel function in the Gaussian process regression model is used to eliminate the front leaf proportion information of the tea pile, and the Gaussian process regression model is used to predict the quality of the front leaf; through the present invention, the influence of the tea bud stacking structure and the difference in the proportion of the front leaf in the mixed spectrum of the tea pile on the accurate inversion of the tea bud quality can be eliminated.

Description

Technical Field

[0001] The present invention relates to the technical field of rapid detection of the quality of tea buds in the field, and particularly relates to a method, device and electronic device for processing spectral data of a tea pile. Background Art

[0002] The traditional chemical method for detecting the quality of tea buds not only takes a long time but also has strong destructiveness. In recent years, with the rapid development of hyperspectral technology, narrow bands can capture subtle changes in absorption characteristics and achieve accurate estimation of the content of physiological and biochemical components of vegetation. However, the establishment of these estimation models depends on the measurement of leaf spectra.

[0003] The tea buds of small-leaf tea trees are very small and the leaf margins are curled. The spectral information of a single tea bud is weak and it is difficult for existing non-imaging spectrometers to measure the spectrum of a single tea bud leaf. Therefore, only the mixed spectrum of the tea pile composed of tea buds can be measured. However, the mixed spectrum of the tea pile is affected by the stacking structure of the tea buds in the tea pile, resulting in the difficulty of the mixed spectrum of the tea pile representing the spectrum of the leaves.

[0004] It can be seen that the tea bud quality detection method in the related art has the technical problem that it is difficult to measure the quality of the tea bud leaves in the tea pile. Summary of the Invention

[0005] The present invention provides a method, device and electronic device for processing spectral data of a tea pile to solve the defect that the tea bud quality detection method in the prior art is difficult to measure the quality of the tea bud leaves in the tea pile, and to eliminate the influence of the tea bud stacking structure and the difference in the proportion of front leaves in the mixed spectrum of the tea pile on the accurate inversion of the tea bud quality.

[0006] The present invention provides a method for processing spectral data of a tea pile, including the following steps.

[0007] Obtain the mixed spectrum of the tea pile. The mixed spectrum of the tea pile is obtained by collecting spectral data of the tea pile based on the measurement wavelength. The tea pile is composed of randomly stacked tea bud leaves, and the tea bud leaves include: front leaves and back leaves. Input the mixed spectrum of the tea pile into the bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model. The bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model using a radial basis kernel function. Specifically, it includes: inputting the mixed spectrum of the tea pile into the bilinear spectral model to obtain the front leaf spectrum of the tea pile output by the bilinear spectral model, where the bilinear spectral model is used to correct the mixed spectrum of the tea pile; taking the derivative of the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, where the derivative is used to eliminate the tea bud stacking structure information of the tea pile; inputting the derivative set into the Gaussian process regression model using the radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using the radial basis kernel function. The radial basis kernel function in the Gaussian process regression model is used to eliminate the front leaf proportion information of the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality.

[0008] According to a method for processing spectral data of a tea pile provided by the present invention, before inputting the mixed spectrum of the tea pile into the bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model, the method further includes: obtaining a front leaf spectrum sample of a front tea bud leaf sample and a back leaf spectrum sample of a back tea bud leaf sample; determining a first linear relationship between the front leaf spectrum sample and the back leaf spectrum sample based on the front leaf spectrum sample and the back leaf spectrum sample; obtaining a mixed spectrum sample of a tea pile sample, the tea bud stacking structure information of the tea pile sample, and the front leaf proportion information of the tea pile sample, where the tea pile sample is stacked by the front tea bud leaf sample and the back tea bud leaf sample; determining a second linear relationship between the front leaf spectrum sample, the back leaf spectrum sample, and the mixed spectrum sample of the tea pile based on the tea bud stacking structure information and the front leaf proportion information.

[0009] According to a method for processing spectral data of a tea pile provided by the present invention, the method further includes: constructing a bilinear spectral model based on the first linear relationship and the second linear relationship; where the bilinear spectral model refers to the following formula:

[0010]

[0011] Where Represents the front leaf spectrum, Represents the proportion information of the front leaves of the tea pile, Represents the first constant, Represents the mixed spectrum of the tea pile, Represents the second constant, Represents the tea bud stacking structure information of the tea pile.

[0012] According to a method for processing spectral data of a tea pile provided by the present invention, taking the derivative of the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum includes: taking the derivative of the measurement wavelength based on the front leaf spectrum according to the bilinear spectral model to obtain a derivative set of the front leaf spectrum, referring to the following formula:

[0013] ;

[0014] Wherein, Represents the derivative set of the front leaf spectrum, Represents the front leaf proportion diagonal matrix, Represents the measurement wavelength, Represents the derivative set of the mixed spectrum of the tea pile, Represents the front leaf spectrum, Represents the proportion information of the front leaves of the tea pile, Represents the first constant, Represents the mixed spectrum of the tea pile, Represents the tea bud stacking structure information of the tea pile.

[0015] According to a method for processing spectral data of a tea pile provided by the present invention, before inputting the derivative set into the Gaussian process regression model using the radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using the radial basis kernel function, the method further includes:

[0016] Obtaining a verification spectrum and quality data set and a training spectrum and quality data set, wherein the verification spectrum and quality data set include a derivative set of the front leaf spectrum of the verification tea pile and a quality data set of the verification tea pile; the training spectrum and quality data set include a derivative set of the front leaf spectrum of the training tea pile and a quality data set of the training tea pile;

[0017] Based on the derivative set of the front leaf spectrum of the verification tea pile and the derivative set of the front leaf spectrum of the training tea pile, determining a similarity covariance function between the derivative set of the front leaf spectrum of the verification tea pile and the derivative set of the front leaf spectrum of the training tea pile, referring to the following formula:

[0018]

[0019] Among them, represents the similarity covariance function, represents the set of derivatives of the front leaf spectrum of the verification tea pile, represents the set of derivatives of the front leaf spectrum of the training tea pile, represents the scaling parameter, represents the front leaf proportion diagonal matrix, represents the set of derivatives of the mixed spectrum of the verification tea pile, represents the set of derivatives of the mixed spectrum of the training tea pile, represents the length scale parameter of the kernel function; represents the noise standard deviation, represents the Kronecker symbol.

[0020] A method for processing spectral data of a tea pile provided by the present invention, the method further includes: determining the mean function of the set of derivatives of the front leaf spectrum; constructing a Gaussian process regression model based on the mean function and the similarity covariance function.

[0021] The present invention also provides a spectral data processing device for a tea pile, including the following modules:

[0022] A spectral acquisition module, configured to acquire the mixed spectrum of the tea pile of the tea pile, wherein the mixed spectrum of the tea pile is obtained by collecting spectral data of the tea pile based on the measurement wavelength, the tea pile is composed of randomly stacked tea bud leaves, and the tea bud leaves include: front leaves and back leaves; a spectral processing module, configured to input the mixed spectrum of the tea pile into a bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model, and the bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model using a radial basis kernel function, wherein, including: inputting the mixed spectrum of the tea pile into the bilinear spectral model to obtain the front leaf spectrum of the tea pile output by the bilinear spectral model, wherein the bilinear spectral model is used to correct the mixed spectrum of the tea pile; differentiating the front leaf spectrum of the tea pile to obtain the set of derivatives of the front leaf spectrum, wherein the differentiation is used to eliminate the tea bud stacking structure information of the tea pile; inputting the set of derivatives into the Gaussian process regression model using the radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using the radial basis kernel function, wherein the radial basis kernel function in the Gaussian process regression model is used to eliminate the front leaf proportion information of the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality.

[0023] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the spectral data processing method of the tea pile as described in any one of the above is implemented.

[0024] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the spectral data processing method of the tea pile as described in any one of the above is implemented.

[0025] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the spectral data processing method of the tea pile as described in any one of the above is implemented.

[0026] The spectral data processing method, device, and electronic device of the tea pile provided by the present invention collect spectral data of a tea pile composed of randomly stacked tea bud leaves (including front leaves and back leaves) through a measurement wavelength, and can obtain a mixed spectrum containing the overall spectral information of the tea pile. The mixed spectrum of the tea pile is input into a bilinear spectral model, and the bilinear spectral model can convert the mixed spectrum of the tea pile into a front leaf spectrum through the correction of the mixed spectrum. By taking the derivative of the front leaf spectrum, a derivative set of the front leaf spectrum is obtained, eliminating the influence of the stacking structure information of the tea buds in the tea pile on the spectral data. The derivative set is input into a Gaussian process regression model, which can further eliminate the influence of the proportion information of the front leaves in the tea pile on the spectral data. Through the prediction ability of Gaussian process regression, a more accurate target front leaf quality can be obtained, thereby solving the technical problem that it is difficult to measure the quality of the tea bud leaves in the tea pile in the tea bud quality detection method in the related art. Description of the Drawings

[0027] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art one by one. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0028] Figure 1 It is a comparison schematic diagram of tea buds of small-leaf tea trees and large-leaf tea trees provided by the present invention.

[0029] Figure 2 It is a schematic diagram of the influencing factors of the mixed spectrum of the tea pile provided by the present invention.

[0030] Figure 3 It is a flow schematic diagram of the spectral data processing method of the tea pile provided by the present invention.

[0031] Figure 4It is a schematic diagram for evaluating the accuracy of the bilinear spectral derivative Gaussian process regression model provided by the present invention.

[0032] Figure 5 It is a schematic structural diagram of the spectral data processing device for tea piles provided by the present invention.

[0033] Figure 6 It is a schematic entity structure diagram of the electronic device provided by the present invention. Detailed implementation manners

[0034] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0035] As one of the most popular beverages, green tea is made from fresh tea buds with one bud and one leaf picked from tea trees. The quality of tea leaves depends on the contents of tea polyphenols, amino acids and soluble sugars. Therefore, the quantitative evaluation of quality parameters in tea buds is a key factor for guiding tea bud picking and producing high-quality green tea. The traditional chemical methods for detecting tea bud quality not only take a long time but also are highly destructive. In recent years, with the rapid development of hyperspectral technology, narrow bands can capture subtle changes in absorption characteristics and achieve accurate estimation of the contents of vegetation physiological and biochemical components. However, the establishment of these estimation models relies on the measurement of leaf spectra.

[0036] Reference Figure 1 , Figure 1 It is a comparative schematic diagram of tea buds of small-leaf tea trees and large-leaf tea trees provided by the present invention, including: (a) large-leaf tea trees and (b) small-leaf tea trees.

[0037] As Figure 1 shown, the tea buds of small-leaf tea trees are very small and the leaf margins are curled. The spectral information of a single tea bud is weak and it is difficult for existing non-imaging spectrometers to measure the spectrum of a single tea bud leaf. Therefore, only the mixed spectrum of the tea pile composed of tea buds can be measured.

[0038] Reference Figure 2 , Figure 2 It is a schematic diagram of influencing factors of the tea pile mixed spectrum provided by the present invention, including: (a) stacking structure and (b) field of view ratio.

[0039] As Figure 2As shown, the accuracy of inverting the contents of tea polyphenols, amino acids, and soluble sugars in tea buds based on the mixed spectrum of tea piles is affected by the stacking structure of tea buds in the tea pile and the proportion of the front and back sides of tea buds in the field of view. The Linear Spectral Mixing Model (LSMM) is a model widely used in remote sensing data analysis. This model assumes that the reflection spectrum of the entire canopy is a linear combination of the spectra of vegetation and soil in the field of view, and significant progress has been made in various aspects such as land cover classification, vegetation cover inversion, water quality, mineral content, and aerosol monitoring. In addition, machine learning algorithms have achieved remarkable success in estimating the physiological and biochemical parameters of vegetation using hyperspectral data. In particular, machine learning algorithms with kernel functions can map data into a high-dimensional space, thus transforming the problem that is non-linearly separable in the input space into a linearly separable form. This enables it to effectively process data with complex and non-linear relationships.

[0040] In summary, the development of LSMM and machine learning methods provides the possibility of eliminating the influence of the stacking structure and the proportion of the front and back sides of tea buds in the field of view in the mixed spectrum of tea piles on the accurate inversion of tea bud quality.

[0041] In related technologies, the tea buds of small-leaf tea trees are very small and have curled leaf margins. The average leaf length is 2.7 cm, and the average leaf width is 1.3 cm. The spectral information of a single tea bud is weak, and it is difficult for existing non-imaging spectrometers to measure the spectrum of a single tea bud leaf. The mixed spectrum of the tea pile is affected by the stacking structure of tea buds in the tea pile. Especially in the near-infrared region where the spectral transmittance is high, it is more susceptible to the influence of the stacking structure (such as stacking thickness), resulting in the mixed spectrum of the tea pile being difficult to represent the spectrum of the leaves. The mixed spectrum of the tea pile is affected by the proportion of the front and back leaves in the field of view. Especially in the visible light region, since the back of the tea bud is covered with white fluff, the reflectance of the back of the tea bud is higher than that of the front.

[0042] According to a spectral data processing method for tea piles provided by the present invention, a Bilinear Spectral Derivative Gaussian Process Regression Model (BSDGPR) is constructed by combining the Linear Spectral Mixing Model (LSMM) and the Gaussian Process Regression (GPR) model, eliminating the influence of the stacking structure information in the mixed spectrum of tea piles and the proportion of the front and back sides of tea buds in the field of view, correcting the mixed spectrum of the tea pile into the leaf spectrum, and realizing rapid, non-destructive, and high-precision monitoring of tea bud quality in the field.

[0043] The method of the present invention directly and rapidly detects the quality of fresh tea buds through a bilinear spectral derivative Gaussian process regression model for the spectra of fresh tea buds directly picked in the field, which helps to determine the most suitable time for picking high-quality tea buds in the field so as to pick them in a timely manner.

[0044] The present invention is different from the methods in the related art: in the related art, after the tea buds are picked, there are multiple processes in the tea processing factory to process the fresh tea buds, and it is necessary to monitor the impact of each process on the quality of the tea buds, and finally determine the processes for producing high-quality tea products (such as precise control of fermentation time, temperature, etc.).

[0045] Coupling the linear spectral mixture model and the first-order and second-order derivative methods eliminates the influence of the stacking structure of tea buds in the tea pile on the mixed spectrum of the tea pile. By scaling the length scale parameter in the squared exponential covariance kernel function of the GPR model, the influence of the proportion of the front and back sides of the tea buds in the field of view on the mixed spectrum of the tea pile is eliminated. The mixed spectrum of the tea pile is corrected to the target front leaf quality (i.e., the spectrum of the tea buds), improving the accuracy of rapid non-destructive monitoring of fresh tea buds in the field.

[0046] Optionally, the method for processing the spectral data of the tea pile in the embodiments of the present application can be executed by a server, or by a terminal device, or jointly executed by the server and the terminal device. Taking the execution of the method for processing the spectral data of the tea pile in this embodiment by the terminal device as an example.

[0047] Figure 3 is a schematic flowchart of the method for processing the spectral data of the tea pile provided by the present invention, as Figure 3 shown, the method includes the following steps.

[0048] Step 301, obtain the mixed spectrum of the tea pile.

[0049] Among them, the mixed spectrum of the tea pile is obtained by collecting spectral data of the tea pile based on the measurement wavelength. The tea pile is composed of randomly stacked tea bud leaves, and the tea bud leaves include: front leaves and back leaves.

[0050] In the embodiments of the present invention, during the annual tea bud picking period (usually from March to April), pick one-bud-one-leaf tea buds in the field, randomly stack the collected tea buds on a pure black non-reflective cloth block (the diameter of the tea pile is about 5 cm and the thickness is about 2 cm), and use an ASD handheld spectrometer corrected with a whiteboard at the target frequency of the measurement wavelength to record the mixed spectrum of the tea pile 11-12 cm above the tea pile within the target time period (for example, 30 minutes).

[0051] In some embodiments, it is also necessary to conduct quality tests on tea buds in the tea pile. For the quality tests of tea buds, each tea bud sample needs to be first blanched after measuring the spectral data and then dried in an oven at 80 °C until a constant weight is reached. After weighing, it is ground into powder using an electric grinder for quality tests. The content of tea polyphenols is determined by the ferrous tartrate colorimetry method and spectrophotometry. The content of free amino acids is determined by the ninhydrin colorimetry method at 570 nm. The content of soluble sugars is determined by the anthrone reagent method. Each sample is tested 3 times repeatedly, and the average value is taken as the final value.

[0052] Step 302: Input the mixed spectrum of the tea pile into the bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model.

[0053] Among them, the bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model using a radial basis kernel function.

[0054] In the embodiments of the present invention, a bilinear spectral derivative Gaussian process regression model is constructed by combining a linear spectral mixture model and a Gaussian process regression model. The bilinear spectral derivative Gaussian process regression model is used to eliminate the influence of the stacking structure information in the mixed spectrum of the tea pile and the proportion of the front and back of the tea buds in the field of view, and correct the mixed spectrum of the tea pile into the target front leaf spectrum, realizing fast, non-destructive and high-precision monitoring of the quality of tea buds in the field.

[0055] The above step 302 specifically includes the following steps.

[0056] Step 3021: Input the mixed spectrum of the tea pile into the bilinear spectral model to obtain the front leaf spectrum of the tea pile output by the bilinear spectral model, where the bilinear spectral model is used to correct the mixed spectrum of the tea pile.

[0057] Since the physiological and biochemical parameter contents of the same tea bud are the same and the leaf cell structures are similar, there is a linear relationship between the front leaf spectrum and the back leaf spectrum. Therefore, through linear relationship conversion, the mixed spectrum of the tea pile can be converted into the front leaf spectrum of the tea pile.

[0058] According to a method for processing spectral data of a tea pile provided by the present invention, before inputting the mixed spectrum of the tea pile into the bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model, the above method further includes:

[0059] Obtain the front leaf spectrum sample of the front tea bud leaf sample and the back leaf spectrum sample of the back tea bud leaf sample;

[0060] Based on the spectral samples of the front leaves and the spectral samples of the back leaves, determine the first linear relationship between the spectral samples of the front leaves and the spectral samples of the back leaves;

[0061] Obtain the tea pile mixed spectral sample of the tea pile sample, the tea bud stacking structure information of the tea pile sample, and the proportion information of the front leaves in the tea pile sample, where the tea pile sample is obtained by stacking the front tea bud leaf samples and the back tea bud leaf samples;

[0062] Based on the tea bud stacking structure information and the proportion information of the front leaves, determine the second linear relationship among the spectral samples of the front leaves, the spectral samples of the back leaves, and the tea pile mixed spectral sample.

[0063] In some embodiments, from the front tea bud leaf samples, use a spectrometer or other relevant equipment to measure and record their spectral data. These data will constitute the spectral samples of the front leaves of the front leaves. Similarly, from the back tea bud leaf samples, use the same equipment or method to measure and record their spectral data. These data will constitute the spectral samples of the back leaves of the back leaves.

[0064] Use statistical methods (such as linear regression analysis) to analyze the linear relationship between the spectral samples of the front leaves and the spectral samples of the back leaves as the first linear relationship.

[0065] From the tea pile sample obtained by stacking the front tea bud leaf samples and the back tea bud leaf samples, use a spectrometer to measure and record its mixed spectral data. These data will constitute the tea pile mixed spectral sample.

[0066] Through physical observation or image analysis of the tea pile sample, determine the structural characteristics such as the arrangement pattern and stacking density of the tea buds as the tea bud stacking structure information. Through sampling analysis of the tea pile sample, or using image recognition technology to estimate the proportion of the front leaves and the back leaves in the tea pile as the proportion information of the front leaves.

[0067] It should be noted that the tea bud stacking structure information and the proportion information of the front leaves can also be used as preset unknown variables and eliminated through subsequent mathematical operation processes. In the actual application process, the tea bud stacking structure information and the proportion information of the front leaves in the tea pile are unknown variables.

[0068] Based on the tea bud stacking structure information and the proportion information of the front leaves, determine the second linear relationship among the spectral samples of the front leaves, the spectral samples of the back leaves, and the tea pile mixed spectral sample.

[0069] In the embodiments of the present invention, the bilinear spectral mixture model assumes that the reflected spectrum of the entire canopy is a linear combination of the vegetation and soil within the field of view. The tea pile is randomly arranged by the front-facing leaves and the back-facing leaves of fresh tea buds. Therefore, the tea pile mixed spectrum ( ) can be represented by the front leaf spectrum ( ) and the back leaf spectrum ( ), and the formula is as shown in (1).

[0070] (1)

[0071] In the formula, represents the tea pile mixed spectrum (sample), , respectively represent the front leaf spectrum (sample) and the back leaf spectrum (sample), represents the proportion information of the front leaves in the sensor's field of view (the proportion of the front of the tea bud in the total area of the tea pile, as shown in (b) of Figure 2 ), so, , represents the stacking structure information of the tea buds in the tea pile (as shown in (a) of Figure 2 ), represents noise or random error.

[0072] Since the physiological and biochemical parameter contents of the same kind of tea buds are the same and the leaf cell structures are similar, there is a linear relationship between the spectra of the front and back of the tea buds, and the formula is as shown in (2).

[0073] (2)

[0074] Among them, , respectively represent the front leaf spectrum (sample) and the back leaf spectrum (sample), and m and n are the slope and intercept of the linear model respectively.

[0075] Through the embodiments of the present invention, by using statistical methods to fit multiple linear equations, the relationship between the tea pile mixed spectrum sample and the front leaf spectrum sample and the back leaf spectrum sample can be described, so as to facilitate the subsequent correction of the tea pile mixed spectrum to the front leaf spectrum.

[0076] According to a method for processing spectral data of a tea pile provided by the present invention, the above method further includes:

[0077] Constructing a bilinear spectral model based on the first linear relationship and the second linear relationship;

[0078] Among them, the bilinear spectral model refers to the following formula:

[0079]

[0080] Among them, represents the front leaf spectrum, represents the proportion information of the front leaves of the tea pile, represents the first constant, Represents the mixed spectrum of the tea pile, Represents the second constant, Represents the information on the stacking structure of tea buds in the tea pile.

[0081] In the embodiment of the present invention, the bilinear spectral model is constructed using the above formula (2) and formula (1). The bilinear spectral model is used to correct the mixed spectrum of the tea pile to the front leaf spectrum , and the calculation formula is as shown in (3).

[0082] (3)

[0083] It can be seen from formula (3) that the front leaf spectrum is affected by the tea bud stacking structure information (SI) and the front leaf proportion information (f).

[0084] Through the embodiment of the present invention, the mixed spectrum of the tea pile can be corrected to the front leaf spectrum by the bilinear spectral model.

[0085] Step 3022: Take the derivative of the front leaf spectrum of the tea pile to obtain a set of derivatives of the front leaf spectrum, where taking the derivative is used to eliminate the tea bud stacking structure information of the tea pile.

[0086] According to a method for processing spectral data of a tea pile provided by the present invention, taking the derivative of the front leaf spectrum of the tea pile to obtain a set of derivatives of the front leaf spectrum includes:[[]]

[0087] According to the bilinear spectral model, take the derivative of the measurement wavelength based on the front leaf spectrum to obtain a set of derivatives of the front leaf spectrum, referring to the following formula:[[]]

[0088]

[0089] Wherein, Represents a set of derivatives of the front leaf spectrum, Represents the front leaf proportion diagonal matrix, Represents the measurement wavelength, Represents a set of derivatives of the mixed spectrum of the tea pile, Represents the front leaf spectrum, Represents the front leaf proportion information of the tea pile, Represents the first constant, Represents the mixed spectrum of the tea pile, Represents the tea bud stacking structure information of the tea pile.

[0090] Since the front leaf proportion information in each tea pile is independent of the measurement wavelength, it can be represented by formula (4) after taking the derivative with respect to the wavelength .

[0091] (4)

[0092] Wherein, a is the first constant, which does not vary with the measurement wavelength Since most of the light energy in the visible light region is absorbed and the transmittance is very small, it is hardly affected by the structure. In the near-infrared region, although it is affected by the structure, due to the consistent transmittance, the change rate of SI with the measurement wavelength is very small.

[0093] Therefore, in the present invention, by calculating the first derivative (First Derivative, FD) and the second derivative (Second Derivative, SD), and are very small and can be ignored, thereby eliminating the influence of the tea bud stacking structure information of the tea pile ( ) on the front leaf spectrum. After simplifying formula (4), it is expressed as formula (5):

[0094] (5)

[0095] In the formula, represents the set composed of the derivatives of the front leaf spectrum; represents the set composed of the derivatives of the tea pile mixed spectrum, both including the first derivative ( ) and the second derivative ( ).

[0096] Wherein, , , m = 1, 2,..., n, represents the derivative of the m-th front leaf spectrum, represents the derivative of the m-th tea pile mixed spectrum, represents the set of real numbers, represents the dimension, represents the total number of derivatives, a is the first constant, represents the front leaf proportion parameter (used to adjust the weight), represents the front leaf proportion parameter of the m-th tea pile, is regulated by the front leaf proportion information , represents the front leaf proportion information in the m-th tea pile.

[0097] Through the embodiments of the present invention, the front leaf spectrum is affected by the tea bud stacking structure information and the front leaf proportion information, and the front leaf proportion information of each tea pile is independent of the measurement wavelength. Therefore, by taking the derivative of the front leaf spectrum with respect to the measurement wavelength and calculating the first derivative and the second derivative, the influence of the tea bud stacking structure information on the front leaf spectrum is eliminated.

[0098] Step 3023: Input the derivative set into a Gaussian process regression model using a radial basis kernel function to obtain the target positive leaf quality output by the Gaussian process regression model using the radial basis kernel function. Among them, the radial basis kernel function in the Gaussian process regression model is used to eliminate the information on the proportion of positive leaves in the tea pile, and the Gaussian process regression model is used to predict the target positive leaf quality.

[0099] The target positive leaf quality is used to subsequently evaluate the tea bud quality of the tea pile (tea tree), where multiple tea bud leaves in the tea pile belong to the same tea tree.

[0100] According to a method for processing spectral data of a tea pile provided by the present invention, the above method further includes:

[0101] Determine the mean function of the derivative set of the positive leaf spectrum;

[0102] Construct a Gaussian process regression model based on the mean function and the similarity covariance function.

[0103] Gaussian Process Regression (GPR) is a Bayesian method for solving general regression problems using kernels, which can map data into a high-dimensional space, thus transforming a problem that is non-linearly separable in the input space into a linearly separable form. Specifically, GPR establishes the form represented by Formula 6-8 between the input data and the output variable ( represents the i-th input data, represents the i-th output variable), and the formula is as follows:

[0104] (6)

[0105] where, represents that the function obeys the Gaussian process , the mean function describes the expected value of the function at any point ; the covariance function (kernel function) describes the correlation between the function at any two points .

[0106] In a Gaussian process, the mean function is usually known and can be any form of function, such as a linear function, a polynomial function, etc. The covariance function determines how the values of the function at different points depend on each other. Common kernel functions include the squared exponential kernel (RBF kernel), the polynomial kernel, the Matern kernel, etc.

[0107] The present invention constructs a bilinear spectral derivative Gaussian process regression model (BSDGPR) based on the bilinear spectral model and the GPR model:

[0108] (7)

[0109] wherein, represents the derivative set of the front leaf spectrum, , respectively represent the derivative set of the front leaf spectrum of the verification tea pile and the derivative set of the front leaf spectrum of the training tea pile.

[0110] Through the embodiments of the present invention, the Gaussian process can flexibly model various complex functional relationships through the mean function and the covariance function.

[0111] According to a method for processing spectral data of a tea pile provided by the present invention, before inputting the derivative set into the Gaussian process regression model using the radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using the radial basis kernel function, the above method further includes:

[0112] Obtain the verification spectral and quality data set and the training spectral and quality data set, wherein the verification spectral and quality data set includes the derivative set of the front leaf spectrum of the verification tea pile and the quality data set of the verification tea pile; the training spectral and quality data set includes the derivative set of the front leaf spectrum of the training tea pile and the quality data set of the training tea pile;

[0113] Based on the derivative set of the front leaf spectrum of the verification tea pile and the derivative set of the front leaf spectrum of the training tea pile, determine the similarity covariance function between the derivative set of the front leaf spectrum of the verification tea pile and the derivative set of the front leaf spectrum of the training tea pile, referring to the following formula:

[0114]

[0115] wherein, represents the similarity covariance function, represents the derivative set of the front leaf spectrum of the verification tea pile, represents the derivative set of the front leaf spectrum of the training tea pile, represents the scaling parameter, represents the front leaf occupancy diagonal matrix, represents the verification tea pile mixed spectral derivative set, represents the training tea pile mixed spectral derivative set, represents the length scale parameter of the kernel function; represents the noise standard deviation, represents the Kronecker symbol.

[0116] The present invention adopts a scaled squared exponential covariance function, and the formula is shown in (8).

[0117] (8)

[0118] In the formula, is the scaling parameter, , is the length scale parameter of the kernel function, which is the way of propagation of training information along the input dimension, is the noise standard deviation, is the Kronecker symbol.

[0119] In the present invention, the diagonal matrix is a positive definite matrix, is equivalent to linearly scaling , without changing its relative position. Therefore, under the adjustment of the length scale parameter of the covariance function, will not be affected by the proportion of the positive leaves. Therefore, the influence of the proportion of the positive leaves on the model accuracy is eliminated, and finally the mixed spectrum of the tea pile is converted into the spectrum of fresh positive leaves.

[0120] and are respectively the set composed of the derivatives of the positive leaf spectrum and the set composed of the derivatives of the tea pile mixed spectrum.

[0121] In some embodiments, the GPR model is trained: using the training set of the derivatives of the tea pile mixed spectrum and the covariance function defined above, the Gaussian process regression model is trained. During the training process, the model will learn the covariance function parameters most suitable for the data (such as , , ).

[0122] Model prediction: Using the trained GPR model ( ) and the validation set of the derivatives of the tea pile mixed spectrum for prediction. The model will output the predicted value of the quality of the validated tea pile, thereby providing an accurate prediction of the tea pile quality.

[0123] Reference Figure 4 , Figure 4 is a schematic diagram for evaluating the accuracy of the bilinear spectral derivative Gaussian process regression model provided by the present invention.

[0124] As Figure 4As shown, the precision of predicting the contents of tea polyphenols, amino acids, and soluble sugars using the bilinear spectral derivative Gaussian process regression model (BSDGPR) for the mixed spectrum of the tea pile is considered. The bilinear spectral derivative Gaussian process regression model includes the bilinear spectral derivative Gaussian process regression model using the first derivative (FD) (BSDGPR-FD) and the bilinear spectral derivative Gaussian process regression model using the second derivative (SD) (BSDGPR-SD).

[0125] Figure 4 In the figure, the abscissa represents the measured values of the quality parameters obtained by chemical analysis, and the ordinate represents the estimated values of the quality parameters of the target front leaf by the BSDGPR model. The quality parameters shown refer to the contents of tea polyphenols, amino acids, and soluble sugars. Figure 4 In Figure (a), it shows the prediction precision of the tea pile mixed spectrum input into the BSDGPR-FD model for the content of tea polyphenols. The coefficient of determination (R2) is 0.81, and the root mean square error (RMSE) is 1.61; Figure 4 In Figure (b), it shows the prediction precision of the tea pile mixed spectrum input into the BSDGPR-SD model for the content of tea polyphenols. The R2 is 0.75, and the RMSE is 1.90; Figure 4 In Figure (c), it shows the prediction precision of the tea pile mixed spectrum input into the BSDGPR-FD model for the content of amino acids. The R2 is 0.70, and the RMSE is 0.30; Figure 4 In Figure (d), it shows the prediction precision of the tea pile mixed spectrum input into the BSDGPR-SD model for the content of amino acids. The R2 is 0.66, and the RMSE is 0.36; Figure 4 In Figure (e), it shows the prediction precision of the tea pile mixed spectrum input into the BSDGPR-FD model for the content of soluble sugars. The R2 is 0.91, and the RMSE is 0.54; Figure 4 In Figure (f), it shows the prediction precision of the tea pile mixed spectrum input into the BSDGPR-SD model for the content of soluble sugars. The R2 is 0.90, and the RMSE is 0.57.

[0126] In summary, the present invention proposes a model (BSDGPR) for rapid non-destructive measurement of the quality of fresh leaves of small-leaf tea trees in the field. First, based on the bilinear spectral model, the mixed spectrum of the tea pile is converted into the front leaf spectrum, and the information of the tea bud stacking structure and the proportion of the front leaf are eliminated respectively through the methods of derivative and GPR, realizing rapid, non-destructive, and accurate monitoring of the quality of tea buds in the field.

[0127] Through the above embodiments of the present invention, by coupling the linear spectral mixture model and the Gaussian process regression model, the mixed spectrum of the randomly stacked fresh tea buds of small-leaf tea trees is corrected to the front leaf spectrum, eliminating the influence of the stacking structure and the proportion of the front and back of the tea buds in the field of view, and having the following advantages.

[0128] The influence of the stacking structure is eliminated: The influence of the stacking structure on the mixed spectrum of the tea pile caused by the high spectral transmittance and low absorption rate in the near-infrared region under the stacked state of tea buds is solved. The influence of the proportion of the front and back tea buds in the field of view is eliminated: The influence of the high reflection of the fluff on the back of the tea buds in the visible light region on the mixed spectrum of the tea pile is eliminated through the kernel function. Rapid non-destructive monitoring in the field: The fusion of the linear spectral mixture model and the Gaussian process regression model corrects the spectrum of the tea pile to the spectrum of tea buds of equal size, realizing the rapid non-destructive monitoring of the quality of fresh tea buds in the field.

[0129] Through the bilinear spectral derivative Gaussian process regression model of the embodiment of the present invention, the difficulty of obtaining the spectrum of a single tea bud by the existing non-imaging spectrometer due to the too small tea buds of the small-leaf tea tree is overcome, and the problem that the mixed spectrum of the tea pile is interfered by the stacking structure and the proportion of the front and back of the tea leaves in the field of view is solved, providing a new method for the rapid non-destructive and accurate monitoring of small-leaf tea trees in the field.

[0130] The spectral data processing device of the tea pile provided by the present invention will be described below. The spectral data processing device of the tea pile described below can be mutually corresponding and referred to the spectral data processing method of the tea pile described above.

[0131] Reference Figure 5 , Figure 5 is a schematic structural diagram of the spectral data processing device of the tea pile provided by the present invention.

[0132] The spectral acquisition module 501 is used to acquire the mixed spectrum of the tea pile of the tea pile. Among them, the mixed spectrum of the tea pile is obtained by collecting spectral data of the tea pile based on the measurement wavelength. The tea pile is composed of randomly stacked tea bud leaves, and the tea bud leaves include: front leaves and back leaves;

[0133] The spectral processing module 502 is used to input the mixed spectrum of the tea pile into the bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model. The bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model using a radial basis kernel function, among which, including:

[0134] Input the mixed spectrum of the tea pile into the bilinear spectral model to obtain the front leaf spectrum of the tea pile output by the bilinear spectral model, where the bilinear spectral model is used to correct the mixed spectrum of the tea pile;

[0135] Derive the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, where the derivation is used to eliminate the tea bud stacking structure information of the tea pile;

[0136] Input the derivative set into the Gaussian process regression model with a radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model with a radial basis kernel function. Among them, the radial basis kernel function in the Gaussian process regression model is used to eliminate the information of the proportion of front leaves in the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality.

[0137] Specifically, the above-mentioned spectral data processing device for the tea pile provided by the present invention can implement all the method steps implemented by the above-mentioned spectral data processing method embodiment of the tea pile, and can achieve the same technical effects. Here, the same parts and beneficial effects as those in the method embodiment will not be specifically described again.

[0138] Figure 6 It is a schematic physical structure diagram of the electronic device provided by the present invention. As Figure 6 shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640. Among them, the processor 610, the communication interface 620, and the memory 630 complete mutual communication through the communication bus 640. The processor 610 can call the logical instructions in the memory 630 to execute the spectral data processing method for the tea pile. The method includes: obtaining the mixed spectral of the tea pile, where the mixed spectral of the tea pile is obtained by collecting spectral data of the tea pile based on the measurement wavelength. The tea pile is composed of randomly stacked tea bud leaves, and the tea bud leaves include: front leaves and back leaves; input the mixed spectral of the tea pile into the bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model. The bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model with a radial basis kernel function. Among them, it includes: inputting the mixed spectral of the tea pile into the bilinear spectral model to obtain the front leaf spectrum of the tea pile output by the bilinear spectral model, where the bilinear spectral model is used to correct the mixed spectral of the tea pile; taking the derivative of the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, where taking the derivative is used to eliminate the tea bud stacking structure information of the tea pile; inputting the derivative set into the Gaussian process regression model with a radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model with a radial basis kernel function. Among them, the radial basis kernel function in the Gaussian process regression model is used to eliminate the information of the proportion of front leaves in the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality.

[0139] In addition, when the logical instructions in the above-mentioned memory 630 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0140] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the spectral data processing method of the tea pile provided by the above-mentioned various methods. The method includes: obtaining the mixed spectrum of the tea pile, where the mixed spectrum of the tea pile is obtained by collecting spectral data of the tea pile based on the measurement wavelength. The tea pile is composed of randomly stacked tea bud leaves, and the tea bud leaves include: front leaves and back leaves; inputting the mixed spectrum of the tea pile into a bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model. The bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model using a radial basis kernel function, where: inputting the mixed spectrum of the tea pile into the bilinear spectral model to obtain the front leaf spectrum of the tea pile output by the bilinear spectral model, where the bilinear spectral model is used to correct the mixed spectrum of the tea pile; taking the derivative of the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, where taking the derivative is used to eliminate the tea bud stacking structure information of the tea pile; inputting the derivative set into the Gaussian process regression model using a radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using a radial basis kernel function, where the radial basis kernel function in the Gaussian process regression model is used to eliminate the proportion information of the front leaves in the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality.

[0141] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements a spectral data processing method for a tea pile provided by the above-mentioned various methods. The method includes: obtaining a mixed spectrum of the tea pile, where the mixed spectrum of the tea pile is obtained by collecting spectral data of the tea pile based on the measurement wavelength, and the tea pile is composed of randomly stacked tea bud leaves, and the tea bud leaves include: front leaves and back leaves; inputting the mixed spectrum of the tea pile into a bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model. The bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model using a radial basis kernel function. Among them, it includes: inputting the mixed spectrum of the tea pile into the bilinear spectral model to obtain the front leaf spectrum of the tea pile output by the bilinear spectral model, where the bilinear spectral model is used to correct the mixed spectrum of the tea pile; taking the derivative of the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, where taking the derivative is used to eliminate the tea bud stacking structure information of the tea pile; inputting the derivative set into the Gaussian process regression model using a radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using a radial basis kernel function, where the radial basis kernel function in the Gaussian process regression model is used to eliminate the proportion information of the front leaves in the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality.

[0142] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.

[0143] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, also by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing spectral data of tea piles, characterized in that: include: Acquire a tea pile mixed spectrum of a tea pile, wherein the tea pile mixed spectrum is obtained by collecting spectral data of the tea pile based on a measurement wavelength, the tea pile is composed of randomly stacked tea buds and leaves, and the tea buds and leaves include: front leaves and back leaves; The mixed spectrum of the tea pile is input into a bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model, wherein the bilinear spectral derivative Gaussian process regression model includes: a bilinear spectral model and a Gaussian process regression model using a radial basis kernel function, wherein: Inputting the mixed spectrum of the tea pile into the bilinear spectrum model to obtain the spectrum of the front leaves of the tea pile output by the bilinear spectrum model, wherein the bilinear spectrum model is used to correct the mixed spectrum of the tea pile; Derivative the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, wherein the derivation is used to eliminate the tea bud stacking structure information of the tea pile; Inputting the derivative set into the Gaussian process regression model using the radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using the radial basis kernel function, wherein the radial basis kernel function in the Gaussian process regression model is used to eliminate the front leaf proportion information of the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality; The bilinear spectral model refers to the following formula: ； ； in, represents the front leaf spectrum, Indicates the proportion of the front leaves of the tea pile, represents the first constant, represents the tea pile mixed spectrum, represents the second constant, Indicates the tea bud stacking structure information of the tea pile, represents the spectrum of the back leaf, Represents noise or random error.

2. The method for processing spectral data of tea pile according to claim 1, characterized in that: Before inputting the mixed spectrum of the tea pile into the bilinear spectral derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectral derivative Gaussian process regression model, the method further includes: Obtaining a front leaf spectrum sample of a front tea bud leaf sample and a back leaf spectrum sample of a back tea bud leaf sample; Based on the front blade spectrum sample and the back blade spectrum sample, determining a first linear relationship between the front blade spectrum sample and the back blade spectrum sample; Obtaining a tea pile mixed spectrum sample of a tea pile sample, tea bud stacking structure information of the tea pile sample, and front leaf ratio information of the tea pile sample, wherein the tea pile sample is obtained by stacking the front tea bud leaf sample and the back tea bud leaf sample; Based on the tea bud stacking structure information and the front leaf proportion information, a second linear relationship between the front leaf spectrum sample, the back leaf spectrum sample and the tea pile mixed spectrum sample is determined.

3. The method for processing spectral data of tea pile according to claim 1, characterized in that: The step of deriving the spectrum of the front leaves of the tea pile to obtain a set of derivatives of the spectrum of the front leaves includes: According to the bilinear spectral model, the measurement wavelength is derived based on the front blade spectrum to obtain a derivative set of the front blade spectrum, referring to the following formula: ； in, represents the set of derivatives of the front leaf spectrum, represents the diagonal matrix of the front leaf proportion, represents the measurement wavelength, represents the set of derivatives of the tea pile mixture spectrum, represents the front leaf spectrum, Indicates the proportion of the front leaves of the tea pile, represents the first constant, represents the tea pile mixed spectrum, Represents the tea bud stacking structure information of the tea pile.

4. The method for processing spectral data of tea pile according to claim 1, characterized in that: Before inputting the derivative set into the Gaussian process regression model using the radial basis kernel function to obtain the target front blade quality output by the Gaussian process regression model using the radial basis kernel function, the method further includes: Acquire a verification spectrum and quality data set and a training spectrum and quality data set, wherein the verification spectrum and quality data set includes a derivative set of the front leaf spectrum of the verification tea pile and a quality data set of the verification tea pile; the training spectrum and quality data set includes a derivative set of the front leaf spectrum of the training tea pile and a quality data set of the training tea pile; Based on the derivative set of the front leaf spectra of the verification tea pile and the derivative set of the front leaf spectra of the training tea pile, the similarity covariance function between the derivative set of the front leaf spectra of the verification tea pile and the derivative set of the front leaf spectra of the training tea pile is determined, referring to the following formula: ； in, represents the similarity covariance function, represents the set of derivatives of the front leaf spectra of the validation tea bunch, represents the set of derivatives of the front leaf spectra of the training tea bunches, represents the scaling parameter, represents the diagonal matrix of the front leaf proportion, It represents the set of derivatives of the mixed spectrum of tea piles. represents the set of spectral derivatives of the training tea pile mixture, represents the length scale parameter of the kernel function; represents the noise standard deviation, Represents the Kronecker symbol.

5. The method for processing spectral data of tea pile according to claim 4, characterized in that: The method further comprises: determining a mean function of a set of derivatives of the front leaf spectrum; A Gaussian process regression model is constructed based on the mean function and the similarity covariance function.

6. A spectral data processing device for tea piles, characterized in that: include: A spectrum acquisition module is used to obtain a tea pile mixed spectrum of a tea pile, wherein the tea pile mixed spectrum is obtained by collecting spectral data of the tea pile based on a measurement wavelength, the tea pile is composed of randomly stacked tea buds and leaves, and the tea buds and leaves include: front leaves and back leaves; The spectrum processing module is used to input the mixed spectrum of the tea pile into a bilinear spectrum derivative Gaussian process regression model to obtain the target front leaf quality output by the bilinear spectrum derivative Gaussian process regression model, wherein the bilinear spectrum derivative Gaussian process regression model includes: a bilinear spectrum model and a Gaussian process regression model using a radial basis kernel function, which includes: Inputting the mixed spectrum of the tea pile into the bilinear spectrum model to obtain the spectrum of the front leaves of the tea pile output by the bilinear spectrum model, wherein the bilinear spectrum model is used to correct the mixed spectrum of the tea pile; Derivative the front leaf spectrum of the tea pile to obtain a derivative set of the front leaf spectrum, wherein the derivation is used to eliminate the tea bud stacking structure information of the tea pile; Inputting the derivative set into the Gaussian process regression model using the radial basis kernel function to obtain the target front leaf quality output by the Gaussian process regression model using the radial basis kernel function, wherein the radial basis kernel function in the Gaussian process regression model is used to eliminate the front leaf proportion information of the tea pile, and the Gaussian process regression model is used to predict the target front leaf quality; The bilinear spectral model refers to the following formula: ； ； in, represents the front leaf spectrum, Indicates the proportion of the front leaves of the tea pile, represents the first constant, represents the tea pile mixed spectrum, represents the second constant, Indicates the tea bud stacking structure information of the tea pile, represents the spectrum of the back leaf, Represents noise or random error.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the method for processing spectral data of a tea pile according to any one of claims 1 to 5 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the spectral data processing method of the tea pile according to any one of claims 1 to 5 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the spectral data processing method of the tea pile according to any one of claims 1 to 5 is implemented.

Images