Tea raw material quality evaluation method and device, tea raw material screening method and tea drink

Abstract

The invention discloses a tea raw material quality evaluation method and device, a tea raw material screening method and a tea beverage. The tea raw material quality evaluation method comprises the following steps: establishing a tea quality evaluation model by integrating a spectrum technology and artificial sensory evaluation in advance; acquiring spectral data of the to-be-detected tea sample based on near infrared spectrum scanning; and determining the quality of the to-be-detected tea by using the spectral data of the to-be-detected tea sample and the tea quality evaluation model. By utilizing the scheme of the invention, the quality of the tea leaves can be detected quickly, accurately and nondestructively.

Description

Technical Field

[0001] The present application relates to the field of food industry, and specifically to a method and device for evaluating the quality of tea raw materials, a method for screening tea raw materials, and tea beverages. Background Art

[0002] In modern agriculture and the food industry, tea, as an important cash crop, is crucial for quality testing and evaluation. Tea quality is a comprehensive reflection of both its appearance and internal qualities. Rapid and accurate evaluation of tea quality is crucial for tea processing and trade.

[0003] Traditional tea quality testing relies primarily on manual sensory flavor evaluation. This method relies on the tea taster's professional experience and sensory judgment, which is subject to subjective errors and poor consistency. With the advancement of technology, spectral analysis and artificial intelligence technologies are increasingly being applied to tea quality testing. These include component analysis, as well as emerging technologies such as computer vision systems, near-infrared spectroscopy, electronic noses, and electronic tongues. Component analysis, which uses analytical instruments to examine the internal components of tea leaves, offers more accurate and objective results, but it cannot measure sensory flavors such as flavor and aroma, and is costly and highly destructive to samples. Computer vision systems use image sensors to acquire target image signals, then use image processing systems to convert these signals into digital signals and extract target feature information. While this method is fast, simple, and non-destructive, it currently primarily focuses on the appearance of fresh leaves and finished tea for quality analysis and classification. Electronic noses primarily detect aroma compounds, while electronic tongues primarily identify flavor compounds.

[0004] Sensory evaluation of tea is a fundamental method for assessing tea quality, offering advantages such as convenience, intuitiveness, and speed. However, this method requires high technical expertise from testers and is subject to significant subjective influences, resulting in subjective errors and poor consistency. Furthermore, from an industrial perspective, it also faces challenges with numerous batches and a heavy workload. Existing chemical analysis and physical testing methods require extensive equipment and specialized personnel, are complex to operate, and require long testing times, failing to meet the demands of modern agriculture and the food industry for rapid and efficient tea quality testing. Furthermore, existing tea quality testing technologies can only assess certain aspects of tea and cannot comprehensively and accurately reflect tea quality.

[0005] Therefore, how to quickly, accurately and non-destructively detect the quality of tea is an important issue facing the current field of tea quality detection technology. Summary of the Invention

[0006] The embodiments of the present application provide a method and device for evaluating the quality of tea raw materials, a method for screening tea raw materials, and tea beverages, so as to quickly, accurately, and non-destructively detect the quality of tea and ensure product quality.

[0007] In one aspect, the present invention provides a method for evaluating the quality of tea raw materials, the method comprising:

[0008] Preliminary integration of spectral technology and manual sensory evaluation to establish a tea quality evaluation model;

[0009] Obtaining spectral data of the tea sample to be tested based on near-infrared spectroscopy scanning;

[0010] The quality of the tea leaves to be tested is determined using the spectral data of the tea leaves to be tested and the tea leaves quality evaluation model.

[0011] Optionally, the integrated spectral technology and manual sensory evaluation to establish a tea quality evaluation model includes:

[0012] Collect tea samples;

[0013] Obtain the quality labels corresponding to each tea sample based on sensory evaluation;

[0014] Acquiring spectral data of the tea sample obtained based on near-infrared spectroscopy scanning;

[0015] extracting key features from the spectral data;

[0016] A tea quality evaluation model is trained based on the key features and corresponding quality labels.

[0017] Optionally, the method further comprises: scanning the tea sample multiple times using an infrared spectrometer.

[0018] Optionally, extracting key features from the spectral data includes: determining the key features of the spectral data using a principal component analysis method.

[0019] Optionally, the integrated spectral technology and manual sensory evaluation to establish a tea quality evaluation model further includes: obtaining component information of the tea sample according to the spectral data, the component information including: molecular structure and / or chemical composition;

[0020] The training of the tea quality evaluation model according to the key features and the corresponding quality labels includes:

[0021] A tea quality evaluation model is trained based on the key features, the ingredient information of the tea samples, and the corresponding quality labels.

[0022] Optionally, the method further includes:

[0023] Obtaining component information of the tea sample to be tested according to the spectral data of the tea sample to be tested;

[0024] Determining the quality of the tea leaves to be tested by using the spectral data of the tea leaves to be tested and the tea leaves quality evaluation model includes:

[0025] The component information and spectral data of the tea sample to be tested are input into the tea quality evaluation model, and the quality of the tea to be tested is determined according to the model output.

[0026] Optionally, the method further includes:

[0027] Establishing a tea digital label library based on the tea quality evaluation model, wherein the digital label library includes digital labels of various different categories of high-quality tea;

[0028] The quality of tea is quickly evaluated based on the tea digital label library.

[0029] Optionally, establishing a tea digital label library according to the tea quality evaluation model includes:

[0030] For a specific category of tea, determining a series of high-quality teas according to the tea quality evaluation model;

[0031] Analyzing the spectral data corresponding to the high-quality tea leaves using unsupervised learning technology;

[0032] A digital label corresponding to the specific category of high-quality tea is determined according to the analysis result.

[0033] Optionally, the analyzing the spectral data corresponding to the high-quality tea leaves by using an unsupervised learning technique includes: clustering the spectral data corresponding to the high-quality tea leaves; or performing anomaly detection on the spectral data corresponding to the high-quality tea leaves.

[0034] Optionally, the rapid evaluation of tea quality based on the tea digital label library includes:

[0035] respectively obtaining spectral data of the new tea samples obtained based on near-infrared spectroscopy scanning and digital labels of high-quality tea of ​​the category to which the new tea samples belong;

[0036] Calculating the distance or similarity between the spectral data of the new tea sample and the digital tag;

[0037] The quality of the new tea sample is determined according to the distance or similarity.

[0038] On the other hand, the present invention also provides a device for evaluating the quality of tea raw materials, the device comprising:

[0039] A data acquisition module is used to acquire spectral data of the tea sample to be tested obtained by near-infrared spectroscopy scanning;

[0040] The evaluation module is used to determine the quality of the tea to be tested by using the spectral data of the tea sample to be tested and a tea quality evaluation model established by pre-integrating spectral technology and artificial sensory evaluation.

[0041] Optionally, the device further comprises:

[0042] A model training module is used to establish a tea quality evaluation model by integrating spectral technology and manual sensory evaluation; the model training module includes:

[0043] a sample determination unit for collecting tea samples;

[0044] A label acquisition unit, configured to acquire a quality label corresponding to each tea sample determined based on sensory evaluation;

[0045] A spectral data acquisition unit, configured to acquire spectral data of the tea sample obtained by near-infrared spectroscopy scanning;

[0046] A feature extraction unit, configured to extract key features from the spectral data;

[0047] The training unit is used to train a tea quality evaluation model based on the key features and corresponding quality labels.

[0048] Optionally, the model training module further includes:

[0049] a component information acquiring unit, configured to acquire component information of the tea sample according to the spectral data, wherein the component information includes: molecular structure and / or chemical composition;

[0050] The training unit trains a tea quality evaluation model based on the key features, the component information of the tea samples, and the corresponding quality labels.

[0051] Optionally, the device further comprises:

[0052] A component information acquisition module, configured to acquire component information of the tea sample to be tested based on the spectral data of the tea sample to be tested;

[0053] The evaluation module is specifically used to input the component information and spectral data of the tea sample to be tested into the tea quality evaluation model, and determine the quality of the tea to be tested according to the model output.

[0054] Optionally, the device further comprises:

[0055] A tea digital label library establishment module is used to establish a tea digital label library according to the tea quality evaluation model, wherein the digital label library includes digital labels of various different categories of high-quality tea;

[0056] The rapid evaluation module is used to quickly evaluate the quality of tea leaves according to the digital label library.

[0057] On the other hand, the present invention also provides a method for screening tea raw materials, comprising:

[0058] Obtaining multiple candidate tea raw materials;

[0059] Using the tea raw material quality evaluation method according to any one of claims 1 to 10 to evaluate the quality of each candidate tea raw material, and obtain a quality evaluation result;

[0060] Determine whether each candidate tea raw material meets the set quality grade according to the quality evaluation result; and use the candidate tea raw materials that meet the set quality grade as target tea raw materials.

[0061] On the other hand, an embodiment of the present application further provides a tea beverage, which includes tea raw materials, and the tea raw materials have a set quality grade, and the quality grade is determined using the tea raw material quality evaluation method.

[0062] On the other hand, an embodiment of the present application also provides a computer-readable storage medium, which is a non-volatile storage medium or a non-transient storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the steps of the tea raw material quality evaluation method or the steps of the tea raw material screening method.

[0063] The tea raw material quality evaluation method and device provided in the embodiments of the present application pre-integrate spectral technology and artificial sensory evaluation to establish a tea quality evaluation model. For the tea sample to be tested, its spectral data is obtained based on near-infrared spectroscopy technology. The spectral data and the tea quality model are used to quickly, accurately and non-destructively detect the quality of the tea.

[0064] Furthermore, by using the solution of the present application, a digital label library of tea can be established, and the quality of tea can be quickly evaluated based on the digital labels in the digital label library of tea.

[0065] Compared with existing technologies, the tea raw material quality evaluation method and device provided in the embodiments of the present application have the following beneficial effects:

[0066] 1. Rapid and non-destructive: This application solution uses near-infrared-visible spectroscopy to rapidly scan tea leaves, combined with artificial intelligence analysis, to quickly and non-destructively determine tea quality. Compared to traditional manual sensory flavor evaluation methods, this application solution significantly shortens testing time and improves efficiency, while also avoiding the subjective errors and poor consistency inherent in manual sensory evaluation.

[0067] 2. Accurate and Comprehensive: This application utilizes the learning and analysis capabilities of artificial intelligence to establish a tea quality assessment model that can more comprehensively and accurately reflect tea quality. Compared to existing chemical analysis and physical testing methods, this application not only tests the external characteristics of tea, but also comprehensively analyzes the internal composition and quality of tea, providing more accurate and comprehensive tea quality testing results.

[0068] 3. No chemical reagents or specialized operators required: This solution does not require any chemical reagents or specialized operators for tea quality testing. Instead, it only requires near-infrared-visible spectrum scanning equipment and artificial intelligence analysis software to quickly and accurately detect tea quality. This not only reduces testing costs but also minimizes the environmental and human health hazards of chemical reagents, improving the convenience and safety of testing.

[0069] 4. Ease of promotion and application: This solution is simple to operate and easy to promote and apply for tea quality testing. Whether it is a small tea production company or a large food processing company, this solution can be used to quickly and accurately test tea quality, thereby improving the quality and market competitiveness of tea products.

[0070] The tea raw material screening method and tea beverage provided in the embodiments of the present application perform accurate and scientific quality evaluation and screening of the tea raw materials required for the product, which can better ensure product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a flow chart of establishing a tea quality evaluation model in an embodiment of the present application;

[0072] Figure 2 This is another flow chart for establishing a tea quality evaluation model in an embodiment of the present application;

[0073] Figure 3 This is a flow chart of the tea raw material quality evaluation method provided in the embodiment of the present application;

[0074] Figure 4 This is a flow chart for quickly evaluating tea quality in the embodiment of the present application;

[0075] Figure 5 This is a structural diagram of a tea raw material quality evaluation device provided in an embodiment of the present application;

[0076] Figure 6 This is a structural diagram of the model training module in the embodiment of the present application;

[0077] Figure 7 This is another structural diagram of the tea raw material quality evaluation device provided in an embodiment of the present application;

[0078] Figure 8 This is another structural diagram of the tea raw material quality evaluation device provided in an embodiment of the present application;

[0079] Figure 9 This is a flow chart of the tea raw material screening method provided in the embodiment of the present application. DETAILED DESCRIPTION

[0080] In order to make the above-mentioned objectives, features and beneficial effects of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0081] Existing tea quality analysis is primarily guided by specific chemical analyses. These analyses typically attempt to correlate the amounts of specific compounds found in tea, such as caffeine, epigallocatechin gallate (EGCG), and moisture, with its quality. However, there are some significant issues with this approach.

[0082] Initially, accurate measurement of these specific compounds could only be performed directly using chromatographic methods, which are typically coupled with either a UV detector (for low-end equipment) or a mass spectrometer (for high-end equipment). While these devices offer high accuracy, they are expensive to operate and maintain, have slow processing speeds, require extensive sample preparation, and require skilled personnel to operate and analyze the results.

[0083] Secondly, the ultimate determinant of tea quality is sensory (flavor and aroma). Knowing the content of only a few compounds does not translate well into the taste of tea, and even if the compounds are accurately analyzed, it is impossible to fully and reliably determine whether the tea sample in question has a certain quality.

[0084] Near-infrared spectroscopy, which uses light to penetrate a sample and generate signals based on the chemical compounds present, has recently been used to quantify these compounds associated with tea quality. However, these studies have required correlating NIR spectra with chromatographic results to create calibration curves to quantify the compounds of interest in blind samples.

[0085] In order to quickly, accurately and non-destructively evaluate the quality of tea, the embodiments of the present application provide a method and device for evaluating the quality of tea raw materials. A tea quality evaluation model is established by pre-integrating spectral technology and manual sensory evaluation. For the tea samples to be tested, their spectral data is obtained based on spectral technology, and the spectral data and the tea quality model are used to evaluate the quality of the tea.

[0086] The following first describes in detail the process of establishing the tea quality evaluation model in the embodiment of this application.

[0087] It should be noted that in some embodiments, a unified evaluation model can be trained for different types of tea. That is, regardless of the type of tea, the same tea quality model is used for evaluation. This tea quality evaluation model needs to be universal. Accordingly, during training, a large number of different types of tea samples need to be collected.

[0088] In some embodiments, tea can be categorized by fermentation level and processing method, for example, into green tea, white tea, yellow tea, oolong tea, black tea, and dark tea. A large number of tea samples can be collected for each category to train a tea quality assessment model for that category. This makes the tea quality assessment model for each category more targeted and enables more granular quality assessment.

[0089] It should be noted that other classification methods may also be used to classify tea, such as origin, etc., which is not limited in the embodiments of the present application.

[0090] The method for establishing the tea quality evaluation model described later is applicable to both of the above two different models. The only difference is the selection of tea samples.

[0091] like Figure 1 FIG. 1 is a flow chart of establishing a tea quality evaluation model in an embodiment of the present application, which includes the following steps:

[0092] Step 101: Collect tea samples.

[0093] Step 102: Obtain the quality label corresponding to each tea sample determined based on sensory evaluation.

[0094] Specifically, experts can conduct a sensory evaluation of tea leaves, comparing them to standard samples to determine pass / fail. For example, ten professionally trained tea tasters could conduct a sensory evaluation of tea samples, comparing them to standard samples, and comprehensively assessing the tea samples' appearance (e.g., shape, tenderness, color, wholeness, clarity, etc.) and internal qualities (e.g., tea soup color, aroma, flavor, and leaf base) to determine whether the tea samples are acceptable. For example, tea with a total score of less than 80 points or an aroma score of less than 20 points against the standard samples would be considered unqualified and unsuitable for use in tea beverages. Tea with a total score of more than 90 points and an aroma and flavor score of 25 or above would be considered high-quality tea raw materials.

[0095] In some embodiments, various scores and standards may be compared to determine refined quality labels, such as using quality scores as labels.

[0096] It does not mean that the quality label can be "qualified" and "unqualified" or "score", which can be determined according to the application scenario of the tea quality evaluation model.

[0097] Step 103: Acquire spectral data of the tea sample obtained by near-infrared spectroscopy scanning.

[0098] Specifically, a specific spectrum analyzer may be used to scan the tea sample multiple times (eg, 2-10 times) to obtain corresponding spectrum data. Step 104: extract key features from the spectrum data.

[0099] In some embodiments, a principal component analysis (PCA) method may be used to extract key features of the spectral data.

[0100] PCA is a multivariate statistical analysis method that transforms a set of potentially correlated variables into a set of linearly uncorrelated variables through an orthogonal transformation. These new variables are called principal components. PCA is a statistical tool used for data dimensionality reduction and feature extraction, rather than directly identifying specific compounds or components. In spectral data analysis, PCA can extract the most important features from hundreds of wavelengths, reducing data dimensionality while retaining the most critical information.

[0101] The main steps of principal component analysis are as follows:

[0102] 1. Data standardization: Since spectral data involves multiple wavelengths, the data must first be standardized, such as using standard normalization or vector normalization, to ensure that the contribution of each wavelength is balanced.

[0103] 2. Calculate the covariance matrix: Calculate the covariance matrix of the spectral data to understand the correlation between different wavelengths.

[0104] 3. Eigenvalue decomposition: Find the principal components by performing eigenvalue decomposition on the covariance matrix.

[0105] As mentioned earlier, principal components do not directly identify specific chemical components or physical entities in the raw data. Instead, they represent abstract directions of variation in the data. Each principal component can be considered a representation of a specific pattern or trend in the data, and these patterns can be the result of multiple different chemical or physical processes. Therefore, principal components provide more of a means of data simplification and interpretation than directly identifying specific compounds.

[0106] 4. Select principal components: You can select the first few principal components for subsequent analysis based on the variance percentage.

[0107] Step 105: training a tea quality evaluation model based on the key features and corresponding quality labels.

[0108] For example, a training dataset includes near-infrared spectral data of at least 100 tea samples and their corresponding sensory evaluation results (i.e., quality labels). Key features are extracted from the near-infrared spectral data of the tea samples. There can be multiple key features, for example, the 3-5 most important ones are selected. These key features and their corresponding quality labels are then used to train a tea quality evaluation model.

[0109] In specific implementation, the tea quality evaluation model may adopt a linear regression model, a support vector machine (SVM), a random forest or a neural network, etc., which is not limited in the present embodiment. Moreover, the training method may adopt the existing training method of the corresponding model.

[0110] Furthermore, during model training, cross-validation techniques can be used to evaluate the model's generalization ability. For example, using the K-fold cross-validation method, the entire dataset is divided into K equal (or nearly equal) sized subsets, commonly referred to as "folds." The model training and validation process is repeated K times, each time using a different fold as the validation set, while the remaining K-1 folds serve as the training set. Cross-validation allows model training to maximize the performance of the model using limited data.

[0111] In addition, the performance of the tea quality evaluation model can be tested based on the test set, and further optimization and adjustment can be made based on the test results to improve the performance of the tea quality evaluation model.

[0112] In some embodiments, the tea quality evaluation model can also be established by comprehensively considering the tea composition information and the above key characteristics. Figure 2 This is explained in detail.

[0113] Reference Figure 2 , Figure 2 This is another flow chart for establishing a tea quality evaluation model in an embodiment of the present application, including the following steps:

[0114] Step 201: Collect tea samples.

[0115] Step 202: Obtain the quality label corresponding to each tea sample determined based on sensory evaluation.

[0116] Step 203: Acquire spectral data of the tea sample based on near-infrared spectroscopy scanning.

[0117] Step 204: extract key features from the spectral data.

[0118] Step 205 : Acquire component information of the tea sample according to the spectral data, wherein the component information includes molecular structure and / or chemical composition.

[0119] The method of using spectral data to obtain the molecular structure and chemical composition of tea can adopt some existing related technologies.

[0120] Step 206 : training a tea quality evaluation model based on the key features, the ingredient information of the tea sample, and the corresponding quality labels.

[0121] The tea quality evaluation model described above can be used to quickly, accurately, and non-destructively evaluate the quality of tea samples. For example, it can determine whether the tea samples are qualified, determine the quality grade of the tea samples, or score the quality of the tea samples to obtain a quality score. It should be noted that when constructing the tea quality evaluation model, a regression model or a classification model can be used, depending on the desired form of the evaluation results, and this embodiment of the present application does not limit this.

[0122] like Figure 3 FIG. 1 is a flow chart of a method for evaluating the quality of tea raw materials provided in an embodiment of the present application, comprising the following steps:

[0123] Step 301: Acquire spectral data of a tea sample to be tested based on near-infrared spectral scanning.

[0124] Specifically, a specific spectrum analyzer may be used to scan the tea sample to obtain corresponding spectrum data.

[0125] Step 302: Determine the quality of the tea leaves to be tested by using the spectral data of the tea leaves to be tested and a pre-established tea quality evaluation model.

[0126] Specifically, the spectral data is input into the tea quality evaluation model (such as Figure 1 The tea quality evaluation model established by the method shown in the figure is used to determine whether the tea to be tested is qualified according to the model output, or to determine the quality grade of the tea to be tested, or to obtain the quality score of the tea to be tested.

[0127] In some embodiments, spectral data may also be used to obtain component information of the tea sample to be tested, where the component information includes molecular structure and / or chemical composition.

[0128] Accordingly, the spectral data and the component information are input into the tea quality evaluation model (such as Figure 2 The tea quality evaluation model established by the method shown in the figure is used to determine whether the tea to be tested is qualified or obtain the quality score of the tea to be tested based on the model output.

[0129] In some embodiments, the tea quality evaluation model can also be used to establish a tea digital label library, and the tea digital label library can be used to achieve rapid evaluation of tea quality.

[0130] like Figure 4 As shown in FIG. 1 , a flow chart for quickly evaluating tea quality according to an embodiment of the present application is shown, which includes the following steps:

[0131] Step 401 : establishing a tea digital label library based on a tea quality evaluation model, wherein the digital label library includes digital labels of various different categories of high-quality tea.

[0132] Specifically, for a specific category of tea, a series of high-quality teas can be determined based on the tea quality evaluation model; the spectral data corresponding to the high-quality teas can be analyzed using unsupervised learning techniques, such as clustering methods, anomaly detection methods, etc.; and the digital labels corresponding to the specific category of high-quality teas can be determined based on the analysis results.

[0133] The digital label may be a multi-dimensional data structure including thousands of floating point numbers, each of which reflects a specific response of the tea sample in the near infrared spectrum.

[0134] For each category of tea, such as green tea, white tea, yellow tea, oolong tea, black tea, dark tea, etc., a digital label for high-quality tea of ​​the corresponding category is established. The digital label can also be regarded as the quality standard for this type of tea.

[0135] The tea digital label library can include digital labels for high-quality teas based on various classifications to meet the needs of different evaluation applications.

[0136] For example, when evaluating the quality of green tea from different origins, digital labels based on high-quality tea across the country can be used for evaluation.

[0137] For example, when evaluating the quality of green tea from the same origin, digital labels based on high-quality tea from that origin can be used for evaluation.

[0138] Step 402: quickly evaluate the quality of the tea leaves according to the tea digital label library.

[0139] Specifically, spectral data of a new tea sample obtained through near-infrared spectroscopy scanning and a digital label of the high-quality tea category to which the new tea sample belongs can be obtained. The distance (e.g., Euclidean distance) or similarity (e.g., Jaccard similarity coefficient, adjusted Rand coefficient, etc.) between the spectral data of the new tea sample and the digital label can be calculated. The quality of the new tea sample can be determined based on the distance or similarity. The closer the distance, the better the quality; the greater the similarity, the better the quality.

[0140] Accordingly, the present application also provides a tea raw material quality evaluation device, such as Figure 5 The figure shows a structural diagram of the device.

[0141] In this embodiment, the tea raw material quality evaluation device 500 includes the following modules:

[0142] The data acquisition module 501 is used to acquire spectral data of the tea sample to be tested obtained by near infrared spectroscopy scanning;

[0143] The evaluation module 502 is used to determine the quality of the tea leaves to be tested by using the spectral data of the tea leaves to be tested and a tea quality evaluation model 50 established by pre-combining spectral technology and manual sensory evaluation.

[0144] In some embodiments, the tea raw material quality evaluation device 500 may further include a model training module, such as Figure 6 The figure shows a structural diagram of the model training module.

[0145] In this embodiment, the model training module 600 includes the following units:

[0146] The sample determination unit 601 is used to collect tea samples;

[0147] The label acquisition unit 602 is used to obtain the quality label corresponding to each tea sample determined based on the sensory evaluation;

[0148] A spectral data acquisition unit 603 is used to acquire spectral data of the tea sample obtained based on near-infrared spectral scanning;

[0149] A feature extraction unit 604 is used to extract key features from the spectral data;

[0150] The training unit 605 is used to train a tea quality evaluation model based on the key features and corresponding quality labels.

[0151] In some embodiments, the model training module 600 may further include: a component information acquisition unit (not shown), which is used to obtain component information of the tea sample based on the spectral data, and the component information includes: molecular structure and / or chemical composition.

[0152] Accordingly, the training unit 605 trains a tea quality evaluation model based on the key features, the component information of the tea samples, and the corresponding quality labels.

[0153] Accordingly, if Figure 7As shown, based on the tea quality evaluation model of this embodiment, the tea raw material quality evaluation device 500 may further include a component information acquisition module 503 for acquiring component information of the tea sample to be tested based on the spectral data of the tea sample to be tested. Accordingly, the evaluation module 502 needs to input the component information and spectral data of the tea sample to be tested into the tea quality evaluation model and determine the quality of the tea to be tested based on the model output.

[0154] like Figure 8 , which is another structural schematic diagram of the tea raw material quality evaluation device provided in an embodiment of the present application.

[0155] Relative to Figure 5 In the embodiment shown, the tea material quality evaluation device 500 in this embodiment further includes: a tea digital label library establishment module 504 and a rapid evaluation module 505.

[0156] The tea digital label library establishment module 504 is used to establish a tea digital label library 70 according to the tea quality evaluation model, wherein the digital label library 70 includes digital labels of various different categories of high-quality tea;

[0157] The rapid evaluation module 505 is used to perform a rapid evaluation on the quality of tea leaves according to the digital label library.

[0158] For the specific implementation of the above modules and units, please refer to the description in the previous embodiment of the method of this application, which will not be repeated here.

[0159] Accordingly, the present invention also provides a method for screening tea raw materials, such as Figure 9 FIG. 1 is a flow chart of the method, comprising the following steps:

[0160] Step 901: Acquire multiple candidate tea raw materials.

[0161] Step 902: Use the pre-established tea quality evaluation model to perform quality evaluation on each candidate tea raw material to obtain a quality evaluation result.

[0162] Specifically, samples can be taken from each candidate tea raw material, and the tea raw material quality evaluation method provided in the previous embodiment of the present application can be used to complete the evaluation of each sample to obtain the quality evaluation results corresponding to each candidate tea raw material.

[0163] Step 903: Determine whether each candidate tea raw material meets the set quality level based on the quality evaluation result.

[0164] Step 904: select candidate tea raw materials that meet the set quality level as target tea raw materials.

[0165] As mentioned above, based on different quality evaluation result requirements, the output of the tea quality evaluation model can be in the form of corresponding acceptance, quality grade, or quality score. Regardless of the output form, in step 903, it can be determined whether each candidate tea raw material meets the set quality grade based on the quality evaluation results.

[0166] For example, in some embodiments, only a coarse-grained division of tea quality is required, and the division is performed according to the quality grade actually required by the product. The tea quality evaluation model outputs the result as whether the candidate tea raw material is qualified or unqualified, so that the qualified candidate tea raw material can be determined as the tea raw material that meets the set quality grade.

[0167] For example, in some embodiments, it is necessary to perform a fine-grained classification of tea quality, such as into three grades: superior, medium, and inferior, and establish a correspondence between each grade and a quality score range. The tea quality evaluation model outputs the quality score of the candidate tea raw material. Based on this correspondence, the quality grade of the candidate tea raw material can be determined, and then whether the candidate tea raw material meets the set quality grade.

[0168] By using the tea raw material screening method provided in the embodiment of the present application, the tea raw materials required for the product can be screened out quickly and accurately, providing effective support and guarantee for product manufacturers in product selection and ensuring product quality.

[0169] Correspondingly, an embodiment of the present application further provides a tea beverage, which includes tea raw materials, wherein the tea raw materials have a set quality grade, and the quality grade is determined using the tea raw material quality evaluation method provided in the embodiment of the present application.

[0170] Since the tea raw materials in the product have undergone accurate and scientific quality evaluation, the quality of the product can be better guaranteed.

[0171] In specific implementations, the various modules / units included in the various devices and products described in the above embodiments may be software modules / units, hardware modules / units, or partially software modules / units and partially hardware modules / units. For example, for various devices and products applied to or integrated into a chip, the various modules / units included therein may all be implemented in the form of hardware such as circuits, or at least some of the modules / units may be implemented in the form of software programs, which run on a processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated into a chip module, the various modules / units included therein may all be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as a chip, circuit module, etc.) or different components of the chip module, or at least some of the modules / units may be located in the same component (such as a chip, circuit module, etc.) or different components of the chip module, or at least some of the modules / units may be located in the same component (such as a chip, circuit module, etc.) or different components of the chip module. The element can be implemented in the form of a software program, which runs on the processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in the terminal, the various modules / units contained therein can all be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or different components in the terminal, or, at least some modules / units can be implemented in the form of a software program, which runs on the processor integrated inside the terminal, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.

[0172] The present application also discloses a storage medium, which is a computer-readable storage medium having a computer program stored thereon. When the computer program is run, the computer program can execute Figure 1 or Figure 2 or Figure 3 or Figure 4 or Figure 9 Some or all of the steps of the method shown. The storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. The storage medium may also include a non-volatile memory or a non-transitory memory.

[0173] It should be understood that the term "and / or" as used herein simply describes an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " as used herein indicates that the related objects are in an "or" relationship.

[0174] The term "plurality" used in the embodiments of the present application refers to two or more.

[0175] The first, second, etc. descriptions appearing in the embodiments of this application are only for illustration and distinction of the description objects. There is no order, nor does it indicate any special limitation on the number of devices in the embodiments of this application, and cannot constitute any limitation on the embodiments of this application.

[0176] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0177] The above embodiments may be implemented in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless means.

[0178] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0179] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may be physically arranged separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0180] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit stored in a storage medium includes a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to perform some steps of the method described in various embodiments of the present application.

[0181] Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims.

Claims

1. A method for evaluating the quality of tea raw materials, characterized in that: The method comprises: Preliminary integration of spectral technology and manual sensory evaluation to establish a tea quality evaluation model; Obtaining spectral data of the tea sample to be tested based on near-infrared spectroscopy scanning; The quality of the tea leaves to be tested is determined using the spectral data of the tea leaves to be tested and the tea leaves quality evaluation model.

2. The method for evaluating the quality of tea raw materials according to claim 1, wherein: The tea quality evaluation model established by integrating spectral technology and manual sensory evaluation includes: Collect tea samples; Obtain the quality labels corresponding to each tea sample based on sensory evaluation; Acquiring spectral data of the tea sample obtained based on near-infrared spectroscopy scanning; extracting key features from the spectral data; A tea quality evaluation model is trained based on the key features and corresponding quality labels.

3. The method for evaluating the quality of tea raw materials according to claim 2, wherein: The method further comprises: The tea sample is scanned multiple times using an infrared spectrometer.

4. The method for evaluating the quality of tea raw materials according to claim 2, wherein: The extracting key features from the spectral data comprises: The principal component analysis method was used to determine the key features of the spectral data.

5. The method for evaluating the quality of tea raw materials according to claim 2, wherein: The tea quality evaluation model established by integrating spectral technology and manual sensory evaluation also includes: Acquiring component information of the tea sample according to the spectral data, wherein the component information includes: molecular structure and / or chemical composition; The training of the tea quality evaluation model according to the key features and the corresponding quality labels includes: A tea quality evaluation model is trained based on the key features, the ingredient information of the tea samples, and the corresponding quality labels.

6. The method for evaluating the quality of tea raw materials according to claim 5, wherein: The method further comprises: Obtaining component information of the tea sample to be tested according to the spectral data of the tea sample to be tested; Determining the quality of the tea leaves to be tested by using the spectral data of the tea leaves to be tested and the tea leaves quality evaluation model includes: The component information and spectral data of the tea sample to be tested are input into the tea quality evaluation model, and the quality of the tea to be tested is determined according to the model output.

7. The method for evaluating the quality of tea raw materials according to any one of claims 1 to 6, characterized in that: The method further comprises: Establishing a tea digital label library based on the tea quality evaluation model, wherein the digital label library includes digital labels of various different categories of high-quality tea; The quality of tea is quickly evaluated based on the tea digital label library.

8. The method for evaluating the quality of tea raw materials according to claim 7, wherein: The establishment of a tea digital label library according to the tea quality evaluation model includes: For a specific category of tea, determining a series of high-quality teas according to the tea quality evaluation model; Analyzing the spectral data corresponding to the high-quality tea leaves using unsupervised learning technology; A digital label corresponding to the specific category of high-quality tea is determined according to the analysis result.

9. The method for evaluating the quality of tea raw materials according to claim 8, wherein: The analyzing of the spectral data corresponding to the high-quality tea leaves by using unsupervised learning technology includes: Clustering the spectral data corresponding to the high-quality tea leaves; or Anomaly detection is performed on the spectral data corresponding to the high-quality tea leaves.

10. The method for evaluating the quality of tea raw materials according to claim 8, wherein: The rapid evaluation of tea quality according to the tea digital label library includes: respectively obtaining spectral data of the new tea samples obtained based on near-infrared spectroscopy scanning and digital labels of high-quality tea of ​​the category to which the new tea samples belong; Calculating the distance or similarity between the spectral data of the new tea sample and the digital tag; The quality of the new tea sample is determined according to the distance or similarity.

11. A tea raw material quality evaluation device, characterized in that: The device comprises: A data acquisition module is used to acquire spectral data of the tea sample to be tested obtained by near-infrared spectroscopy scanning; The evaluation module is used to determine the quality of the tea to be tested by using the spectral data of the tea sample to be tested and a tea quality evaluation model established by pre-integrating spectral technology and artificial sensory evaluation.

12. The tea raw material quality evaluation device according to claim 11, characterized in that: The device further comprises: A model training module is used to establish a tea quality evaluation model by integrating spectral technology and manual sensory evaluation; the model training module includes: a sample determination unit for collecting tea samples; A label acquisition unit, configured to acquire a quality label corresponding to each tea sample determined based on sensory evaluation; A spectral data acquisition unit, configured to acquire spectral data of the tea sample obtained by near-infrared spectroscopy scanning; A feature extraction unit, configured to extract key features from the spectral data; The training unit is used to train a tea quality evaluation model based on the key features and corresponding quality labels.

13. The tea raw material quality evaluation device according to claim 12, characterized in that: The model training module also includes: a component information acquiring unit, configured to acquire component information of the tea sample according to the spectral data, wherein the component information includes: molecular structure and / or chemical composition; The training unit trains a tea quality evaluation model based on the key features, the component information of the tea samples, and the corresponding quality labels.

14. The tea raw material quality evaluation device according to claim 13, characterized in that: The device further comprises: A component information acquisition module, configured to acquire component information of the tea sample to be tested based on the spectral data of the tea sample to be tested; The evaluation module is specifically used to input the component information and spectral data of the tea sample to be tested into the tea quality evaluation model, and determine the quality of the tea to be tested according to the model output.

15. The tea raw material quality evaluation device according to any one of claims 11 to 14, characterized in that: The device further comprises: A tea digital label library establishment module is used to establish a tea digital label library according to the tea quality evaluation model, wherein the digital label library includes digital labels of various different categories of high-quality tea; The rapid evaluation module is used to quickly evaluate the quality of tea leaves according to the digital label library.

16. A method for screening tea raw materials, characterized in that: The method comprises: Obtaining multiple candidate tea raw materials; Using the tea raw material quality evaluation method according to any one of claims 1 to 10 to evaluate the quality of each candidate tea raw material, and obtain a quality evaluation result; Determine whether each candidate tea raw material meets the set quality grade according to the quality evaluation result; and use the candidate tea raw materials that meet the set quality grade as target tea raw materials.

17. A tea beverage, characterized in that: The tea beverage comprises tea raw materials, and the tea raw materials have a set quality grade, and the quality grade is determined by using the tea raw material quality evaluation method according to any one of claims 1 to 10.

18. A computer-readable storage medium, wherein the computer-readable storage medium is a non-volatile storage medium or a non-transient storage medium, and a computer program is stored thereon, wherein: When the computer program is executed by a processor, the steps of the method for evaluating the quality of tea raw materials according to any one of claims 1 to 10 are executed, or the steps of the method for screening tea raw materials according to claim 16 are executed.

Images