Tobacco leaf chemical component content prediction method, device, equipment, medium and product

Abstract

The invention discloses a tobacco leaf chemical component content prediction method, device and equipment, a medium and a product. The method comprises the following steps: determining a historical climate data matrix corresponding to a target tobacco leaf producing area in a historical duration; wherein the historical climate data matrix comprises historical climate data corresponding to a plurality of target climate factors; and processing the historical climate data matrix according to a pre-trained chemical component content prediction model corresponding to the to-be-predicted tobacco leaf grade, and predicting to obtain predicted chemical component data corresponding to the tobacco leaves of the to-be-predicted tobacco leaf grade produced in the target tobacco leaf producing area within the historical duration. According to the technical scheme provided by the embodiment of the invention, the effect of predicting the content of the chemical components contained in the tobacco leaves of the corresponding tobacco leaf grade produced in the tobacco leaf producing area can be realized only according to the neural network model and the historical climate data about the climate factors of any tobacco leaf producing area without acquiring an actual sample of the tobacco leaf.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device, equipment, medium and product for predicting the content of chemical components in tobacco leaves. Background Art

[0002] Tobacco is a leaf crop and one of the most important cash crops. The quality of tobacco leaves has a decisive influence on the quality of cigarettes, and climate factors (temperature, humidity, rainfall, etc.) have a key impact on the characteristics and quality of tobacco leaves. The characteristics and quality of tobacco leaves are directly related to their chemical composition (nicotine, total nitrogen, total sugar, reducing sugar, etc.). Therefore, how to predict the chemical composition of tobacco leaves based on climate factors is of great value, which can provide important decision-making support for the production of flue-cured tobacco.

[0003] In the related art, most of the existing methods for predicting the chemical composition of tobacco leaves are based on near-infrared spectral data. In this way, tobacco leaf samples need to be obtained to measure the chemical composition of tobacco leaves in a certain year from a certain origin, making it impossible to predict the chemical composition of tobacco leaves before obtaining the tobacco leaves. This method of predicting the chemical composition of tobacco leaves may have problems such as low efficiency and low accuracy, and consumes manpower and material costs, and reduces the timeliness of the prediction of the content of tobacco leaf chemical components. Summary of the invention

[0004] The present invention provides a method, device, equipment, medium and product for predicting the content of chemical components in tobacco leaves, so as to achieve the effect of predicting the content of chemical components contained in tobacco leaves of corresponding tobacco grade produced in any tobacco producing area based only on a neural network model and historical climate data on climate factors of the tobacco producing area without obtaining actual tobacco leaf samples.

[0005] According to one aspect of the present invention, a method for predicting the content of chemical components in tobacco leaves is provided, the method comprising:

[0006] Determine a historical climate data matrix corresponding to a target tobacco leaf producing area within a historical period; wherein the historical period at least includes a time interval associated with the tobacco leaf production law of the target tobacco leaf producing area; the historical climate data matrix includes historical climate data corresponding to a plurality of target climate factors;

[0007] The historical climate data matrix is ​​processed according to a chemical component content prediction model corresponding to the tobacco leaf grade to be predicted obtained by pre-training, and the predicted chemical component data corresponding to the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period are predicted;

[0008] Among them, the chemical component content prediction model is used to predict the chemical composition data of tobacco of the to-be-predicted tobacco grade produced in the tobacco producing area within the historical period according to the historical climate data of the tobacco producing area within the historical period; the chemical component content prediction model is obtained by training a pre-constructed deep learning model based on the sample climate data matrix of the sample tobacco producing area within the historical period and the real chemical composition data of tobacco of the sample tobacco grade produced in the sample tobacco producing area within the historical period.

[0009] According to another aspect of the present invention, a device for predicting the content of chemical components in tobacco leaves is provided, the device comprising:

[0010] A climate data matrix determination module is used to determine the historical climate data matrix corresponding to the target tobacco producing area within a historical period; wherein the historical period at least includes a time interval associated with the tobacco production law of the target tobacco producing area; the historical climate data matrix includes historical climate data corresponding to multiple target climate factors;

[0011] A chemical composition prediction module is used to process the historical climate data matrix according to a pre-trained chemical composition content prediction model corresponding to the tobacco grade to be predicted, and predict the predicted chemical composition data corresponding to the tobacco of the tobacco grade to be predicted produced in the target tobacco producing area within the historical period; wherein the chemical composition content prediction model is used to predict the chemical composition data of the tobacco of the tobacco grade to be predicted produced in the tobacco producing area within the historical period according to the historical climate data of the tobacco producing area within the historical period; the chemical composition content prediction model is obtained by training a pre-constructed deep learning model based on the sample climate data matrix of the sample tobacco producing area within the historical period and the real chemical composition data of the tobacco of the tobacco grade to be predicted produced in the sample tobacco producing area within the historical period.

[0012] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for predicting the content of chemical components in tobacco leaves described in any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for predicting the content of chemical components in tobacco leaves described in any embodiment of the present invention when executed.

[0017] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program, and the computer program, when executed by a processor, implements the method for predicting the content of chemical components in tobacco leaves as described in any embodiment of the present invention.

[0018] The technical solution of the embodiment of the present invention provides a data basis for the subsequent prediction of the content of chemical components in tobacco leaves by determining the historical climate data matrix corresponding to the target tobacco producing area within the historical period, wherein the historical climate data matrix includes historical climate data corresponding to multiple target climate factors. In addition, the constructed historical climate data matrix can facilitate the characterization of the correlation between climate changes and seasonal and date changes, which helps to improve the prediction accuracy of the content of chemical components in tobacco leaves. Furthermore, the historical climate data matrix is ​​processed according to the pre-trained prediction model of the chemical component content corresponding to the tobacco leaf grade to be predicted, and the predicted chemical component data corresponding to the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period are predicted. This solves the problem in the related technology that the chemical composition of the tobacco leaves cannot be predicted before the tobacco leaves are obtained, which leads to low efficiency and accuracy in the prediction of the chemical composition of the tobacco leaves, and wastes manpower and material costs. This achieves the effect of predicting the chemical component content contained in the tobacco leaves of the corresponding tobacco leaf grade produced by any tobacco leaf producing area without obtaining actual tobacco leaf samples, only based on the neural network model and the historical climate data on climate factors of the tobacco leaf producing area, which greatly improves the timeliness of the prediction of the chemical component content of the tobacco leaves, improves the prediction efficiency and prediction accuracy of the chemical component content of the tobacco leaves, and to a certain extent provides important decision-making support for the production of flue-cured tobacco, and has important practical significance and application prospects.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1is a flow chart of a method for predicting the content of chemical components in tobacco leaves provided in Example 1 of the present invention;

[0022] Figure 2 is a flow chart of a method for predicting the content of chemical components in tobacco leaves provided in Example 2 of the present invention;

[0023] Figure 3 is a flow chart of a method for predicting the content of chemical components in tobacco leaves provided in Example 3 of the present invention;

[0024] Figure 4 2 is a schematic diagram of the structure of a device for predicting the content of chemical components in tobacco leaves provided according to a fourth embodiment of the present invention;

[0025] Figure 5 It is a schematic diagram of the structure of an electronic device for implementing the method for predicting the content of chemical components in tobacco leaves according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Embodiment 1

[0029] Figure 1This is a flowchart of a method for predicting the content of chemical components in tobacco leaves provided by the first embodiment of the present invention. This embodiment can be applied to predicting the content of chemical components contained in tobacco leaves of any grade of tobacco leaves produced in any tobacco leaf producing area. The method can be executed by a tobacco leaf chemical component content prediction device, which can be implemented in the form of hardware and / or software, and can be configured in a terminal and / or server. Figure 1 As shown, the method includes:

[0030] S110, determining a historical climate data matrix corresponding to a target tobacco leaf producing area within a historical period; wherein the historical climate data matrix includes historical climate data corresponding to a plurality of target climate factors.

[0031] Among them, the target tobacco leaf producing area can be a tobacco leaf producing area for which the tobacco leaf chemical component content is to be predicted. The tobacco leaf producing area generally refers to the area where tobacco leaves are planted and produced, and these areas generally have climatic conditions, soil environment and planting techniques suitable for tobacco leaf growth. The historical duration can be any duration before the current moment. In this embodiment, the historical duration at least includes a time interval associated with the tobacco leaf production law of the target tobacco leaf producing area. That is to say, the historical duration can be a time interval determined according to the tobacco leaf growth law of the target tobacco leaf producing area. Exemplarily, it is assumed that the tobacco leaf growth law of the target tobacco leaf producing area is: greenhouse seedling cultivation in March and April, transplanting in May, and harvesting tobacco leaves in September and October. According to this tobacco leaf growth law, the corresponding historical duration can be every day in the five months of May, June, July, August and September in a year, for example, 153 days. The historical climate data matrix can be a matrix constructed based on historical climate data. The historical climate data matrix can include historical climate data corresponding to multiple target climate factors. The target climate factor can be a climate factor that has a greater impact on the tobacco leaf growth process. In general, all climatic factors that affect tobacco growth can be found first, and further, climatic factors that have a greater impact on tobacco growth can be screened out from these climatic factors. These climatic factors may affect the chemical content of tobacco leaves, other tobacco quality indicators such as smoking quality, and these climatic factors can be used as target climatic factors. It should be noted that the introduction of climatic factors that have a smaller impact on tobacco growth may introduce too much interference, which may directly affect the prediction results of the chemical content of tobacco leaves. Therefore, screening out target climatic characteristics that have a greater impact on the tobacco growth process can improve the prediction accuracy of the chemical content of tobacco leaves. Optionally, multiple target climatic factors include daily maximum temperature, daily average speed, daily average atmospheric pressure, daily total precipitation, daily average net sunshine intensity, and daily average wind speed. It should be noted that the reason for selecting daily maximum temperature and daily average temperature is that the daily maximum temperature reflects the peak heat within a day, which has a significant impact on the physiological response of crops, and may affect the photosynthesis efficiency together with the daily average net sunshine intensity, thereby affecting the growth of crops and the accumulation of chemical components; the daily average temperature provides a comprehensive description of the heat throughout the day and is a good indicator for measuring the growth environment. The reason for choosing daily average atmospheric pressure is that different regions have unique climate characteristics due to different altitudes. There is a complex invisible connection between climate and altitude. In addition, since atmospheric pressure is a direct reflection of altitude characteristics, daily average atmospheric pressure is additionally introduced into the target climate factor to improve the richness of the data. In addition, the introduction of daily average atmospheric pressure as a climate factor to characterize the characteristics of different tobacco producing areas enhances the cross-regional applicability and generalization ability of the model.The reason for choosing the daily average wind speed is that for tobacco leaves, the wind speed will change the temperature gradient on the leaf surface, affect the water vapor exchange rate between the leaf and the air, stimulate the opening and closing of stomata, affect the absorption of carbon dioxide, and further affect the rate of transpiration and photosynthesis. Therefore, the daily average wind speed is additionally introduced into the target climate factor to improve the richness of the data. The historical climate data can be the climate data corresponding to the target climate factor in the target tobacco producing area over a historical period of time.

[0032] In this embodiment, the historical climate data matrix can be constructed based on the historical climate data corresponding to multiple target climate factors. Furthermore, the historical climate data corresponding to multiple target climate factors corresponding to the target tobacco producing area within the historical period can be first obtained. Further, the historical climate data matrix can be constructed based on the historical climate data corresponding to multiple target climate factors.

[0033] Optionally, determining a historical climate data matrix corresponding to a target tobacco producing area within a historical period includes: determining multiple target climate factors from multiple candidate climate factors according to a preset climate factor screening standard; obtaining historical climate data corresponding to the multiple target climate factors in the target tobacco producing area within a historical period; for multiple target climate factors, arranging the historical climate data corresponding to the target climate factors in chronological order to obtain a historical climate data sequence corresponding to the target climate factors; splicing the historical climate data sequences corresponding to the multiple target climate factors according to a preset sequence splicing standard to obtain a historical climate data matrix.

[0034] Among them, the climate factor screening criteria can be a screening condition for screening out the climate factors that have a greater impact on the tobacco leaf growth process. Exemplarily, the climate factor screening criteria can include the altitude of the target tobacco leaf production area, the longitude and latitude of the location and / or the characteristics of the climate factor itself. The candidate climate factor can be a climate factor to be selected. The candidate climate factor can be a climate factor that affects tobacco growth. Optionally, the candidate climate factor can include the daily maximum temperature, the daily minimum temperature, the daily average temperature, the accumulated temperature, the atmospheric pressure, the precipitation, the sunshine intensity and the wind speed. The historical climate data sequence can be a one-dimensional matrix composed of a row of historical climate data or a column of historical climate data. Exemplarily, it is assumed that the target climate factor is the daily maximum temperature, and the historical duration is 153 days. Then, the number of historical climate data corresponding to the daily maximum temperature is 153. In this case, the historical climate data sequence corresponding to the daily maximum temperature can be a column of historical climate data arranged in the order of the time of these 153 days by these 153 historical climate data, that is, a matrix with 153 rows and 1 columns.

[0035] As an optional implementation of this embodiment, multiple candidate climate factors can be determined. Further, multiple candidate climate factors can be screened according to preset climate factor screening criteria to determine multiple target climate factors from multiple candidate climate factors. Further, the historical duration associated with the tobacco production law of the target tobacco producing area can be determined from the duration before the current moment, and the historical climate data corresponding to multiple target climate factors in the target tobacco producing area within the historical duration can be obtained. Further, for multiple target climate factors, the historical climate data corresponding to the target climate factors can be arranged in chronological order of the historical duration, and the sequence obtained after the arrangement is used as the historical climate data sequence corresponding to the target climate factor. Further, the historical climate data sequences corresponding to multiple target climate factors are spliced ​​together to obtain a historical climate data matrix. Exemplarily, assuming that the number of target climate factors is 6, the historical duration is 153 days, and the historical climate data sequence corresponding to one target climate factor is a 153×1 matrix. Furthermore, the historical climate data matrix obtained by concatenating the historical climate data sequences corresponding to the six target climate factors is a 153×6 matrix, where 153 represents the number of days in the historical period and 6 represents the number of target climate factors.

[0036] It should be noted that the historical climate data series can be spliced ​​randomly, that is, the arrangement order of multiple target climate factors is not limited.

[0037] It should also be noted that the historical climate data structure based on chronological order has rich time series information, which can better fit the climate change relationship in the real environment. In the process of prediction using models that are good at capturing time series information, it can establish a more complete climate change relationship and improve the accuracy of prediction of the chemical composition content of tobacco leaves.

[0038] S120, processing the historical climate data matrix according to the chemical component content prediction model corresponding to the tobacco leaf grade to be predicted obtained in advance, and predicting the predicted chemical component data corresponding to the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period.

[0039] Among them, the chemical component content prediction model can take the historical climate data of the target tobacco leaf producing area as an input object, and use it as a neural network model to predict the chemical component data in the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area based on the input object. The chemical component content prediction model can be corresponding to the tobacco leaf grade to be predicted, that is, the chemical component content prediction model can only predict the chemical component content in the tobacco leaves of the corresponding tobacco leaf grade to be predicted. The chemical component prediction model is used to predict the chemical component data of the tobacco leaves of the tobacco leaf grade to be predicted produced in the tobacco leaf producing area within the historical length according to the historical climate data of the tobacco leaf producing area within the historical length. The tobacco leaf grade to be predicted can be the tobacco leaf grade for which the chemical component content of the tobacco leaf is to be predicted. Optionally, the tobacco leaf grade to be predicted may include B2F, C2F, C3F and X2F, etc. The chemical component content prediction model can be a deep learning model of any model structure, and the model structure of the chemical component content prediction model is not limited here, and can be customized according to actual needs. In this embodiment, the chemical component content prediction model may include a feature extraction module and a chemical component content prediction module. The feature extraction module may be a neural network model for extracting local features of input data. The feature extraction module may include any neural network model capable of extracting data features. Optionally, the feature extraction module includes a convolutional neural network. It is understood that convolutional neural networks are good at extracting local features of input data, especially in multidimensional time series data, and can capture important feature information. The chemical component content prediction module may be a neural network model for predicting the content of each chemical component contained in tobacco leaves. The chemical component content prediction module may include any neural network model capable of predicting the content of chemical components in tobacco leaves. Optionally, the chemical component content prediction module includes a long short-term memory network and a fully connected layer connected to the long short-term memory network. The long short-term memory network can retain information over a long period of time, thereby effectively processing long-term dependencies in time series data. By combining the advantages of the two neural networks, the convolutional neural network and the long short-term memory network, the chemical component content prediction model can capture the complex spatiotemporal interaction relationship in the climate factors, and provide a more accurate model for predicting the content of chemical components in tobacco leaves. The predicted chemical component data may be the content of chemical components contained in tobacco leaves predicted by the neural network model based on historical climate data. The predicted chemical component data includes the predicted content of multiple tobacco chemical components. The chemical composition of tobacco leaves generally refers to the various chemical substances contained in tobacco leaves, which have an important impact on the quality, taste and burning characteristics of tobacco leaves. Multiple tobacco leaf chemical components include total sugar, reducing sugar, nicotine, potassium and total nitrogen. It can be understood that total sugar refers to the sum of all soluble sugars in tobacco leaves, including monosaccharides (such as glucose, fructose), disaccharides (such as sucrose, maltose) and polysaccharides (such as starch). Reducing sugar refers to sugars with reducing properties, mainly monosaccharides and some disaccharides.Nicotine is a nitrogen-containing alkaline compound in tobacco leaves and the main component of tobacco alkaloids. Potassium is an important inorganic salt element in tobacco leaves, which plays a decisive role in the combustibility and fire-holding power of tobacco leaves. Total nitrogen refers to the sum of all nitrogen-containing compounds in tobacco leaves, including proteins, amino acids, amides, etc.

[0040] As an optional implementation of the first embodiment of this embodiment, when the historical climate data matrix corresponding to the target tobacco leaf producing area within the historical length of time is obtained, the tobacco leaf grade to be predicted can be determined, and the chemical component content prediction model corresponding to the tobacco leaf grade to be predicted can be retrieved from the chemical component content prediction model corresponding to at least one sample tobacco leaf grade obtained by pre-training. Further, the historical climate data matrix can be input into the retrieved chemical component content prediction model. Further, the historical climate data matrix can be processed in sequence based on the feature extraction module and the chemical component content prediction module in the chemical component content prediction model, and the predicted chemical component data corresponding to the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical length of time can be output.

[0041] In this embodiment, in order to clearly and intuitively observe the correlation between the climate data of the target climate factor and the component content of the chemical components of tobacco leaves, and provide a data basis for improving the quality of tobacco leaves, after obtaining the predicted chemical composition data, a correlation diagram can also be generated based on the historical climate data matrix and the predicted chemical composition data.

[0042] Optionally, after predicting the predicted chemical composition data corresponding to the tobacco of the tobacco grade to be predicted produced in the target tobacco producing area within the historical period, it also includes: generating a predicted chemical composition content curve corresponding to the tobacco of the tobacco grade to be predicted based on the predicted chemical composition data; and / or, generating a climate chemical composition content correlation diagram corresponding to the target tobacco producing area within the historical period based on the predicted chemical composition data and the historical climate data matrix.

[0043] Among them, the predicted chemical component content curve can be a curve that characterizes the distribution law of the predicted chemical component content. The predicted chemical component content curve can be a curve with time as the horizontal axis and the chemical components of each tobacco leaf as the vertical axis. The climate chemical component content correlation diagram can be an image that characterizes the correlation between the chemical components of tobacco leaves and climate data. The climate chemical component content correlation image can be any graph that can characterize the correlation between two types of data. Optionally, the graphic type of the climate chemical component content correlation diagram can be at least one of a scatter plot, a bubble plot, a heat map, a network diagram, and a parallel coordinate diagram.

[0044] As an optional implementation of this embodiment, after obtaining the predicted chemical composition data, data analysis is performed on it and converted from data form to graphic form to obtain a predicted chemical component content curve corresponding to the tobacco leaves of the tobacco leaf grade to be predicted. Further, data analysis can be performed on the predicted chemical composition data and the historical climate data matrix to generate a climate chemical component content correlation diagram corresponding to the target tobacco leaf producing area within the historical period.

[0045] It should be noted that the benefit of generating a predicted chemical component content curve is that the change trend of the predicted content of each tobacco leaf chemical component can be clearly and intuitively observed based on the predicted chemical component content curve. The benefit of generating a climate chemical component content correlation diagram is that the correlation between the tobacco leaf chemical components and climate data can be clearly and intuitively observed based on the climate chemical component content correlation diagram.

[0046] The technical solution of the embodiment of the present invention provides a data basis for the subsequent prediction of the content of chemical components in tobacco leaves by determining the historical climate data matrix corresponding to the target tobacco producing area within the historical period, wherein the historical climate data matrix includes historical climate data corresponding to multiple target climate factors. In addition, the constructed historical climate data matrix can facilitate the characterization of the correlation between climate changes and seasonal and date changes, which helps to improve the prediction accuracy of the content of chemical components in tobacco leaves. Furthermore, the historical climate data matrix is ​​processed according to the pre-trained prediction model of the chemical component content corresponding to the tobacco leaf grade to be predicted, and the predicted chemical component data corresponding to the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period are predicted. This solves the problem in the related technology that the chemical composition of the tobacco leaves cannot be predicted before the tobacco leaves are obtained, which leads to low efficiency and accuracy in the prediction of the chemical composition of the tobacco leaves, and wastes manpower and material costs. This achieves the effect of predicting the chemical component content contained in the tobacco leaves of the corresponding tobacco leaf grade produced by any tobacco leaf producing area without obtaining actual tobacco leaf samples, only based on the neural network model and the historical climate data on climate factors of the tobacco leaf producing area, which greatly improves the timeliness of the prediction of the chemical component content of the tobacco leaves, improves the prediction efficiency and prediction accuracy of the chemical component content of the tobacco leaves, and to a certain extent provides important decision-making support for the production of flue-cured tobacco, and has important practical significance and application prospects.

[0047] Embodiment 2

[0048] Figure 2It is a flowchart of a method for predicting the content of chemical components in tobacco leaves provided by the second embodiment of the present invention. On the basis of the above-mentioned embodiment, the process of determining the predicted chemical component data is further refined. Optionally, the historical climate data matrix is ​​processed according to the chemical component content prediction model corresponding to the tobacco grade to be predicted obtained by pre-training, and the predicted chemical component data corresponding to the tobacco of the tobacco grade to be predicted produced in the target tobacco producing area within the historical duration are predicted, including: calling the chemical component content prediction model corresponding to the tobacco grade to be predicted from at least one chemical component content prediction model obtained by pre-training; inputting the historical climate data matrix into the chemical component content prediction model, extracting features from the historical climate data matrix based on the feature extraction module in the chemical component content prediction model, and obtaining a climate feature matrix; performing chemical component prediction on the climate feature matrix based on the chemical component content prediction module, so as to obtain the predicted chemical component data corresponding to the tobacco of the tobacco grade to be predicted produced in the target tobacco producing area within the historical duration. Its specific implementation method can be referred to the technical solution of this embodiment. Among them, the technical terms that are the same or similar to the above-mentioned embodiments are not repeated here.

[0049] like Figure 2 As shown, the method includes:

[0050] S210. Determine a historical climate data matrix corresponding to a target tobacco producing area within a historical period; wherein the historical climate data matrix includes historical climate data corresponding to a plurality of target climate factors.

[0051] S220. Retrieve a chemical component content prediction model corresponding to the tobacco leaf grade to be predicted from at least one chemical component content prediction model obtained through pre-training.

[0052] In this embodiment, there are multiple implementations for retrieving the chemical component content prediction model corresponding to the tobacco leaf grade to be predicted. Optionally, the grade identifier of the tobacco leaf grade to be predicted is determined, and the chemical component content prediction model corresponding to the tobacco leaf grade to be predicted is retrieved based on the grade identifier. The grade identifier may be identification information for identifying the corresponding tobacco leaf grade. The grade identifier may be identification information in any form. Optionally, the grade identifier may be a grade name and / or a grade code, etc.

[0053] As an optional implementation of this embodiment, a chemical component content prediction model corresponding to at least one sample tobacco leaf grade can be pre-trained, and the grade identifier corresponding to the sample tobacco leaf grade is associated with the chemical component content prediction model corresponding to it and stored. Further, when determining the tobacco leaf grade to be predicted, the grade identifier corresponding to the tobacco leaf grade to be predicted can be determined. Further, the chemical component content prediction model corresponding to the tobacco leaf grade to be predicted can be retrieved from at least one pre-stored chemical component content prediction model based on the grade identifier.

[0054] S230, inputting the historical climate data matrix into the chemical component content prediction model, performing feature extraction on the historical climate data matrix based on the feature extraction module in the chemical component content prediction model, and obtaining a climate feature matrix.

[0055] Among them, the feature extraction module can be a neural network model for extracting features from input data. Optionally, the feature extraction module includes a convolutional neural network. The convolutional neural network is a type of feedforward neural network that includes convolution calculations and has a deep structure. It is one of the representative algorithms of deep learning. At the same time, the convolutional neural network has representation learning capabilities and can perform translation-invariant classification of input information according to its hierarchical structure. In this embodiment, the convolutional neural network may include at least one convolutional layer, and the convolution kernel size of each convolutional layer may be any value, optionally 1×6, 1×7 or 1×8. It should be noted that the convolution kernel size is associated with the number of target climate factors. Exemplarily, assuming that the number of target climate factors is 6, the convolution kernel size of the convolutional layer in the convolutional neural network can be set to 1×6. The advantage of this setting is that the convolutional neural network can be allowed to extract the historical climate data of the 6 target climate factors as one data feature. The climate feature matrix can be understood as the local data features of the historical climate data matrix. Compared with the global data features, the local data features can be the local expression of the data features, reflecting the important features on the historical climate data matrix. The characteristic data included in the climate characteristic matrix are important climate characteristics in the historical climate data matrix. It should be noted that the advantage of extracting characteristics from the historical climate data matrix is ​​that it can speed up the model calculation speed and improve the prediction accuracy of tobacco leaf chemical composition data.

[0056] As an optional implementation of this embodiment, after obtaining the historical climate data matrix, the historical climate data matrix can be input into the chemical component content prediction model. Further, the historical climate data matrix can be feature extracted based on the feature extraction module in the chemical component content prediction model to obtain a climate feature matrix.

[0057] For example, it is assumed that the historical climate data matrix is ​​a 153×6 matrix, the number of target climate factors is 6, and then the convolution kernel size of the convolution layer in the convolutional neural network in the feature extraction module is 1×6. Furthermore, the historical climate data matrix is ​​input into the feature extraction module, and the output climate feature matrix is ​​a 153×1 feature matrix.

[0058] S240, performing chemical composition prediction on the climate characteristic matrix based on the chemical composition content prediction module to obtain predicted chemical composition data corresponding to tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within a historical period of time.

[0059] Among them, the chemical component content prediction module can be a neural network model for predicting the content of each tobacco chemical component in tobacco leaves. The chemical component content prediction module may include a long short-term memory network and a fully connected layer connected to the long short-term memory network. Long Short-Term Memory (LSTM) can be used to capture the temporal dependencies of climate characteristics. Long short-term memory networks are usually able to understand the temporal dynamics of data well and make predictions based on time series data. The fully connected layer is used to map the learned features to specific output types or values ​​to complete tasks such as classification, regression, and numerical prediction. At the same time, the fully connected layer usually performs a nonlinear transformation on the results of the linear combination under the action of the activation function, introduces nonlinear properties, and enables the smoking quality prediction model to fit more complex nonlinear patterns. In addition, the fully connected layer can also be used to reduce the dimension of the data for subsequent processing.

[0060] As an optional implementation of the first embodiment of this invention, after obtaining the climate characteristic matrix, the climate characteristic matrix can be input into the chemical component content prediction module. Furthermore, the climate characteristic matrix can be processed based on the long short-term memory network in the chemical component content prediction module, and the characteristic values ​​corresponding to the chemical components of each tobacco leaf can be output, and the characteristic values ​​corresponding to the outputted multiple tobacco leaf chemical components can be used as predicted chemical component features. Furthermore, the predicted chemical component features can be input into the fully connected layer, and then, the predicted chemical component features can be transformed based on the fully connected layer to obtain the predicted chemical component data corresponding to the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period.

[0061] The technical solution of the embodiment of the present invention is to retrieve the chemical component content prediction model corresponding to the tobacco leaf grade to be predicted from at least one chemical component content prediction model obtained by pre-training; further, input the historical climate data matrix into the chemical component content prediction model, and extract features from the historical climate data matrix based on the feature extraction module in the chemical component content prediction model to obtain a climate feature matrix; further, perform chemical component prediction on the climate feature matrix based on the chemical component content prediction module to obtain the predicted chemical component data corresponding to the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period, thereby achieving the effect of predicting the chemical component content contained in the tobacco leaves of the tobacco leaf grade to be predicted produced in any tobacco leaf producing area based on the neural network model corresponding to the tobacco leaf grade to be predicted and the historical climate data on climate factors of the tobacco leaf producing area, and the deep learning model combining the convolutional neural network and the long short-term memory network can fully extract the time series characteristics of the climate data, capture the complex nonlinear relationship between the climate characteristics and the chemical composition of the tobacco leaves, and further improve the accuracy of the prediction of the chemical composition of the tobacco leaves.

[0062] Embodiment 3

[0063] Figure 3 This is a flowchart of a method for predicting the content of chemical components in tobacco leaves provided in Example 3 of the present invention. Based on the above-mentioned embodiments, before applying the chemical component content prediction model, a pre-built deep learning model can be trained based on training sample data to obtain a chemical component content prediction model. The specific implementation method can refer to the technical solution of this embodiment. Among them, the technical terms that are the same or similar to those in the above-mentioned embodiments are not repeated here.

[0064] like Figure 3 As shown, the method includes:

[0065] S310, training to obtain at least one chemical component content prediction model.

[0066] It should be noted that before applying the chemical component content prediction model provided in this embodiment, the pre-constructed deep learning model can be trained in a supervised or unsupervised manner. Before training the deep learning model, multiple training sample data can be constructed to train the model based on multiple training sample data. In order to improve the prediction accuracy of the chemical component content prediction model, as many and rich training samples as possible can be constructed.

[0067] Optionally, training to obtain at least one chemical component content prediction model includes: obtaining multiple training sample data corresponding to at least one sample tobacco leaf grade, wherein the training sample data include a sample climate data matrix corresponding to the sample tobacco leaf origin within a historical period and real chemical composition data of tobacco of the sample tobacco leaf grade produced in the sample tobacco leaf origin within a historical period; at least one sample tobacco leaf grade includes a tobacco leaf grade to be predicted; for at least one sample tobacco leaf grade, training a pre-constructed deep learning model based on the multiple training sample data corresponding to the sample tobacco leaf grade to obtain a chemical component content prediction model corresponding to the sample tobacco leaf grade.

[0068] The training sample data include the sample climate data matrix corresponding to the sample tobacco leaf producing area within the historical period and the real chemical composition data of the sample tobacco leaf grade produced in the sample tobacco leaf producing area within the historical period.

[0069] Among them, the origin of the sample tobacco leaves can be any area where tobacco leaves can be grown. The sample climate data matrix may include sample climate data corresponding to multiple target climate factors. The sample climate data corresponding to each target climate factor may be pre-stored climate data retrieved from a climate database, or may be simulated climate data generated based on a data generation model, etc. The sample tobacco leaf grade may be any tobacco leaf grade. Optionally, the sample tobacco leaf grades include B2F, C2F, C3F, and X2F, etc. At least one sample tobacco leaf grade includes a tobacco leaf grade to be predicted. In other words, the chemical component content prediction model corresponding to at least one sample tobacco leaf grade obtained by training may include a chemical component content prediction model corresponding to the tobacco leaf grade to be predicted. The real chemical composition data may include the real contents of multiple tobacco leaf chemical components.

[0070] In this embodiment, in order to construct a large number of training samples, at least one sample tobacco grade of tobacco produced in a plurality of sample tobacco producing areas within a historical period can be obtained. Further, the chemical component content of each sample tobacco grade produced in each sample tobacco producing area is detected, and the real chemical component data corresponding to the tobacco is obtained. And, the sample climate data matrix of multiple sample tobacco producing areas within a historical period can be obtained. Further, for at least one sample tobacco producing area, a plurality of training sample data corresponding to the sample tobacco producing area can be constructed based on the sample climate data matrix and the real chemical component data of the multiple sample tobacco producing areas. Further, for at least one sample tobacco producing area, a pre-constructed deep learning model can be trained based on the multiple training sample data corresponding to the sample tobacco producing area to obtain a trained deep learning model. Further, a chemical component content prediction model corresponding to the sample tobacco producing area can be determined based on the trained deep learning model.

[0071] Optionally, a pre-constructed deep learning model is trained based on multiple training sample data corresponding to the sample tobacco leaf grade to obtain a chemical component content prediction model corresponding to the sample tobacco leaf grade, including: for multiple training sample data corresponding to the sample tobacco leaf grade, the sample climate data matrix in the training sample data is input into the pre-constructed deep learning model to obtain chemical component prediction data corresponding to the training sample data; based on the chemical component prediction data and the actual chemical component data in the training sample data, a loss value is determined, model parameters in the deep learning model are corrected based on the loss value, and the convergence of the loss function in the deep learning model is used as a training target to obtain at least one model to be tested; at least one model to be tested is tested based on the test sample data, and the model to be tested whose test results meet preset conditions is used as the chemical component content prediction model corresponding to the sample tobacco leaf grade.

[0072] Among them, the pre-built deep learning model can be a neural network model with model parameters as default values ​​or initial values. The chemical composition prediction data can be the predicted chemical composition data output after processing the sample climate data matrix through the deep learning model. The loss value can be a numerical value that characterizes the degree of difference between the model output and the true output. The loss function can be a function determined based on the loss value and used to characterize the degree of difference between the predicted output and the actual output. The model to be tested can be a model obtained by correcting the model parameters in the deep learning model according to the loss value. The model to be tested can be a model to be evaluated for model performance. The number of models to be tested can be consistent with the number of training iterations (or the number of model parameter corrections). The test sample data can be part of the sample data in the training sample data, or it can be sample data reconstructed based on the construction method of the training sample data. The preset conditions can include at least one of the mean square error (MSE) reaching the first error threshold, the mean absolute error (MAE) reaching the second error threshold, and the mean absolute percentage error (MAPE) reaching the third error threshold. Among them, mean square error is usually used to measure the average square difference between the predicted value and the true value. Mean absolute error is usually used to measure the average absolute difference between the predicted value and the true value. Mean absolute percentage error is usually used to measure the average absolute percentage of the predicted value deviating from the true value.

[0073] As an optional implementation of this embodiment, after obtaining a plurality of training sample data corresponding to at least one sample tobacco leaf grade, for a plurality of training sample data corresponding to the sample tobacco leaf grade, the sample climate data matrix in the training sample data can be input into a pre-built deep learning model to process the sample climate data matrix based on the deep learning model, and output the chemical composition prediction data corresponding to the training sample data. Further, the chemical composition prediction data can be compared with the real chemical composition data in the training sample data to obtain a loss value. Further, in the case of obtaining a loss value, the model parameters in the deep learning model can be corrected based on the loss value, the corrected model can be used as the first model to be processed, and the first model to be processed can be backed up, and the backup model can be saved. Further, the process of processing the training sample data and determining the loss value can be repeated. Afterwards, the model parameters in the model to be processed can be corrected based on the obtained loss value, the corrected model can be used as the second model to be processed, and the second model to be processed can be backed up, and the backup model can be saved. Furthermore, when it is detected that the loss function has converged, such as when the training error of the loss function is less than the preset error, or when the error change trend tends to be stable, the iterative training can be stopped at this time, and the model obtained at this time can be used as the model to be tested. Furthermore, the saved backup model can be retrieved, and both the retrieved backup model and the model to be tested obtained at this time can be used as the model to be tested to obtain at least one model to be tested. Furthermore, in order to evaluate the model performance of each model to be tested, test sample data can be obtained, and for at least one model to be tested, the model to be tested can be tested and verified based on the test sample data, and a test result corresponding to the model to be tested can be obtained. Furthermore, it can be determined whether the test result of at least one model to be tested meets the preset conditions. Furthermore, the model to be tested whose test result meets the preset conditions can be used as a chemical component content prediction model corresponding to the grade of the sample tobacco leaves.

[0074] S320, determining a historical climate data matrix corresponding to a target tobacco leaf producing area within a historical period; wherein the historical climate data matrix includes historical climate data corresponding to a plurality of target climate factors.

[0075] S330, processing the historical climate data matrix according to the chemical component content prediction model corresponding to the tobacco leaf production area to be predicted obtained in advance, and predicting the predicted chemical component data corresponding to the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf production area within the historical period.

[0076] The technical solution of the embodiment of the present invention is achieved by acquiring multiple training sample data corresponding to at least one sample tobacco leaf grade, wherein the training sample data include a sample climate data matrix corresponding to the sample tobacco leaf origin within a historical period and real chemical composition data of tobacco of the sample tobacco leaf grade produced in the sample tobacco leaf origin within a historical period; at least one sample tobacco leaf grade includes a tobacco leaf grade to be predicted; further, for at least one sample tobacco leaf grade, a pre-constructed deep learning model is trained based on multiple training sample data corresponding to the sample tobacco leaf grade to obtain a chemical component content prediction model corresponding to the sample tobacco leaf grade, thereby achieving the effect of training a chemical component content prediction model that can accurately predict the chemical component content of tobacco leaves of different sample tobacco leaf grades based on the constructed training sample data, effectively improving the accuracy and robustness of the model, and effectively improving the model performance.

[0077] Embodiment 4

[0078] Figure 4 Schematic diagram of a device for predicting the content of chemical components in tobacco leaves provided by the fourth embodiment of the present invention. Figure 4 As shown, the device includes: a climate data matrix determination module 410 and a chemical composition prediction module 420.

[0079] Among them, the climate data matrix determination module 410 is used to determine the historical climate data matrix corresponding to the target tobacco producing area within the historical time period; wherein the historical time period at least includes the time interval associated with the tobacco production law of the target tobacco producing area; the historical climate data matrix includes historical climate data corresponding to multiple target climate factors; the chemical composition prediction module 420 is used to process the historical climate data matrix according to the pre-trained chemical composition content prediction model corresponding to the tobacco grade to be predicted, and predict the predicted chemical composition data corresponding to the tobacco of the tobacco grade to be predicted produced by the target tobacco producing area within the historical time period; wherein the chemical composition content prediction model is used to predict the chemical composition data of the tobacco of the tobacco grade to be predicted produced by the tobacco producing area within the historical time period based on the historical climate data of the tobacco producing area within the historical time period; the chemical composition content prediction model is obtained by training a pre-constructed deep learning model based on the sample climate data matrix of the sample tobacco producing area within the historical time period and the real chemical composition data of the tobacco of the tobacco grade to be predicted produced by the sample tobacco producing area within the historical time period.

[0080] The technical solution of the embodiment of the present invention provides a data basis for the subsequent prediction of the content of chemical components in tobacco leaves by determining the historical climate data matrix corresponding to the target tobacco producing area within the historical period, wherein the historical climate data matrix includes historical climate data corresponding to multiple target climate factors. In addition, the constructed historical climate data matrix can facilitate the characterization of the correlation between climate changes and seasonal and date changes, which helps to improve the prediction accuracy of the content of chemical components in tobacco leaves. Furthermore, the historical climate data matrix is ​​processed according to the pre-trained prediction model of the chemical component content corresponding to the tobacco leaf grade to be predicted, and the predicted chemical component data corresponding to the tobacco leaves of the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period are predicted. This solves the problem in the related technology that the chemical composition of the tobacco leaves cannot be predicted before the tobacco leaves are obtained, which leads to low efficiency and accuracy in the prediction of the chemical composition of the tobacco leaves, and wastes manpower and material costs. This achieves the effect of predicting the chemical component content contained in the tobacco leaves of the corresponding tobacco leaf grade produced by any tobacco leaf producing area without obtaining actual tobacco leaf samples, only based on the neural network model and the historical climate data on climate factors of the tobacco leaf producing area, which greatly improves the timeliness of the prediction of the chemical component content of the tobacco leaves, improves the prediction efficiency and prediction accuracy of the chemical component content of the tobacco leaves, and to a certain extent provides important decision-making support for the production of flue-cured tobacco, and has important practical significance and application prospects.

[0081] Optionally, the multiple target climate factors include daily maximum temperature, daily average speed, daily average atmospheric pressure, daily total precipitation, daily average net sunshine intensity and daily average wind speed; the predicted chemical composition data include predicted contents of multiple tobacco leaf chemical components, and the multiple tobacco leaf chemical components include total sugar, reducing sugar, nicotine, potassium and total nitrogen.

[0082] Optionally, the chemical composition content prediction model includes a feature extraction module and a chemical composition content prediction module; the chemical composition prediction module 420 includes: a model retrieval unit, a feature extraction unit and a chemical composition prediction unit.

[0083] A model retrieval unit, used to retrieve a chemical component content prediction model corresponding to the tobacco leaf grade to be predicted;

[0084] A feature extraction unit, used for inputting the historical climate data matrix into the chemical component content prediction model, performing feature extraction on the historical climate data matrix based on the feature extraction module in the chemical component content prediction model, and obtaining a climate feature matrix;

[0085] A chemical composition prediction unit is used to perform chemical composition prediction on the climate characteristic matrix based on the chemical composition content prediction module to obtain predicted chemical composition data corresponding to the tobacco of the tobacco grade to be predicted produced in the target tobacco producing area within the historical period.

[0086] Optionally, the feature extraction module includes a convolutional neural network; the chemical component content prediction module includes a long short-term memory network and a fully connected layer connected to the long short-term memory network.

[0087] Optionally, the climate data matrix determination module 410 includes: a climate factor screening unit, a climate data acquisition unit, a climate data sequence construction unit and a climate data matrix determination unit.

[0088] A climate factor screening unit, used to determine a plurality of target climate factors from a plurality of candidate climate factors according to a preset climate factor screening standard;

[0089] A climate data acquisition unit, used to acquire historical climate data corresponding to the target climate factors in a target tobacco producing area within a historical period;

[0090] A climate data sequence construction unit is used to arrange the historical climate data corresponding to the target climate factors in chronological order for a plurality of the target climate factors, so as to obtain a historical climate data sequence corresponding to the target climate factors;

[0091] The climate data matrix determination unit is used to splice the historical climate data sequences corresponding to the plurality of target climate factors according to a preset sequence splicing standard to obtain a historical climate data matrix.

[0092] Optionally, the device further includes: a curve generating module and / or a correlation graph generating module.

[0093] a curve generating module, for generating a predicted chemical component content curve corresponding to the tobacco leaf of the tobacco leaf grade to be predicted according to the predicted chemical component data; and / or,

[0094] The association diagram generation module is used to generate a climate chemical component content association diagram corresponding to the target tobacco leaf producing area within a historical period based on the predicted chemical component data and the historical climate data matrix.

[0095] Optionally, the device also includes: a model training module.

[0096] Model training module, used to train and obtain a chemical component content prediction model;

[0097] The model training module includes: a training sample data acquisition unit and a model training unit.

[0098] A training sample data acquisition unit is used to acquire a plurality of training sample data corresponding to at least one sample tobacco leaf grade, wherein the training sample data includes a sample climate data matrix corresponding to a sample tobacco leaf producing area within a historical period and real chemical composition data of tobacco leaves of the sample tobacco leaf grade produced in the sample tobacco leaf producing area within the historical period; at least one sample tobacco leaf grade includes a tobacco leaf grade to be predicted;

[0099] A model training unit is used to train a pre-constructed deep learning model for at least one sample tobacco leaf grade based on a plurality of training sample data corresponding to the sample tobacco leaf grade, so as to obtain a chemical component content prediction model corresponding to the sample tobacco leaf grade.

[0100] Optionally, the model training unit includes: a prediction data determination subunit, a to-be-tested model determination subunit and a model determination unit.

[0101] A prediction data determination subunit is used for inputting the sample climate data matrix in the training sample data into a pre-built deep learning model for a plurality of the training sample data corresponding to the sample tobacco leaf grade, so as to obtain chemical composition prediction data corresponding to the training sample data;

[0102] A model to be tested determination subunit is used to determine a loss value according to the chemical composition prediction data and the real chemical composition data in the training sample data, and to modify the model parameters in the deep learning model according to the loss value, and to take the convergence of the loss function in the deep learning model as a training target to obtain at least one model to be tested;

[0103] The model determination unit is used to test at least one of the models to be tested based on the test sample data, and use the model to be tested whose test results meet preset conditions as a chemical component content prediction model corresponding to the sample tobacco grade.

[0104] The tobacco leaf chemical component content prediction device provided in the embodiment of the present invention can execute the tobacco leaf chemical component content prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0105] Embodiment 5

[0106] Figure 5A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0107] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0108] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0109] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a method for predicting the content of chemical components in tobacco leaves.

[0110] In some embodiments, the method for predicting the content of chemical components in tobacco leaves may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for predicting the content of chemical components in tobacco leaves described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the method for predicting the content of chemical components in tobacco leaves in any other appropriate manner (e.g., by means of firmware).

[0111] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0112] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0113] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0114] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0115] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a target blockchain network, and the Internet.

[0116] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0117] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0118] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting the content of chemical components in tobacco leaves, characterized in that: include: Determine a historical climate data matrix corresponding to a target tobacco leaf producing area within a historical period; wherein the historical period at least includes a time interval associated with the tobacco leaf production law of the target tobacco leaf producing area; the historical climate data matrix includes historical climate data corresponding to a plurality of target climate factors; The historical climate data matrix is ​​processed according to a chemical component content prediction model corresponding to the tobacco leaf grade to be predicted obtained by pre-training, and the predicted chemical component data corresponding to the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period are predicted; Among them, the chemical component content prediction model is used to predict the chemical composition data of tobacco leaves of the grade to be predicted produced in the tobacco producing area within the historical period according to the historical climate data of the tobacco producing area within the historical period; the chemical component content prediction model is obtained by training a pre-constructed deep learning model based on the sample climate data matrix of the sample tobacco producing area within the historical period and the real chemical composition data of tobacco leaves of the grade to be predicted produced in the sample tobacco producing area within the historical period.

2. The method for predicting the content of chemical components in tobacco leaves according to claim 1, characterized in that: The multiple target climate factors include daily maximum temperature, daily average speed, daily average atmospheric pressure, daily total precipitation, daily average net sunshine intensity and daily average wind speed; the predicted chemical composition data include predicted contents of multiple tobacco leaf chemical components, and the multiple tobacco leaf chemical components include total sugar, reducing sugar, nicotine, potassium and total nitrogen.

3. The method for predicting the content of chemical components in tobacco leaves according to claim 1, characterized in that: The chemical component content prediction model includes a feature extraction module and a chemical component content prediction module; the chemical component content prediction model corresponding to the tobacco leaf grade to be predicted obtained by pre-training processes the historical climate data matrix to predict the predicted chemical component data corresponding to the tobacco leaf grade to be predicted produced in the target tobacco leaf producing area within the historical period, including: Retrieving a chemical component content prediction model corresponding to the tobacco leaf grade to be predicted from at least one chemical component content prediction model obtained by pre-training; Inputting the historical climate data matrix into the chemical component content prediction model, performing feature extraction on the historical climate data matrix based on the feature extraction module in the chemical component content prediction model, and obtaining a climate feature matrix; The chemical composition prediction module is used to predict the chemical composition of the climate characteristic matrix, so as to obtain the predicted chemical composition data corresponding to the tobacco leaves of the tobacco grade to be predicted produced in the target tobacco producing area within the historical period.

4. The method for predicting the content of chemical components in tobacco leaves according to claim 3, characterized in that: The feature extraction module includes a convolutional neural network; the chemical component content prediction module includes a long short-term memory network and a fully connected layer connected to the long short-term memory network.

5. The method for predicting the content of chemical components in tobacco leaves according to claim 1, characterized in that: The historical climate data matrix corresponding to the target tobacco leaf producing area within the historical period is determined, including: Determine multiple target climate factors from multiple candidate climate factors according to preset climate factor screening criteria; Acquire historical climate data corresponding to the target climate factors in a target tobacco producing area within a historical period; For a plurality of the target climate factors, the historical climate data corresponding to the target climate factors are arranged in chronological order to obtain a historical climate data sequence corresponding to the target climate factors; The historical climate data sequences corresponding to the plurality of target climate factors are spliced ​​together according to a preset sequence splicing standard to obtain a historical climate data matrix.

6. The method for predicting the content of chemical components in tobacco leaves according to claim 1, characterized in that: Also includes: generating a predicted chemical component content curve corresponding to the tobacco leaves of the tobacco leaf grade to be predicted according to the predicted chemical component data; and / or, Based on the predicted chemical composition data and the historical climate data matrix, a correlation diagram of the climate chemical composition content corresponding to the target tobacco leaf producing area within a historical period is generated.

7. The method for predicting the content of chemical components in tobacco leaves according to claim 1, characterized in that: Also includes: Training and obtaining at least one chemical component content prediction model; The training obtains at least one chemical component content prediction model, including: Acquire a plurality of training sample data corresponding to at least one sample tobacco leaf grade, wherein the training sample data include a sample climate data matrix corresponding to a sample tobacco leaf producing area within a historical period and real chemical composition data of tobacco leaves of the sample tobacco leaf grade produced in the sample tobacco leaf producing area within the historical period; at least one sample tobacco leaf grade includes a tobacco leaf grade to be predicted; For at least one sample tobacco leaf grade, a pre-constructed deep learning model is trained based on a plurality of the training sample data corresponding to the sample tobacco leaf grade to obtain a chemical component content prediction model corresponding to the sample tobacco leaf grade.

8. The method for predicting the content of chemical components in tobacco leaves according to claim 7, characterized in that: The pre-built deep learning model is trained based on the plurality of training sample data corresponding to the sample tobacco leaf grade to obtain a chemical component content prediction model corresponding to the sample tobacco leaf grade, including: For the plurality of training sample data corresponding to the sample tobacco leaf grades, the sample climate data matrix in the training sample data is input into a pre-built deep learning model to obtain chemical composition prediction data corresponding to the training sample data; Determine a loss value according to the chemical composition prediction data and the actual chemical composition data in the training sample data, modify the model parameters in the deep learning model according to the loss value, take the convergence of the loss function in the deep learning model as the training goal, and obtain at least one model to be tested; At least one of the models to be tested is tested based on the test sample data, and the model to be tested whose test result meets the preset conditions is used as a chemical component content prediction model corresponding to the sample tobacco leaf grade.

9. A device for predicting the content of chemical components in tobacco leaves, characterized in that: include: A climate data matrix determination module is used to determine the historical climate data matrix corresponding to the target tobacco producing area within a historical period; wherein the historical period at least includes a time interval associated with the tobacco production law of the target tobacco producing area; the historical climate data matrix includes historical climate data corresponding to multiple target climate factors; A chemical composition prediction module is used to process the historical climate data matrix according to a pre-trained chemical composition content prediction model corresponding to the tobacco grade to be predicted, and predict the predicted chemical composition data corresponding to the tobacco of the tobacco grade to be predicted produced in the target tobacco producing area within the historical period; wherein the chemical composition content prediction model is used to predict the chemical composition data of the tobacco of the tobacco grade to be predicted produced in the tobacco producing area within the historical period according to the historical climate data of the tobacco producing area within the historical period; the chemical composition content prediction model is obtained by training a pre-constructed deep learning model based on the sample climate data matrix of the sample tobacco producing area within the historical period and the real chemical composition data of the tobacco of the tobacco grade to be predicted produced in the sample tobacco producing area within the historical period.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for predicting the content of chemical components in tobacco leaves according to any one of claims 1-8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for predicting the content of chemical components in tobacco leaves according to any one of claims 1 to 8 when executed.

12. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the method for predicting the content of chemical components in tobacco leaves according to claims 1-8.

Images