Automated greenhouse pollen production method based on tobacco flower development stage determination results

Abstract

The application discloses an automatic greenhouse pollen production method based on tobacco flower development stage judgment results, which comprises the following steps: planting tobacco plants to be collected with pollen in a greenhouse environment, and the fertilization parameters in the greenhouse environment are as follows: the EC value of the fertilizer applied when the tobacco plants are in the clump stage is 0.24-0.50; the EC value of the fertilizer applied when the tobacco plants are in the clump stage to the full bloom stage is 0.37-0.85, and the EC value of the fertilizer applied when the tobacco plants are in the full bloom stage to the harvesting stage is 0.30-0.52; the air temperature regulation range in the greenhouse environment is 18-26 DEG C; the air relative humidity regulation range in the greenhouse environment is 50%-70%; and the CO2 concentration in the greenhouse environment is 300-400 ppm. The method can effectively improve the activity of the pollen formed by the tobacco plants, and can effectively improve the pollen activity by 16% compared with field planting.

Description

Technical Field

[0001] This application relates to the field of tobacco pollen production technology, and in particular to an automated greenhouse pollen production method based on the results of judging the development stage of tobacco flowers. Background Technology

[0002] Tobacco leaves are the material foundation for the development of Chinese-style cigarettes and cigarette brands, directly affecting the stability and improvement of cigarette product quality. Tobacco pollen, as the key to hybrid seed production of tobacco sterile lines and the source of tobacco leaf raw materials, is crucial for ensuring the quality of tobacco pollen production and is one of the core factors for achieving tobacco seed industry security and sustainable development of tobacco leaf production.

[0003] Current pollen production mainly relies on manual collection of pollen directly from tobacco flowers in the field. This method is limited by issues such as dwindling land resources, labor shortages, and severe diseases, resulting in limited yields and poor quality. To increase tobacco leaf yield, in vitro mutagenesis of pollen is necessary, such as the highly efficient in vitro liquid culture method for tobacco pollen disclosed in CN201010584722.8. However, the source of the pollen used in this method is not specified.

[0004] Existing research on pollen production mainly focuses on pollen media, pollen drying and collection methods, and pollen viability measurement, but no research has been conducted on how to achieve large-scale production of tobacco pollen in a greenhouse environment.

[0005] Meanwhile, when pollen is produced in a greenhouse environment, it is impossible to accurately and quickly determine the amount of fertilizer and the greenhouse environment control conditions to effectively improve pollen yield and pollen germination rate.

[0006] The information disclosed in the background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] This application addresses the aforementioned technical problems by providing an automated greenhouse pollen production method based on the results of tobacco flower development stage assessment. This method can effectively improve the germination rate of tobacco pollen in a greenhouse environment, increasing it by 16% compared to field planting environments.

[0008] This application provides an automated greenhouse pollen production method based on the results of tobacco flower development stage determination, including the following steps:

[0009] Tobacco plants intended for pollen collection were grown in a greenhouse environment. The fertilization parameters under these conditions were as follows: EC values ​​of 0.24–0.50 for fertilizer applied during the rosette stage; 0.37–0.85 for fertilizer applied from the rosette stage to full bloom; and 0.30–0.52 for fertilizer applied from full bloom to harvest. Using these control parameters increased the pollen germination rate of greenhouse-grown tobacco by 16%.

[0010] The air temperature control range in the greenhouse environment is 18℃~26℃; the air relative humidity control range in the greenhouse environment is 50%~70%; and the CO2 concentration in the greenhouse environment is 300~400ppm.

[0011] This method can effectively control the growth of tobacco plants, thereby effectively improving the activity of pollen produced by the tobacco plants. Compared with field planting, it can effectively increase pollen activity by 16%.

[0012] Preferably, coconut coir is used as the substrate for planting tobacco plants. The coconut coir is filled into the planting containers to a depth of 20-25 cm, with any two adjacent planting containers spaced apart. Coconut coir has good air permeability, is sterilized, and has fewer pests and diseases, thus reducing the damage and adverse effects of pests and diseases on pollen production from the source. The coconut coir used can be reused up to 4 times.

[0013] Preferably, the fertilization device used in the greenhouse includes: a three-level pipe network, a fertilizer applicator, a water storage tank, and multiple fertilizer mother liquor storage tanks; the water inlet of the fertilizer applicator is connected to the water storage tank and multiple fertilizer mother liquor storage tanks respectively; the water outlet of the fertilizer applicator is connected to the three-level pipe network; the three-level pipe network includes: a main pipe, branch pipes and multiple drip ribbons; the main pipe is connected to the water outlet of the fertilizer applicator; multiple branch pipes are spaced apart on the extension section of the main pipe; drip ribbons are installed on each branch pipe, and the drip ribbons are laid next to the tobacco plants in the greenhouse environment.

[0014] Preferably, the multiple fertilizer mother liquor storage tanks include: a first tank, a second tank, and a third tank; the first tank contains fertilizer mother liquor with a calcium ammonium nitrate concentration of 1900-2000 mg / L and a potassium nitrate concentration of 1300-1340 mg / L;

[0015] The second tank contains fertilizer mother liquor including: potassium dihydrogen phosphate at a concentration of 270–275 mg / L, magnesium sulfate at a concentration of 980–990 mg / L, chelated iron at a concentration of 35–45 mg / L, potassium iodide at a concentration of 1.00–2.00 mg / L, boric acid at a concentration of 40–45 mg / L, chelated manganese at a concentration of 110–120 mg / L, chelated zinc at a concentration of 45–55 mg / L, sodium molybdate at a concentration of 0.2–1.0 mg / L, chelated copper at a concentration of 15–25 mg / L, and cobalt chloride at a concentration of 0.02–1.00 mg / L.

[0016] The third tank contains phosphoric acid with a purity of over 90%.

[0017] Preferably, the multiple fertilizer mother liquor storage tanks include: a first tank, a second tank, and a third tank; the first tank contains fertilizer mother liquor with a calcium ammonium nitrate concentration of 1950 mg / L and a potassium nitrate concentration of 1320 mg / L;

[0018] The second tank contains fertilizer mother liquor including: potassium dihydrogen phosphate (272 mg / L), magnesium sulfate (986 mg / L), chelated iron (40 mg / L), potassium iodide (1.66 mg / L), boric acid (44.6 mg / L), chelated manganese (116 mg / L), chelated zinc (50 mg / L), sodium molybdate (0.5 mg / L), chelated copper (20 mg / L), and cobalt chloride (0.05 mg / L).

[0019] The third tank contains phosphoric acid with a purity of 90%.

[0020] The amount of each substance added to the fertilizer stock solution in each of the above tanks can be determined based on the loading volume. For example, when preparing 1L of fertilizer stock solution, simply add the corresponding mass of each substance.

[0021] Preferably, the fertilization device used in the greenhouse includes: a solenoid valve and a delivery pump; a solenoid valve and a delivery pump are respectively installed on the outlet pipes of the water storage tank and multiple fertilizer mother liquor storage tanks.

[0022] Preferably, the irrigation parameters under greenhouse conditions are as follows: the maximum substrate moisture content is 50%–60% when the tobacco plants are in the rosette stage; the maximum substrate moisture content is 70%–80% when the tobacco plants are in the rosette stage to full bloom stage; and the maximum substrate moisture content is 60%–75% when the tobacco plants are in full bloom to harvest stage. Using these control parameters can increase the pollen germination rate of tobacco produced in greenhouses by 16%.

[0023] Preferably, the temperature and humidity in the greenhouse environment are controlled as follows:

[0024] Step S11: Obtain the measured plant height growth Δh, maximum leaf length growth Δl, maximum leaf width growth Δw, and effective leaf number growth Δc on the nth day of tobacco planting relative to the first day of planting;

[0025] Step S12: Calculate the predicted plant height growth Δh, maximum leaf length growth Δl, maximum leaf width growth Δw, and effective leaf number growth Δc using the cumulative function based on the following tobacco plant growth values:

[0026] Δh=f1(x1,x2,x3,x4)

[0027] Δl=f2(x1,x2,x3,x4)

[0028] Δw=f3(x1,x2,x3,x4)

[0029] Δc = f3(x1,x2,x3,x4)

[0030] Where x1 represents the number of days since transplanting, x2 represents the number of days since the growth continued, x3 represents the average atmospheric temperature during the interval, and x4 represents the atmospheric humidity during the interval.

[0031] Step S13: Fit the predicted and measured values ​​respectively to obtain the predicted fitting curve and the measured fitting curve. Determine the atmospheric temperature and humidity under the greenhouse environment based on the similarity between the measured fitting curve and the predicted fitting curve.

[0032] After determining the temperature and humidity of the tobacco plant's growing environment using the above method, the accuracy of temperature and humidity control can be improved. This model has been developed into software for automatic operation; see details below. Figure 3 After adjusting the results obtained by fitting using the above method, effective regulation of tobacco plant growth can be achieved under greenhouse conditions, enabling reliable production of tobacco pollen in greenhouse environments.

[0033] Preferably, step S14: using the coefficient of determination R 2 The coefficient of determination R is used to evaluate the error between the model's predicted values ​​and the mean. 2 Calculate using the following formula:

[0034]

[0035] in, These are the predicted values ​​for plant height, maximum leaf length, maximum leaf width, or number of effective leaves obtained using the above model. y represents the average sample height, maximum leaf length, maximum leaf width, or effective leaf count of tobacco plants on day n of planting. i The measured values ​​of plant height, maximum leaf length, maximum leaf width, or effective leaf number on the nth day of tobacco planting are the actual measurements.

[0036] Step S14: If R 2 =0 indicates that the model's predicted value is 0. Compared with the true value y i Equal; if R 2 =1 indicates that the model's predicted value is 1. and sample mean Equal; when predicted values The closer to the sample mean That is, R 2 The closer the value is to 1, the better the model performance.

[0037] Using R 2After evaluation, the model can be effectively optimized and improved continuously, thereby increasing the accuracy of the model's prediction results.

[0038] Preferably, the greenhouse is equipped with: a central controller, a substrate EC value sensor, an automatic humidity monitoring sensor, an air temperature and humidity sensor, a CO2 sensor, automatic valves, automatic shading devices, an automatic high-pressure spray system, an automatic air conditioner, and an automatic CO2 concentration control device. The substrate EC value sensor, automatic humidity monitoring sensor, air temperature and humidity sensor, CO2 sensor, automatic valves, automatic shading devices, automatic high-pressure spray system, automatic air conditioner, and automatic CO2 concentration control device are all electrically connected to the central controller. Through the electrical connections of these devices, accurate control of the tobacco plant growth within the greenhouse is achieved based on the aforementioned parameters.

[0039] Preferably, when using recycled coconut coir, it undergoes disinfection treatment, which includes the following steps: spreading the substrate to a thickness of 3-6 cm, spraying with a 7-12% concentration of calcium hypochlorite solution while turning the substrate, covering with a film and fumigating for 24 hours, rinsing with clean water for about 5 minutes, and naturally drying until no moisture seeps out before packaging. Coconut coir treated in this way can be reused multiple times, reducing production costs.

[0040] Preferably, multiple cameras are installed in the upper part of the greenhouse. After capturing images of tobacco flowers through the cameras, the following operations are performed to determine the timing of flower harvesting:

[0041] Step S1: Obtain the flower length l on day i. i ;

[0042] Step S21: Determine if the flower length on day i is greater than 4-5cm. If not, return to step S1; if yes, obtain the flower length l on day i+1. i+1 Determine if l i+1 —l i ≤0.1cm, if so, the flower is in stage 7;

[0043] Step S3: Harvest flowers at stage 7 or wait for the top of the corolla to split before harvesting flowers at stage 8.

[0044] Determining the harvesting period using this method can avoid pollen that cannot germinate in vitro at harvesting time 6 and earlier, thereby increasing the proportion of harvested pollen that can germinate in vitro.

[0045] Preferably, step S3 further includes the following steps:

[0046] Step S31: Acquire a live image of the top of the flower.

[0047] Step S32: Determine if the tip of the flower's corolla is split. If the result is yes, then the harvesting period is 8.

[0048] If the result of the judgment is negative, return to step S31.

[0049] The above steps enable precise harvesting of flowers in periods 7 and 8, effectively increasing the activity of pollen in the harvested flowers and avoiding the harvesting of flowers that cannot bloom.

[0050] Preferably, step S2 further includes the following steps:

[0051] Step S22 determines whether the flower length on day i is greater than 3cm. If not, return to step S1; if yes, wait 20-30 hours and measure the real-time flower length l. j Determine the length of the flower l j If the length is greater than 4-5cm, then obtain the flower length l on day j+1. j+1 Determine if l j+1 —l j If the diameter is ≤0.1cm, then the flower is in its growth period.

[0052] 7. Combining time indicators can more accurately identify flowers in period 7.

[0053] The beneficial effects that this application can produce include:

[0054] 1) The automated greenhouse pollen production method provided in this application based on the results of tobacco flower development stage judgment can effectively reduce the occurrence of tobacco pests and diseases in pollen production by completing the pollen planting and harvesting process in the greenhouse and physically blocking the transmission of pathogens; and by using soilless cultivation technology to avoid continuous cropping obstacles, improve land utilization, achieve rational allocation of labor resources, and efficient utilization of natural resources.

[0055] 2) The automated greenhouse pollen production method provided in this application, based on the determination of tobacco flower development stages, uses a cumulative function based on tobacco plant growth values ​​to determine the temperature and humidity values ​​of the environment during the greenhouse pollen production process, effectively improving the yield and quality of tobacco pollen in the greenhouse environment. Simultaneously, the use of an evaluation index in conjunction with this function continuously optimizes the reliability of the evaluation results.

[0056] 3) The automated greenhouse pollen production method provided in this application based on the results of tobacco flower development stage determination can effectively avoid pollen contamination when tobacco is grown in a greenhouse environment using this method. The incidence of TLCV, TMV, tobacco bacterial wilt and black shank disease in each experimental tobacco plant is zero, and the activity of the harvested pollen is much higher than that of pollen harvested from field planting. The germination ability of tobacco pollen produced in greenhouse is improved by 16%.

[0057] 4) The automated greenhouse pollen production method based on the results of tobacco flower development stage judgment provided in this application can determine the timing of flower harvesting, effectively avoid harvesting flowers that cannot germinate, and effectively increase the germination rate of pollen in harvested flowers. The germination rate of harvested flowers can reach more than 45%. Attached Figure Description

[0058] Figure 1 This is a graph showing the correlation between tobacco plant growth indicators and environmental factors in Example 1 of this application;

[0059] Figure 2 The following is a comparison chart of the data fitting effect and the actual value in Embodiment 1 of this application: a) is the chart of the first batch of measurement results; b) is the chart of the second batch of measurement results.

[0060] Figure 3 Here are schematic diagrams of the growth model interface in Embodiment 1 of this application; a) is a graph; b) is a simulation diagram;

[0061] Figure 4 This is a schematic diagram of the fertilizer application device used in Embodiment 2 provided in this application;

[0062] Figure 5 Photographs of flowers at various stages in at least one embodiment provided in this application;

[0063] Figure 6 These are photographs of the germination status of tobacco pollen at different developmental stages in Example 1 of this application; a) is a photograph of pollen germinating on the pistil (stigma) after fluorescent staining at stages 5-8; b) is a photograph of pollen germinating on the culture medium at stages 5-8; as can be seen from Figure a, pollen at stage 6 on the pistil has already begun to germinate. See Figure b, pollen isolated from the culture medium will not germinate until stage 8. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0065] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0066] The controller used in this implementation scheme is an existing structure, and the control circuit can be implemented by those skilled in the art through simple programming. It is common knowledge in the field, and it is only used without modification. Therefore, the control method and circuit connection will not be described in detail.

[0067] Technical means not detailed in this application and not used to solve the technical problems of this application are all set according to common general knowledge in the field, and multiple common general knowledge setting methods can be implemented.

[0068] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments, but this does not limit the present invention in any way. Any modifications or improvements made based on the teachings of the present invention shall fall within the protection scope of the present invention.

[0069] Example

[0070] Unless otherwise specified, all materials and instruments used in the following embodiments were obtained through commercial channels; and all detection methods used are existing methods unless otherwise specified.

[0071] Example 1: Construction of a growth and development model of Yunyan 87 propagated through hydroponics

[0072] 1. Materials and Methods

[0073] Using Yunyan 87 as the research object and guided by the seed propagation technology scheme of the winter propagation base, 10 tobacco plants were planted in pots in Jinning greenhouse and Jinghong winter propagation base. 25 days after transplanting, 5 representative tobacco plants were selected and their agronomic traits (plant height, number of leaves, maximum leaf length, maximum leaf width, stem circumference) were measured weekly.

[0074] A total of 108 data entries, comprising 940 data points, were collected. Atmospheric temperature and humidity were collected using a field environment intelligent device provided by Shanghe Technology Co., Ltd. This device recorded atmospheric data every hour, and the daily average value was used as the meteorological factor for calculation in the experiment.

[0075] 2 Results and Analysis

[0076] 2.1 Data Processing

[0077] The plant height, maximum leaf length, and maximum leaf width of the tobacco plants were selected as the observation targets. Data were measured at regular intervals, and the growth of the tobacco plants in each interval was calculated based on the observed data, denoted as Δplant height, Δmaximum leaf length, and Δmaximum leaf width. The first batch of flue-cured tobacco was transplanted on July 13th. The number of days since transplanting was recorded at each measurement, and the interval between each observation was also calculated. The processed results are shown in Table 1. In the table, id represents the tobacco plant number, Δplant height is the increment between the current measurement and the previous measurement, and Δtransplanting days is the number of days between the current measurement and the previous measurement. The day of transplanting was recorded as day 0, and the measurement data on the day of transplanting was recorded as 0.

[0078] Table 1 shows an example of the processed data.

[0079]

[0080] 2.2 Correlation Analysis

[0081] There is a significant correlation between tobacco plant growth indicators such as plant height, stem circumference, maximum leaf length, maximum leaf width, and their growth rates and environmental factors. The Pearson correlation coefficients among them are shown in the figure. Figure 1 As shown, plant height is negatively correlated with Δtransplanting days, but Δplant height is positively correlated with both atmospheric temperature and Δtransplanting days. This result is consistent with actual growth conditions. Plant height initially grows rapidly, but the rate of increase slows down as the growing period progresses, hence the negative correlation. Although Δplant height increases more slowly, it continues to increase, thus showing a positive correlation. Stem circumference, maximum leaf length, and maximum leaf width also exhibit the same pattern as plant height.

[0082] The Δ plant height is positively correlated with atmospheric temperature, indicating that higher temperatures are beneficial to plant growth; however, the correlation coefficient is only 0.028, suggesting that temperature has a very limited impact on plant height. The Δ number of days after transplanting is positively correlated with Δ plant height, indicating that the plant height is higher at each measurement compared to the previous one.

[0083] The Δstem circumference, Δmaximum leaf length, Δmaximum leaf width, and Δplant height showed similar patterns: they were positively correlated with atmospheric temperature and Δtransplanting days, and negatively correlated with atmospheric humidity and Δtransplanting days. The correlation between Δplant height and atmospheric temperature and atmospheric humidity was less than 0.1, while the correlation with Δtransplanting days and Δtransplanting days was relatively strong. This indicates that atmospheric temperature and humidity have a relatively small impact on plant height, while growth time has a significant impact on height.

[0084] 2.3 Algorithm Description

[0085] This study investigates the relationship between plant height growth, maximum leaf length growth, maximum leaf width growth, and effective leaf number growth and atmospheric temperature, atmospheric humidity, transplanting time, and interval time. The model takes into account the number of days since transplanting, the number of days since transplanting, the average atmospheric temperature during the interval, and the average atmospheric humidity during the interval. It outputs the growth rates of these tobacco growth indicators, and finally sums the growth rates at different times to obtain the plant's growth value.

[0086] Prediction models were established for the plant height increase Δplant height, the maximum leaf length increase Δl, the maximum leaf width increase Δw, and the effective leaf number increase Δc.

[0087] The growth value of the tobacco plant on day n is calculated using a cumulative function, and the formula is as follows:

[0088] Δh=f1(x1,x2,x3,x4)

[0089] Δl=f2(x1,x2,x3,x4)

[0090] Δw=f3(x1,x2,x3,x4)

[0091] Δc = f3(x1,x2,x3,x4)

[0092] In the formula, x1 represents the number of days after transplanting, x2 represents the number of days of continued growth interval, x3 represents the average atmospheric temperature during the interval, and x4 represents the atmospheric humidity during the interval. Δh=f1(x1,x2,x3,x4) means that the flue-cured tobacco plant continues to grow for x2 days starting from day x1 after transplanting, during which the average atmospheric temperature is x3, the average atmospheric humidity is x4, and the plant height increases by Δh during this period.

[0093] Model performance uses the coefficient of determination R. 2 To evaluate, R 2 The definition is as follows, R 2 The error between the predicted value and the mean is calculated using the mean as a benchmark. If R 2 =0 indicates that the model's predicted value is 0. Compared with the true value y i They are equal. If R 2 =1 indicates that the model's predicted value is 1. and sample mean Equal. The main objective of this study is to fit the average value of tobacco plant growth indicators; therefore, the predicted values ​​are... The closer to the sample mean The better the model performance. R 2 The closer the value is to 1, the better the model performance.

[0094]

[0095] In the formula, For predicted values, Let y be the sample mean. i These are measured values.

[0096] 2.4 Model Effects

[0097] By accumulating the simulation outputs, the growth value for each day can be obtained. A regression model for the growth index is constructed using the GradientBoostingRegressor method, and the growth index for day n is calculated using a cumulative function. The model is validated using raw data in the experiment, and the results are as follows: Figure 2 As shown, the gray dashed line represents the actual measured data, while the red line represents the model's predictions. It can be observed that the model has a high degree of fit for plant height, maximum leaf width, and the number of effective leaves.

[0098] Depend on Figure 2 (a) It can be seen that the results predicted using the atmospheric temperature and humidity measured in the first batch have relatively high fitting degrees for plant height, maximum leaf length, maximum leaf width and effective number of leaves. Figure 2 (b) It can be seen that the results predicted using the atmospheric temperature and humidity measured in the second batch have a high degree of fit for plant height, maximum leaf width and effective number of leaves, while the fit for maximum leaf length is slightly lower.

[0099] 3. Conclusion

[0100] 3.1 The model has a high degree of fit.

[0101] Using R 2 The performance of the model was evaluated and analyzed, and the results are shown in Table 2. The plant height, maximum leaf width, and number of effective leaves were all greater than 0.8, indicating that the model's predictions were close to the measured data, and the model had a good fit. The maximum leaf length showed a high R-value in the fitting of the first batch of data. 2 The goodness of fit was 0.96 in the first batch of data, but dropped to 0.69 in the second batch, indicating that the goodness of fit of this indicator fluctuated significantly. (Observation) Figure 2 (b) shows that the measured maximum leaf length changed significantly 75 days after transplanting, which is the main reason for the model fluctuation.

[0102] Table 2 shows the first batch of data and the fitted R-value. 2 value

[0103] growth indicators Batch 1 Batch 2 Δplant height / cm 0.83 0.95 Δ Maximum leaf length / cm 0.96 0.69 Δ Maximum leaf width / cm 0.98 0.96 Δ number of effective leaves / blades 0.94 0.95

[0104] 3.2 The model has good generalization performance.

[0105] Two batches of experimental data were collected from different planting plots with varying atmospheric temperatures and humidity. However, the models constructed based on gradient descent regression effectively fitted the flue-cured tobacco growth indicators of both plots. As shown in Table 2, the R² value for Δplant height / cm... 2 The differences were 0.83 and 0.95, respectively, with a difference of 0.12, and the R value for Δmaximum leaf width / cm was... 2 The values ​​are 0.98 and 0.96 respectively, with a difference of only 0.02. The R-value for Δ effective number of blades / blade is... 2 The values ​​were 0.94 and 0.95 respectively, a difference of only 0.01. The two sets of data had different temperatures and humidity levels, but their R values ​​were... 2 The relatively small values ​​indicate that the model has good generalization performance and can reasonably predict the growth caused by changes in temperature and humidity.

[0106] 3.3 Model Function

[0107] This model can generate suitable air environment parameters (air temperature, air humidity) based on changes in tobacco plant agronomic traits (plant height, leaf width, number of leaves). The user interface is shown below. Figure 3 As shown in a~b), the temperature and humidity in the greenhouse are regulated by temperature and humidity control equipment to ensure the normal growth and development of tobacco plants.

[0108] Example 2: Greenhouse cultivation of tobacco plants

[0109] 1. Cultivation substrate

[0110] Desalinated compressed coconut coir can be used as a cultivation substrate and can be reused no more than 4 times.

[0111] 2. Installation of integrated water and fertilizer equipment

[0112] 2.1 Filtration Equipment

[0113] Use disc filters with a thickness of 125µm or larger as inlet filters. Wrap the ends of the water suction pipes in the water storage tank and the fertilizer mother liquor suction pipes with filter screens of about 0.1mm to prevent impurities from entering the irrigation system. The water suction position of the water supply pipe in the water storage tank should be at least 30cm above the bottom of the tank to prevent silt from being sucked in.

[0114] 2.2 Pipeline Network

[0115] A three-tiered pipe network is adopted, consisting of main pipes, branch pipes, and drip irrigation tape. The main and branch pipes are made of rigid PVC pipes and fittings. The drip irrigation tape is laid along the support structure, with each outlet spaced 50cm apart; multiple rows can be installed depending on the size of the support structure. Four-claw drip irrigation tapes with flow stabilizers are used, with a capillary tube length of 50cm. The rated working pressure is 80–100 kPa, and the drip irrigation tape flow rate is (4.0–6.0) L / h.

[0116] 2.3 Fertilizer application device

[0117] The fertilization equipment used, such as Figure 4 As shown, select a corrosion-resistant plastic storage tank, choosing an appropriate size based on the number of tobacco plants planted. The fertilizer mother liquor storage tank is equipped with a stirring device, a water supply device, and a fertilizer dispensing port. Add the appropriate fertilizer according to the fertilizer concentration, dissolving it in water to adjust the concentration. There are three fertilizer mother liquor storage tanks, labeled No. 1, No. 2, and No. 3. The specific types and concentrations of fertilizers stored are shown in Table 2. During application, tanks No. 1 and No. 2 draw in equal amounts of fertilizer mother liquor, and the fertilizer applicator automatically extracts phosphate to adjust the pH of the mixed fertilizer solution to 5.8–6.0, then delivers it to the drip irrigation system via pipeline.

[0118] Table 2. Fertilizer stored and concentration in each mother liquor storage tank.

[0119]

[0120] 2.4 Irrigation Equipment

[0121] Irrigation water originates from a reservoir and is delivered to the drip irrigation system via a pipeline network. The irrigation water is controlled by a solenoid valve. The size of the reservoir is determined based on the number of tobacco plants planted.

[0122] 2.5 Power Water Pump

[0123] A power water pump consists of a pump and a power unit. The appropriate pump should be selected based on the head and flow rate of the irrigation water in the greenhouse, and should be slightly larger than the maximum head and flow rate during operation. Its operating point should ideally be within the high-efficiency range, and a suitable matching power unit should be chosen.

[0124] 3. Automatic control system and parameters

[0125] The water and fertilizer supply and air environment control for flue-cured tobacco growth are all completed through computer-automated monitoring and control programs. Based on pre-set fertilization parameters (EC value 0.24–0.50 during the rosette stage, 0.37–0.85 from the rosette stage to full bloom, and 0.30–0.52 from full bloom to harvest), irrigation parameters (maximum substrate humidity 50%–60% during the rosette stage, 70%–80% from the rosette stage to full bloom, and 60%–75% from full bloom to harvest), and air environment parameters determined according to the aforementioned model (air temperature control range 18℃–26℃, relative humidity control range 50%–70%, CO2 concentration 300–400 ppm), automatic control of water, fertilizer, and air environment is achieved. The automatic control system mainly consists of a central controller, automatic substrate EC value and humidity monitoring sensors, air temperature, humidity, and CO2 sensors, automatic valves, automatic shading, automatic high-pressure spraying, automatic air conditioning, and an automatic CO2 concentration control device.

[0126] 4. Cultivation and Management

[0127] 4.1 Substrate Preparation

[0128] If using unused desalinated compressed coconut coir, soak it in water at a ratio of 1:8, stirring constantly to ensure it is fully soaked. If using coconut coir harvested from tobacco plants, it must be sterilized before use.

[0129] Coconut coir disinfection method: Spread the substrate evenly to a thickness of about 5cm. While spraying with a 10% calcium hypochlorite solution, turn the substrate over. Cover with a film and fumigate for 24 hours, then rinse with clean water for about 5 minutes. After the coconut coir has been fully soaked or disinfected, let it air dry naturally until no moisture seeps out before packaging. Use planting containers with a diameter of 45cm and a height of 40cm for packaging the coconut coir. Shake the containers from side to side during packaging to ensure that there are no large gaps in the coconut coir substrate. Fill the planting containers to a depth of 20-25cm. Place the packaged planting containers on the cultivation rack, with each container spaced 50cm apart.

[0130] 4.2 Transplanting

[0131] Manually turn on the irrigation system to maintain substrate moisture at over 90% of maximum humidity. Using a tool, plant the tobacco seedlings deep into the center of the planting container up to the first true leaf node. After transplanting, insert a 4-pronged drip irrigation arrow about 5cm away from the seedling and drip water until the moisture level reaches over 90% of maximum humidity.

[0132] 4.3 Water, fertilizer and air environment management

[0133] According to the set control parameters, water, fertilizer and air environment management is carried out through an intelligent control system.

[0134] Example 3: Study on pollen diseases and quality of tobacco plants grown in greenhouses

[0135] Tobacco plants were planted in a greenhouse environment according to the method in Example 2, and the temperature and humidity in the greenhouse environment were controlled according to the method in Example 1. The results are as follows:

[0136] 1. Survey of Tobacco Diseases in Greenhouse Pollen Production and Conventional Field Production

[0137] The tobacco pollen greenhouse production trial officially commenced in June 2021 and continued until April 2023, encompassing five seasons of soilless tobacco propagation trials over a period of two and a half years. During these six planting seasons, the incidence rates of TLCV, TMV, bacterial wilt, and black shank were all zero for all tobacco plants. Specific results are shown in the table below.

[0138] Table 3. Incidence rates of four diseases in soilless tobacco seedlings over six seasons in the past two years.

[0139] Incidence rate TLCV% TLCV% bacterial wilt % Black shank % June 2021 - September 2021 0 0 0 0 October 2021 - February 2022 0 0 0 0 March 2022 - July 2022 0 0 0 0 August 2022 - November 2022 0 0 0 0 December 2022 - April 2023 0 0 0 0 May 2023 - September 2023 0 0 0 0

[0140] Analysis of the incidence rates of TLCV, TMV, bacterial wilt, and black shank in field-grown tobacco plants during the same period revealed that TLCV, TMV, and bacterial wilt all occurred during the same period, with TLCV showing a particularly high incidence rate. Specific results are shown in the table below.

[0141] Table 4. Incidence rates of four diseases in tobacco plants planted in the field during the same period (2021-2022).

[0142]

[0143]

[0144] Table 5. Incidence rates of four diseases in tobacco plants planted in the field during the same period (2022-2023).

[0145] Investigation time Tobacco black shank Tobacco bacterial wilt TLCV TMV Clustering period 0.79％ 0 0 0.00％ Prosperous Long-Term 1.98％ 0 5.94％ 0.99％ Budding stage 1.98％ 0 7.92％ 0.99％ Peak bloom 1.98％ 0 8.91％ 1.32％ Harvest period 1.98％ 0 10.86％ 1.75％

[0146] The above analysis revealed that none of the six tobacco plants propagated by hydroponics developed TLCV, TMV, tobacco bacterial wilt, or black shank disease occurred. This preliminarily indicates that the integrated pest management measures adopted in hydroponics can effectively block the spread and infection of pathogens and prevent the occurrence of the above four diseases.

[0147] 2. Comparison of pollen viability between greenhouse production and conventional field production

[0148] Pollen from greenhouse-grown and conventional field-grown flowers was collected during the peak flowering period. The collection was repeated three times, with a 7-day interval between each collection. Three flowers that were about to bloom were collected from each treatment. The pollen was stored in ice boxes and brought back to the laboratory for testing of pollen viability.

[0149] Methods for detecting pollen viability:

[0150] (1) Collect pollen: On a sunny morning between 10:00 and 11:00, select healthy plants that are free from pests and diseases, collect flowers in the large bud stage, quickly put them in an ice box, and bring them back to the laboratory.

[0151] (2) Preparation of liquid culture medium: Take 1 mL of mother liquor I and add 99 mL of distilled water. After mixing evenly, add 0.1 mL of mother liquor II, 10 g of sucrose and 15 g of PEG4000, and stir to dissolve.

[0152] (3) Pollen culture: Add 1 ml of liquid culture medium to a 1.5 ml centrifuge tube, weigh 10 mg of pollen and add it to the centrifuge tube, cover and shake well. Perform 3 replicates for each pollen sample. Place the centrifuge tube in a light and temperature incubator, set the light intensity to 300-400 μmol·m-2·s-1, the temperature to 25-26℃, and incubate for 2 h;

[0153] (4) Pollen freezing: The pollen cultured for 2 hours was immersed in liquid nitrogen for 5 minutes, and the quick-frozen pollen was placed in a -20℃ freezer for storage.

[0154] (5) Pollen thawing: Take out the frozen pollen and thaw it at room temperature for 30 minutes until the culture medium is completely thawed;

[0155] (6) Microscopic examination: 60 μL of pollen culture medium was added to a grooved glass slide and photographed using a microscope. The pollen tube length being greater than the pollen grain diameter was used as the standard for pollen germination. Three fields of view were observed each time, and the pollen germination rate was counted. The result was obtained by repeating the observation three times and taking the average value. The pollen germination rate is the proportion of pollen with a pollen tube length greater than the pollen grain diameter in the field of view. The results are shown in the table below.

[0156] Table 6 Comparison of pollen germination rates between greenhouse and conventional field production.

[0157]

[0158] The results showed that the pollen germination ability of tobacco produced in greenhouses was improved by 16% compared with conventional field production.

[0159] Example 4

[0160] The difference from Example 2 is that the concentration of calcium ammonium nitrate and potassium nitrate in the fertilizer mother liquor loaded in the first tank is 2000 mg / L and 1340 mg / L, respectively.

[0161] The second tank contains fertilizer mother liquor including: potassium dihydrogen phosphate (275 mg / L), magnesium sulfate (990 mg / L), chelated iron (45 mg / L), potassium iodide (2.00 mg / L), boric acid (45 mg / L), chelated manganese (120 mg / L), chelated zinc (55 mg / L), sodium molybdate (1.0 mg / L), chelated copper (25 mg / L), and cobalt chloride (1.00 mg / L).

[0162] The third tank contains phosphoric acid with a purity of over 95%.

[0163] Disinfection of reclaimed coconut coir includes the following steps: spread the substrate to a thickness of 3cm, and while spraying with a 7% concentration of calcium hypochlorite solution, turn the substrate over.

[0164] Example 5

[0165] The difference from Example 2 is that the concentration of calcium ammonium nitrate and potassium nitrate in the fertilizer mother liquor loaded in the first tank is 1900 mg / L and 1300 mg / L, respectively.

[0166] The second tank contains fertilizer mother liquor including: potassium dihydrogen phosphate (270 mg / L), magnesium sulfate (980 mg / L), chelated iron (35 mg / L), potassium iodide (1.00 mg / L), boric acid (40 mg / L), chelated manganese (110 mg / L), chelated zinc (45 mg / L), sodium molybdate (0.2 mg / L), chelated copper (15 mg / L), and cobalt chloride (0.02 mg / L).

[0167] Disinfection of reclaimed coconut coir includes the following steps: spread the substrate to a thickness of 6cm, and while spraying a 12% concentration of calcium hypochlorite solution, turn the substrate over.

[0168] Example 6: Determination of flower morphology and pollen germination performance.

[0169] 1.1 Test Materials

[0170] The experimental materials were anthers of K326 at different developmental stages, such as Figure 5 As shown.

[0171] 1.2 Test Methods

[0172] a) Observation of microspore or pollen cell structure using transmission electron microscopy (TEM)

[0173] Tobacco anthers at different developmental stages (stages 1, 3, 5, and 8) were fixed overnight at 4°C in phosphate-buffered saline (PBS) containing 5% glutaraldehyde. They were then rinsed three times in 0.1M PBS and fixed for 3 hours in 1% OsO4 (dissolved in 0.1M PBS). After washing the material three times with PBS, it was dehydrated for 15 minutes in 50%, 70%, and 90% ethanol, 1 / 2% 90% ethanol:1 / 2 acetone (V:V), and pure acetone. The anthers were embedded in Spurr resin and ultrathin sections (70 nm) were cut using a diamond scalpel on a Leica Ultracut microtome (Leica EMUC6). The sections were double-stained with saturated uranyl acetate and lead citrate, and then examined using a transmission electron microscope (JEM1230, JEOL, Japan).

[0174] 2. Changes in the germination rate of tobacco pollen at different developmental stages

[0175] During pollen germination, pollen from stages 5-8 was extracted in vitro and stained on the stigma for photographing. The results are as follows: Figure 6 As shown in a) and b).

[0176] Stigma germination conditions: Apply pollen directly to the stigma at the tip of the pistil and allow it to germinate for 6 hours at 25°C and 70% relative humidity.

[0177] Germination in culture medium: Culture medium formula: 10% sucrose + 50 mg / L boric acid + 20 mg / L calcium chloride; use deionized water to prepare the solution, and the pH of the culture medium is 6.0-6.5. Measurement method: Add 1 mL of liquid culture medium to a 2 mL centrifuge tube, add 2.5 mg of the pollen to be tested, mix well to prepare a pollen suspension with a concentration of 2.5 mg / mL, and incubate at 25℃ in the dark for 3 hours to allow germination. Take 40-50 μL of the pollen suspension and drop it onto the center of a glass slide. Slowly cover the droplet with a coverslip from one side to prevent air bubbles. Examine under a 10× objective lens. A field of view containing 80-120 pollen grains evenly distributed is considered a valid field of view. Each sample is tested in triplicate, and three valid fields of view are randomly observed in each replicate. The pollen germination rate is then calculated.

[0178] A study comparing the correlation between pollen germination ability in vitro and in vivo at different developmental stages and the development of pollen microstructures revealed that pollen from stage 5 and earlier stages could not germinate on the stigma. Morphological observations showed that pollen contents were not sufficiently deposited before stage 6, and cell membrane electron transparency was low. Pollen from stages 6, 7, and 8 could germinate on the stigma, and the germination rate gradually increased. In the in vitro germination experiment, only pollen from stage 8 had an in vitro germination rate >85%, while the germination rates of pollen from stages 6 and 7 decreased sharply, indicating that the dehydration process of pollen after stage 6 had a significant impact on the in vitro germination rate.

[0179] 3. Summary

[0180] Before stage 6, pollen contents are not fully deposited, and cell membrane electron transparency is low. Pollen from stages 5 and earlier cannot germinate on the stigma, while pollen from stages 6, 7, and 8 can germinate on the stigma, with the germination rate gradually increasing. Based on the experimental results, stages 7 and 8 can be preliminarily used as one of the indicators for judging pollen maturity through flower morphology.

[0181] Period 7 Flower morphological characteristics description: The corolla has fully grown to its longest length, but the top is still closed.

[0182] Period 8 Flower morphological characteristics description: from the top of the corolla splitting to the corolla opening to its maximum diameter.

[0183] Example 7: Obtaining images of tobacco plant flowers under greenhouse conditions to determine harvesting time.

[0184] In the greenhouse tobacco cultivation environment of Example 2, multiple cameras were installed inside the greenhouse to acquire images of the flowers. Then, the following steps were performed to determine the timing of flower harvesting and to obtain pollen with a high germination rate:

[0185] See the criteria for judging the different stages of flower blooming. Figure 5 .

[0186] Step S1: Obtain the flower length l on day i. i In step S1, the length of the flower is measured after the image is acquired by the camera.

[0187] Step S21: Determine if the flower length on day i is greater than 4-5cm. If not, return to step S1; if yes, obtain the flower length l on day i+1. i+1 Determine if l i+1 —l i If the flower length is ≤0.1cm, then the flower is in stage 7; determine if the flower length on day i is greater than 3cm. If not, return to step S1; if so, wait 20-30 hours and measure the real-time flower length l. j Determine the length of the flower l j If the length is greater than 4-5cm, then obtain the flower length l on day j+1. j+1 Determine if l j+1 —l j If the value is ≤0.1cm, then the flower is in stage 7.

[0188] By combining time indicators, the flowers in period 7 can be obtained more accurately.

[0189] Step S22 determines whether the flower length on day i is greater than 3cm. If not, return to step S1; if yes, wait 20-30 hours and measure the real-time flower length l. j Determine the length of the flower l j If the length is greater than 4-5cm, then obtain the flower length l on day j+1. j+1 Determine if l j+1 —l j If the diameter is ≤0.1cm, then the flower is in stage 7. Combining this with time indicators allows for a more accurate identification of flowers in stage 7.

[0190] Step S3: Harvest flowers at stage 7 or wait for the top of the corolla to split before harvesting flowers at stage 8.

[0191] Step S31: Acquire a live image of the top of the flower.

[0192] Step S32: Determine whether the top of the flower's corolla is cracked. If the result is yes, harvest the flowers at stage 8. If the result is no, return to step S31.

[0193] Step S3 also includes the following steps:

[0194] Step S33: After the flower is determined to be in stage 7, wait 20-30 hours and observe whether the top of the flower has cracked. If it has cracked, then it is determined to be in stage 8 and harvested.

[0195] The tobacco plant is K326.

[0196] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automated greenhouse pollen production method based on the results of tobacco flower development stage determination, characterized in that, Includes the following steps: For tobacco plants to be pollinated in a greenhouse environment, the fertilization parameters under greenhouse conditions are as follows: EC value of fertilizer applied when the tobacco plants are in the rosette stage is 0.24-0.50; EC value of fertilizer applied when the tobacco plants are in the rosette stage to the full bloom stage is 0.37-0.85; and EC value of fertilizer applied when the tobacco plants are in the full bloom stage to the harvest stage is 0.30-0.

52. The air temperature control range in the greenhouse environment is 18℃~26℃; the air relative humidity control range in the greenhouse environment is 50%~70%; the CO2 concentration in the greenhouse environment is 300~400ppm. Temperature and humidity in the greenhouse environment are controlled as follows: Step S11: Obtain the measured plant height growth Δh, maximum leaf length growth Δl, maximum leaf width growth Δw, and effective leaf number growth Δc on the nth day of tobacco planting relative to the first day of planting; Step S12: Using the cumulative function model, calculate the predicted values ​​of plant height growth Δh, maximum leaf length growth Δl, maximum leaf width growth Δw, and effective leaf number growth Δc based on the following tobacco plant growth values: Δh=f1(x1,x2,x3,x4) Δl=f2(x1,x2,x3,x4) Δw=f3(x1,x2,x3,x4) Δc = f4(x1,x2,x3,x4) Where x1 represents the number of days since transplanting, x2 represents the number of days since the growth continued, x3 represents the average atmospheric temperature during the interval, and x4 represents the average atmospheric humidity during the interval. Step S13: Fit the predicted and measured values ​​respectively to obtain the predicted fitting curve and the measured fitting curve. Determine the atmospheric temperature and humidity under the greenhouse environment based on the similarity between the measured fitting curve and the predicted fitting curve. Irrigation parameters under greenhouse conditions: The maximum substrate moisture content is 50%~60% when the tobacco plants are in the rosette stage; the maximum substrate moisture content is 70%~80% when the tobacco plants are in the rosette stage to full bloom stage; and the maximum substrate moisture content is 60%~75% when the tobacco plants are in full bloom to harvest stage.

2. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 1, characterized in that, The substrate used for planting tobacco plants is coconut coir. The coconut coir is filled into the planting container to a depth of 20-25cm, with any two adjacent planting containers spaced apart.

3. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 2, characterized in that, When using recycled coconut coir, it is disinfected. The disinfection process includes the following steps: spread the substrate to a thickness of 3-6 cm, spray with a 7-12% calcium hypochlorite solution while turning the substrate, cover with a film and fumigate for 24 hours, rinse with clean water for about 5 minutes, and air dry naturally until no moisture seeps out before packaging.

4. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 1, characterized in that, The fertilization equipment used in the greenhouse includes: a three-level pipe network, a fertilizer applicator, a water storage tank, and multiple fertilizer mother liquor storage tanks; the water inlet of the fertilizer applicator is connected to the water storage tank and multiple fertilizer mother liquor storage tanks respectively; the water outlet of the fertilizer applicator is connected to the three-level pipe network; the three-level pipe network includes: a main pipe, branch pipes and multiple drip ribbons; the main pipe is connected to the water outlet of the fertilizer applicator; multiple branch pipes are installed at intervals on the extension section of the main pipe; drip ribbons are installed on each branch pipe and are laid next to the tobacco plants in the greenhouse environment.

5. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 4, characterized in that, The multiple fertilizer mother liquor storage tanks include: a first tank, a second tank, and a third tank; the first tank contains fertilizer mother liquor with a calcium ammonium nitrate concentration of 1900-2000 mg / L and a potassium nitrate concentration of 1300-1340 mg / L; The second tank contains fertilizer mother liquor including: potassium dihydrogen phosphate at a concentration of 270–275 mg / L, magnesium sulfate at a concentration of 980–990 mg / L, chelated iron at a concentration of 35–45 mg / L, potassium iodide at a concentration of 1.00–2.00 mg / L, boric acid at a concentration of 40–45 mg / L, chelated manganese at a concentration of 110–120 mg / L, chelated zinc at a concentration of 45–55 mg / L, sodium molybdate at a concentration of 0.2–1.0 mg / L, chelated copper at a concentration of 15–25 mg / L, and cobalt chloride at a concentration of 0.02–1.00 mg / L. The third tank contains phosphoric acid with a purity of over 90%.

6. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 5, characterized in that, The multiple fertilizer mother liquor storage tanks include: a first tank, a second tank, and a third tank; the first tank contains fertilizer mother liquor with a calcium ammonium nitrate concentration of 1950 mg / L and a potassium nitrate concentration of 1320 mg / L; The second tank contains fertilizer mother liquor including: potassium dihydrogen phosphate (272 mg / L), magnesium sulfate (986 mg / L), chelated iron (40 mg / L), potassium iodide (1.66 mg / L), boric acid (44.6 mg / L), chelated manganese (116 mg / L), chelated zinc (50 mg / L), sodium molybdate (0.5 mg / L), chelated copper (20 mg / L), and cobalt chloride (0.05 mg / L); the third tank contains 90% pure phosphoric acid.

7. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 4, characterized in that, The fertilization equipment used in the greenhouse also includes: solenoid valves and delivery pumps; solenoid valves and delivery pumps are respectively installed on the outlet pipes of the water storage tank and multiple fertilizer mother liquor storage tanks.

8. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 5, characterized in that, It also includes step S14: using the coefficient of determination R 2 The coefficient of determination R is used to evaluate the error between the model's predicted values ​​and the mean. 2 Calculate using the following formula: in, The predicted plant height, maximum leaf length, maximum leaf width, or effective leaf length obtained according to the above model are... Predicted number of pieces These represent the sample mean of plant height, maximum leaf length, maximum leaf width, or effective leaf count on the nth day after tobacco planting. y i The measured values ​​of plant height, maximum leaf length, maximum leaf width, or effective leaf number on the nth day of tobacco planting are the actual measurements. When R 2 When R approaches 0, it indicates that the model's predictive ability is weak. 2 When the value is close to 1, it indicates that the model's predicted value and the measured value are similar. y i The fit is good.

9. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 4, characterized in that, The greenhouse is equipped with: a central controller, automatic monitoring sensors for substrate EC value and humidity, air temperature, humidity and CO2 sensors, automatic valves, automatic shading, automatic high-pressure spraying, automatic air conditioning, and an automatic CO2 concentration control device. The automatic monitoring sensors for substrate EC value and humidity, air temperature, humidity and CO2 sensors, automatic valves, automatic shading, automatic high-pressure spraying, automatic air conditioning, and automatic CO2 concentration control device are all electrically connected to the central controller. Through the electrical connection of the above devices, the growth of tobacco plants in the greenhouse can be accurately controlled according to the aforementioned parameters.

10. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 1, characterized in that, Multiple cameras are installed in the upper part of the greenhouse. After capturing images of tobacco flowers using these cameras, the following steps are taken to determine the timing of flower harvesting: Step S1: Obtain the flower length l on day i. i ; Step S2: Determine if the flower length on day i is greater than 4cm. If not, return to step S1. If so, then obtain the flower length l on the (i+1)th day. i+1 Determine if l i+1 —l i ≤0.1cm, if so, the flower is in stage 7; Step S3: Harvest flowers at stage 7 or wait for the top of the corolla to split open before harvesting flowers at stage 8; Period 7 Flower morphological characteristics: The corolla has fully grown to its longest length, but the apex is still closed; Period 8 Flower morphological characteristics: from the tip of the corolla splitting to the corolla opening to its maximum diameter.

11. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 10, characterized in that, Step S3 also includes the following steps: Step S31: Acquire a live image of the top of the flower. Step S32: Determine whether the top of the flower's corolla is cracked. If the result is yes, harvest the flowers at stage 8. If the result is no, return to step S31.

12. The automated greenhouse pollen production method based on the tobacco flower development stage determination result according to claim 10, characterized in that, Step S2 also includes the following steps: Step S21 determines whether the flower length on day i is greater than 3cm. If not, return to step S1; if yes, wait 20-30 hours and measure the real-time flower length l. j Determine the length of the flower l j Is it greater than 4cm? If so, get the flower length l on day j+1. j+1 Determine if l j+1 —l j If the value is ≤0.1cm, then the flower is in period 7. Combining time indicators can more accurately identify flowers in period 7.

Images