Nutrition diagnosis system and method based on tobacco leaves

Abstract

The invention relates to the technical field of tobacco nutrition diagnosis, and particularly discloses a tobacco-based nutrition diagnosis system, which comprises a data acquisition module, a feature extraction module, an environment calibration module, an identification module, a judgment module and a data storage module. The method comprises the following steps: acquiring growth parameters, illumination intensity and leaf reflection spectrum data of tobacco leaves in a target area through a data acquisition module; the feature extraction module extracts spectral feature parameters in the leaf reflection spectrum data; the environment calibration module is used for calibrating spectral characteristic parameters, so that the influence of low illumination intensity on spectral data is eliminated, and the data accuracy is improved; the identification module identifies the growth periods of the tobacco leaves, and the judgment module adopts different model algorithms for the tobacco leaves in different growth periods, so that the accuracy of the diagnosis result is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of tobacco leaf nutritional diagnosis, and in particular to a tobacco leaf-based nutritional diagnosis system and method. Background Art

[0002] Tobacco is one of my country's important cash crops, widely cultivated across provinces and regions in northern and southern my country. In tobacco production, the soil environment has a significant impact on its growth, development, yield, and quality. Soil with an appropriate organic matter content and rich in phosphorus, potassium, and trace elements is crucial for producing high-quality tobacco. Low organic matter content and soil fertility can lead to nutrient deficiencies during tobacco growth, resulting in weak growth, short plants, small, thin leaves, and poor yield and quality. High soil fertility can produce thick leaves with coarse veins, increased content of nitrogenous compounds such as protein and nicotine, and poor quality. Therefore, precise fertilization during tobacco cultivation, tailored to soil fertility and tobacco nutritional patterns, can coordinate tobacco plant nutrition, directly reducing fertilizer costs, indirectly improving fertilizer utilization and increasing farmers' incomes, while also addressing a range of environmental issues associated with traditional fertilization methods.

[0003] Crop nutrition diagnosis is a crucial technical support for precision fertilization. Crop nutrition diagnosis involves scientifically analyzing crop nutritional status to assess crop growth and nutrient availability. It serves as a fundamental prerequisite for scientific fertilization and a crucial means of regulating the exchange of nutrients and energy, such as nitrogen, phosphorus, and potassium, between crops and the soil. Plant nutrition diagnosis technology has evolved through three phases: empirical diagnosis, chemical testing, and physical methods. Traditional empirical methods offer advantages such as simplicity and speed, but are inherently difficult to replicate, require extensive practical experience, and have a high rate of false positives. Chemical diagnosis, while advantageous in terms of accuracy, is associated with high testing costs, cumbersome procedures, delayed results, and the need for disruptive sampling, which are key drawbacks. Physical diagnosis, as an emerging diagnostic method, offers advantages such as speed and the absence of disruptive sampling. However, existing physical diagnostic systems can be severely impacted by environmental factors and the varying growth stages of tobacco leaves, significantly impacting accuracy.

[0004] Therefore, it is necessary to design a tobacco leaf-based nutritional diagnosis system to solve the problem that the accuracy of the existing physical diagnosis system is seriously affected by environmental factors and the different growth periods of tobacco leaves. Summary of the Invention

[0005] In view of this, the present invention proposes a tobacco leaf-based nutritional diagnosis system to solve the problem that the accuracy of the existing physical diagnosis system is seriously affected by environmental factors and the different growth periods of tobacco leaves.

[0006] In one aspect, the present invention provides a tobacco leaf-based nutritional diagnosis system, comprising:

[0007] The data acquisition module is used to collect the growth parameters and light intensity of tobacco leaves in the target area and obtain leaf reflectance spectrum data through a multispectral imaging device;

[0008] a feature extraction module for performing spectral reflectance analysis on the growth parameters and reflectance spectrum data collected by the data acquisition module, separating the spectral response curves of the visible light band and the near-infrared band, and extracting the spectral characteristic parameters of the leaves at this time;

[0009] An environmental calibration module, configured to input the light intensity collected by the data acquisition module and the spectral characteristic parameters extracted by the feature extraction module into a preset reflectance compensation model to eliminate the influence of different light intensities on the spectral characteristic parameters;

[0010] an identification module for determining the tobacco growth period in the target area based on the growth parameters collected by the data collection module;

[0011] a diagnosis module, configured to obtain the nitrogen content, potassium content, and phosphorus content of the tobacco leaves based on the tobacco leaf growth period determined by the recognition module and the corresponding spectral characteristic parameters of the tobacco leaves;

[0012] The data storage module is used to store historical data.

[0013] Furthermore, the data acquisition module is configured to: preset a standard blade deployment angle B2 and acquire the blade deployment angle B1 in real time;

[0014] When B1 < B2, adjust the shooting angle of the multispectral imaging device to be perpendicular to the main vein direction; when B1 ≥ B2, keep the shooting angle in the canopy plane looking down direction;

[0015] After each adjustment, the reflectance spectrum data of three adjacent leaves are collected, the samples with the maximum and minimum reflectance values are eliminated, and the average reflectance of the remaining data is calculated as the effective spectral data input value.

[0016] Furthermore, the feature extraction module is configured to: preset a coefficient K, and extract the visible light band reflectance fluctuation amplitude C1 and the near infrared band reflectance fluctuation amplitude C2 when separating the spectral response curves of the visible light band and the near infrared band;

[0017] When C1>C2, the visible light band data is compensated and the processed spectral response curve is output, where the compensation amount = (C1-C2)×K;

[0018] When C1≤C2, the original spectral response curve is directly output.

[0019] Furthermore, the environmental correction module is preset with a first reflectivity compensation model R1=R×ω1 and a second reflectivity supplement model R1=R×ω2, where R is the collected spectral reflectivity, R1 is the corrected spectral reflectivity, ω1 is the first weight coefficient, and ω2 is the second weight coefficient.

[0020] Furthermore, the Wherein, D2 is the real-time light intensity collected by the data collection unit, D1 is the historical average light intensity of the same period stored by the data storage unit, K1 is a preset first proportional coefficient, K2 is a second proportional coefficient, and K2<K1;

[0021] The environmental calibration module is configured to:

[0022] When D1<0.6×D2, the first reflectance compensation model is used to compensate and calibrate the input spectral characteristic parameters;

[0023] When 0.6×D2≤D1<D2, the second reflectance compensation model is used to compensate and calibrate the input spectral characteristic parameters;

[0024] When D1≥D2, no compensation calibration is performed on the input spectral characteristic parameters.

[0025] Furthermore, the growth parameters of the tobacco leaves include plant height, number of leaves and leaf color RGB value; the recognition module is preset with plant height matching score models, leaf number matching score models and leaf color RGB value matching score models for different growth periods.

[0026] Furthermore, the identification module is configured to: input the plant height, number of leaves and leaf color RGB value information into the corresponding matching score model respectively, obtain the matching scores of the plant height, number of leaves and leaf color RGB values with all growth periods, perform weighted calculation on the scores, obtain the comprehensive scores of tobacco leaves for different growth periods, and identify the growth period with the highest score as the growth period of tobacco leaves.

[0027] Furthermore, the diagnosis module is configured to: preset a mapping relationship between the nitrogen content, potassium content and phosphorus content in leaves of different growth periods and the spectral characteristic parameters, match the corresponding mapping relationship according to the growth period of the tobacco leaves identified in the identification module, and obtain the corresponding real-time nitrogen content, potassium content and phosphorus content according to the corresponding spectral characteristic parameters of the tobacco leaves.

[0028] Furthermore, the diagnostic module is further configured to: preset standard value ranges and optimal values of nitrogen content, potassium content, and phosphorus content in tobacco leaves corresponding to the growth period, and compare the real-time nitrogen content, potassium content, and phosphorus content with the standard value ranges;

[0029] When any one of the real-time nitrogen content, potassium content and phosphorus content is outside the standard value range, the tobacco leaf is judged to be in an abnormal nutritional state, and a fertilization strategy is generated based on the difference between the real-time nitrogen content, potassium content and phosphorus content and the optimal value.

[0030] On the other hand, the present invention also provides a method for nutritional diagnosis based on tobacco leaves, comprising the following steps:

[0031] Collect the growth parameters and light intensity of tobacco leaves in the target area, and obtain leaf reflectance spectrum data through a multispectral imaging device;

[0032] Performing spectral reflectance analysis on the leaf reflectance spectrum data, separating the spectral response curves of the visible light band and the near-infrared band, and extracting the spectral characteristic parameters of the leaf at this time;

[0033] Calibrate the spectral characteristic parameters according to the collected light intensity data using a preset reflectance compensation model to obtain calibrated spectral characteristic parameters;

[0034] The growth period of tobacco leaves in the target area is determined based on the growth parameters of the leaves, and the nitrogen content, potassium content and phosphorus content of the tobacco leaves are obtained based on the growth period of the tobacco leaves and the spectral characteristic parameters.

[0035] Compared with the prior art, the beneficial effects of the present invention are: by setting an environmental calibration module to compensate for the spectral characteristic parameters collected in a low-light environment, the influence of low light intensity on the spectral characteristic parameters is eliminated, thereby greatly improving the judgment accuracy of the present application; at the same time, the tobacco leaf growing period is identified through the recognition module, and the judgment module automatically matches the corresponding model for tobacco leaves in different growing periods, further improving the accuracy of its own judgment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0037] Figure 1 This is a functional block diagram of a tobacco-based nutritional diagnosis system provided by an embodiment of the present invention.

[0038] Figure 2 This is a flow chart of a tobacco leaf-based nutritional diagnosis method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0040] Reference Figure 1 As shown, in some embodiments of the present application, a tobacco leaf-based nutritional diagnosis system includes:

[0041] The data acquisition module is used to collect the growth parameters and light intensity of tobacco leaves in the target area and obtain leaf reflectance spectrum data through a multispectral imaging device;

[0042] a feature extraction module for performing spectral reflectance analysis on the growth parameters and reflectance spectrum data collected by the data acquisition module, separating the spectral response curves of the visible light band and the near-infrared band, and extracting the spectral characteristic parameters of the leaves at this time;

[0043] An environmental calibration module, configured to input the light intensity collected by the data acquisition module and the spectral characteristic parameters extracted by the feature extraction module into a preset reflectance compensation model to eliminate the influence of different light intensities on the spectral characteristic parameters;

[0044] an identification module for determining the tobacco growth period in the target area based on the growth parameters collected by the data collection module;

[0045] a diagnosis module, configured to obtain the nitrogen content, potassium content, and phosphorus content of the tobacco leaves based on the tobacco leaf growth period determined by the recognition module and the corresponding spectral characteristic parameters of the tobacco leaves;

[0046] The data storage module is used to store historical data.

[0047] Specifically, the data acquisition module acquires images of tobacco plants in the target area through an image acquisition device, thereby obtaining growth parameters of the tobacco leaves; and collects light intensity information in real time by setting up multiple light intensity sensors;

[0048] It can be understood that the setting of the environmental calibration module can compensate for the spectral characteristic parameters collected in a low-light environment, eliminate the influence of low light intensity on the spectral characteristic parameters, and greatly improve the judgment accuracy of this application; at the same time, the recognition module can identify the growing period of tobacco leaves, and the judgment module automatically matches the corresponding model for tobacco leaves in different growing periods, further improving the accuracy of its own judgment results.

[0049] like Figure 1 As shown, in some embodiments of the present application, the data acquisition module is configured to: preset a standard blade deployment angle B2 and acquire the blade deployment angle B1 in real time;

[0050] When B1 < B2, adjust the shooting angle of the multispectral imaging device to be perpendicular to the main vein direction; when B1 ≥ B2, keep the shooting angle in the canopy plane looking down direction;

[0051] After each adjustment, the reflectance spectrum data of three adjacent leaves are collected, the samples with the maximum and minimum reflectance values are eliminated, and the average reflectance of the remaining data is calculated as the effective spectral data input value.

[0052] It can be understood that when B1 is less than B2, the shooting angle is adjusted to be perpendicular to the main vein direction. This can ensure that when the leaves are relatively upright, the multispectral imaging device can more accurately obtain the reflected spectrum data of the leaves, avoid uneven light reflection or shadow influence caused by angle problems, and thus improve the accuracy of the data; when B1 ≥ B2, the shooting angle is kept in the direction of looking down at the canopy plane, which is suitable for the situation where the leaves are relatively flat, and can fully obtain the spectral information of the leaf group, which also helps to ensure data quality.

[0053] It is understood that after collecting reflectance spectral data from three adjacent leaves, the samples with the maximum and minimum reflectance values are removed, and the average reflectance of the remaining data is calculated as the valid spectral data input value. This method can effectively remove abnormal data that may be caused by factors such as local leaf disease, impurities, and light interference. The final valid spectral data input is more representative of the true reflectance characteristics of the tobacco leaves, further improving the accuracy and reliability of the data.

[0054] like Figure 1 As shown, in some embodiments of the present application, the feature extraction module is configured to: preset a coefficient K, and extract the visible light band reflectance fluctuation amplitude C1 and the near infrared band reflectance fluctuation amplitude C2 when separating the spectral response curves of the visible light band and the near infrared band;

[0055] When C1>C2, the visible light band data is compensated and the processed spectral response curve is output, where the compensation amount = (C1-C2)×K;

[0056] When C1≤C2, the original spectral response curve is directly output.

[0057] It is understandable that in the process of tobacco leaf nutritional diagnosis, the spectral response curves of the visible light band and the near-infrared band are crucial for determining the nutritional status of tobacco leaves. However, due to various reasons such as environmental factors, measurement errors, and the characteristics of the tobacco leaves themselves, the reflectivity fluctuation amplitudes of the two bands may differ. When the visible light band reflectivity fluctuation amplitude C1 is greater than the near-infrared band reflectivity fluctuation amplitude C2, it means that the visible light band data is relatively more disturbed or has a larger deviation. In order to make the data of the two bands more comparable, stable, and accurate, so as to perform nutritional diagnosis more accurately, the visible light band data needs to be processed in a targeted manner. Therefore, the compensation amount is quantitatively controlled by the preset coefficient K, and compensation is performed according to the difference in the fluctuation amplitudes of the two. When C1≤C2, it means that the visible light band data is relatively stable and no additional processing is required. The original spectral response curve is directly output.

[0058] It is understandable that this solution can dynamically adjust the stability and consistency of the spectral data in the visible light band and the near-infrared band. When C1>C2, compensating the visible light band data can effectively eliminate the errors and interference caused by data fluctuation differences, so that the spectral response curves of the two bands can more truly reflect the nutritional characteristics of tobacco leaves and improve the quality of spectral data; when C1≤C2, no redundant processing is performed, which avoids unnecessary data changes and ensures the integrity of the original valid data. Ultimately, through this processing method, the accuracy and reliability of feature extraction are significantly improved, thereby providing a more scientific and accurate basis for subsequent tobacco leaf nutritional diagnosis based on spectral data, helping to more accurately judge the nutritional status of tobacco leaves and assist in formulating reasonable fertilization and management strategies.

[0059] like Figure 1 As shown, in some embodiments of the present application, the environmental correction module is preset with a first reflectance compensation model R1=R×ω1 and a second reflectance supplement model R1=R×ω2, where R is the collected spectral reflectance, R1 is the corrected spectral reflectance, ω1 is the first weight coefficient, and ω2 is the second weight coefficient;

[0060] described Wherein, D2 is the real-time light intensity collected by the data collection unit, D1 is the historical average light intensity of the same period stored by the data storage unit, K1 is a preset first proportional coefficient, K2 is a second proportional coefficient, and K2<K1;

[0061] The environmental calibration module is configured to:

[0062] When D1<0.6×D2, the first reflectance compensation model is used to compensate and calibrate the input spectral characteristic parameters;

[0063] When 0.6×D2≤D1<D2, the second reflectance compensation model is used to compensate and calibrate the input spectral characteristic parameters;

[0064] When D1≥D2, no compensation calibration is performed on the input spectral characteristic parameters.

[0065] It's understandable that a graded light intensity correction mechanism is designed to address nonlinear interference issues in low light intensity environments like cloudy skies. By dividing the ratio of D1 to D2 into different intervals and matching them with corresponding reflectivity compensation models, dynamic adaptation of compensation weights is achieved. This inverse relationship conforms to the physical law that reflectivity attenuates in low-light environments.

[0066] like Figure 1 As shown, in some embodiments of the present application, the growth parameters of the tobacco leaves include plant height, number of leaves, and leaf color RGB value; the recognition module is pre-set with plant height matching score models, leaf number matching score models, and leaf color RGB value matching score models for different growth stages;

[0067] The recognition module is configured to input the plant height, number of leaves and leaf color RGB value information into the corresponding matching score model respectively, obtain the matching scores of the plant height, number of leaves and leaf color RGB values with all growth periods, perform weighted calculation on the scores, obtain the comprehensive scores of tobacco leaves for different growth periods, and identify the growth period with the highest score as the growth period of tobacco leaves.

[0068] Specifically, the tobacco leaf growth period can be divided into the seedling stage, root extension stage, vigorous growth stage and maturity stage. The standard growth parameters of different growth stages are shown in Table 1.

[0069] Table 1 Standard growth parameters of tobacco leaves during growth period

[0070]

[0071] Specifically, for the number of leaves, it is directly judged whether the measured number of leaves falls within the standard leaf number interval of each growth period. If it falls within the interval, 1 point is scored, otherwise 0 point is scored;

[0072] Specifically, for plant height and leaf color index:

[0073] Among them, x min is the minimum value of the standard growth parameter range, x max is the maximum value of the standard growth parameter range, x avg is the average value of the standard growth parameter, and x is the growth parameter value of the detected target tobacco leaf.

[0074] Set the weights of each parameter and calculate the comprehensive score of each growth period:

[0075] Comprehensive score = 0.3 × plant height score + 0.4 × leaf number score + 0.3 × leaf color index score.

[0076] like Figure 1 As shown, in some embodiments of the present application, the diagnosis module is configured to: preset a mapping relationship between the nitrogen content, potassium content and phosphorus content in leaves of different growth periods and the spectral characteristic parameters, match the corresponding mapping relationship according to the growth period of the tobacco leaves identified in the identification module, and obtain the corresponding real-time nitrogen content, potassium content and phosphorus content according to the spectral characteristic parameters corresponding to the tobacco leaves.

[0077] Specifically, the mapping relationship is mainly obtained through experimental fitting. First, the spectral characteristic parameters of tobacco leaf samples at different growth stages are collected, and then the tobacco leaf samples are taken through chemical experimental methods (such as Kjeldahl nitrogen determination method, molybdenum antimony colorimetry and flame photometry) to obtain the nitrogen, phosphorus and potassium content of the corresponding tobacco leaves. The spectral characteristic parameter data and nitrogen, phosphorus and potassium content data of different growth stages are fitted by the least squares method to obtain the corresponding mapping relationship, which is then stored in the diagnosis module.

[0078] like Figure 1 As shown, in some embodiments of the present application, the diagnostic module is further configured to: preset standard value ranges and optimal values of nitrogen content, potassium content, and phosphorus content in tobacco leaves corresponding to the growth period, and compare the real-time nitrogen content, potassium content, and phosphorus content with the standard value ranges;

[0079] When any one of the real-time nitrogen content, potassium content and phosphorus content is outside the standard value range, the tobacco leaf is judged to be in an abnormal nutritional state, and a fertilization strategy is generated based on the difference between the real-time nitrogen content, potassium content and phosphorus content and the optimal value.

[0080] Specifically, when any one of the detected nitrogen content, potassium content and phosphorus content is outside the standard value range, it is judged that the tobacco leaves are nutritionally abnormal, and farmers are reminded in time. At the same time, the data storage module stores the nitrogen, phosphorus and potassium contents of different fertilizers and the absorption rate of the plants. When the tobacco leaves are judged to be nutritionally abnormal, the judgment module can generate a precise fertilization strategy based on the difference between the detected nitrogen content, potassium content and phosphorus content and the optimal value, the nitrogen, phosphorus and potassium contents of different fertilizers and the absorption rate of the plants, which greatly improves the practicality of this application.

[0081] like Figure 2 As shown, in some embodiments of the present application, a tobacco leaf-based nutritional diagnosis method comprises the following steps:

[0082] Collect the growth parameters and light intensity of tobacco leaves in the target area, and obtain leaf reflectance spectrum data through a multispectral imaging device;

[0083] Performing spectral reflectance analysis on the leaf reflectance spectrum data, separating the spectral response curves of the visible light band and the near-infrared band, and extracting the spectral characteristic parameters of the leaf at this time;

[0084] Calibrate the spectral characteristic parameters according to the collected light intensity data using a preset reflectance compensation model to obtain calibrated spectral characteristic parameters;

[0085] The growth period of tobacco leaves in the target area is determined based on the growth parameters of the leaves, and the nitrogen content, potassium content and phosphorus content of the tobacco leaves are obtained based on the growth period of the tobacco leaves and the spectral characteristic parameters.

[0086] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or a combination of software and hardware embodiments. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0088] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A tobacco leaf-based nutritional diagnosis system, characterized in that: include: The data acquisition module is used to collect the growth parameters and light intensity of tobacco leaves in the target area and obtain leaf reflectance spectrum data through a multispectral imaging device; a feature extraction module for performing spectral reflectance analysis on the growth parameters and reflectance spectrum data collected by the data acquisition module, separating the spectral response curves of the visible light band and the near-infrared band, and extracting the spectral characteristic parameters of the leaves at this time; An environmental calibration module, configured to input the light intensity collected by the data acquisition module and the spectral characteristic parameters extracted by the feature extraction module into a preset reflectance compensation model to eliminate the influence of different light intensities on the spectral characteristic parameters; an identification module for determining the tobacco growth period in the target area based on the growth parameters collected by the data collection module; a diagnosis module, configured to obtain the nitrogen content, potassium content, and phosphorus content of the tobacco leaves based on the tobacco leaf growth period determined by the recognition module and the corresponding spectral characteristic parameters of the tobacco leaves; The data storage module is used to store historical data.

2. The tobacco leaf-based nutritional diagnosis system according to claim 1, characterized in that: The data acquisition module is configured to: preset a standard blade deployment angle B2 and acquire the blade deployment angle B1 in real time; When B1 < B2, adjust the shooting angle of the multispectral imaging device to be perpendicular to the main vein direction; when B1 ≥ B2, keep the shooting angle in the canopy plane looking down direction; After each adjustment, the reflectance spectrum data of three adjacent leaves are collected, the samples with the maximum and minimum reflectance values are eliminated, and the average reflectance of the remaining data is calculated as the effective spectral data input value.

3. The tobacco leaf-based nutritional diagnosis system according to claim 1, characterized in that: The feature extraction module is configured to: preset a coefficient K, and when separating the spectral response curves of the visible light band and the near infrared band, extract the visible light band reflectance fluctuation amplitude C1 and the near infrared band reflectance fluctuation amplitude C2; When C1>C2, the visible light band data is compensated and the processed spectral response curve is output, where the compensation amount = (C1-C2)×K; When C1≤C2, the original spectral response curve is directly output.

4. The tobacco leaf-based nutritional diagnosis system according to claim 1, characterized in that: The environmental correction module is preset with a first reflectivity compensation model R1=R×ω1 and a second reflectivity supplement model R1=R×ω2, where R is the collected spectral reflectivity, R1 is the corrected spectral reflectivity, ω1 is the first weight coefficient, and ω2 is the second weight coefficient.

5. The tobacco leaf-based nutritional diagnosis system according to claim 4, characterized in that: described Wherein, D2 is the real-time light intensity collected by the data collection unit, D1 is the historical average light intensity of the same period stored by the data storage unit, K1 is a preset first proportional coefficient, K2 is a second proportional coefficient, and K2<K1; The environmental calibration module is configured to: When D1<0.6×D2, the first reflectance compensation model is used to compensate and calibrate the input spectral characteristic parameters; When 0.6×D2≤D1<D2, the second reflectance compensation model is used to compensate and calibrate the input spectral characteristic parameters; When D1≥D2, no compensation calibration is performed on the input spectral characteristic parameters.

6. The tobacco leaf-based nutritional diagnosis system according to claim 1, characterized in that: The growth parameters of the tobacco leaves include plant height, number of leaves and leaf color RGB value; the recognition module is preset with plant height matching score models, leaf number matching score models and leaf color RGB value matching score models for different growth periods.

7. The tobacco leaf-based nutritional diagnosis system according to claim 6, characterized in that: The recognition module is configured to: input the plant height, number of leaves and leaf color RGB value information into the corresponding matching score model respectively, obtain the matching scores of the plant height, number of leaves and leaf color RGB values with all growth periods, perform weighted calculation on the scores, obtain the comprehensive scores of tobacco leaves for different growth periods, and identify the growth period with the highest score as the growth period of tobacco leaves.

8. The tobacco leaf-based nutritional diagnosis system according to claim 1, characterized in that: The diagnostic module is configured to: preset a mapping relationship between the nitrogen content, potassium content and phosphorus content in leaves of different growth periods and spectral characteristic parameters, match the corresponding mapping relationship according to the growth period of the tobacco leaves identified in the recognition module, and obtain the corresponding real-time nitrogen content, potassium content and phosphorus content according to the corresponding spectral characteristic parameters of the tobacco leaves.

9. The tobacco leaf-based nutritional diagnosis system according to claim 1, characterized in that: The diagnostic module is further configured to: preset standard value ranges and optimal values of nitrogen content, potassium content, and phosphorus content in tobacco leaves corresponding to the growth period, and compare the real-time nitrogen content, potassium content, and phosphorus content with the standard value ranges; When any one of the real-time nitrogen content, potassium content and phosphorus content is outside the standard value range, the tobacco leaf is judged to be in an abnormal nutritional state, and a fertilization strategy is generated based on the difference between the real-time nitrogen content, potassium content and phosphorus content and the optimal value.

10. A nutritional diagnosis method used in a tobacco leaf-based nutritional diagnosis system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Collect the growth parameters and light intensity of tobacco leaves in the target area, and obtain leaf reflectance spectrum data through a multispectral imaging device; Performing spectral reflectance analysis on the leaf reflectance spectrum data, separating the spectral response curves of the visible light band and the near-infrared band, and extracting the spectral characteristic parameters of the leaf at this time; Calibrate the spectral characteristic parameters according to the collected light intensity data using a preset reflectance compensation model to obtain calibrated spectral characteristic parameters; The growth period of tobacco leaves in the target area is determined based on the growth parameters of the leaves, and the nitrogen content, potassium content and phosphorus content of the tobacco leaves are obtained based on the growth period of the tobacco leaves and the spectral characteristic parameters.

Images