Plant cultivation method and system incorporating root pathogen monitoring

Abstract

The application provides a plant cultivation method and system combined with root disease and pest monitoring, relates to the technical field of disease and pest monitoring, and the method comprises the following steps: acquiring a target plant planting time point and a monitoring time interval, and constructing multiple monitoring time points; at each monitoring time point, collecting the soil temperature, soil humidity, soil oxygen and root rot fungus quantity of a target area, and acquiring a root rot fungus predicted quantity of a next monitoring time point; comparing the root rot fungus quantity collected at the current monitoring time point with the root rot fungus predicted quantity of the last monitoring time point, and obtaining an accumulated prediction error of the current monitoring time point; correcting the root rot fungus predicted quantity of the next monitoring time point according to the accumulated prediction error, and obtaining a corrected root rot fungus quantity; and when the corrected root rot fungus quantity is greater than or equal to a preset root rot fungus threshold, beneficial bacteria are put into the target area. The technical problems of low root rot fungus monitoring precision and lagging prevention and control measures in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of pest and disease monitoring, and in particular to a plant cultivation method and system that combines root pest and disease monitoring. Background Technology

[0002] Root diseases caused by root rot fungi have long hindered the healthy growth of cultivated crops, severely impacting yield and quality. However, in traditional cultivation management, monitoring of root diseases and pests relies heavily on regular manual experience, leading to problems such as inconsistent monitoring timing and limited parameter collection. This often results in the rapid spread of diseases due to missed optimal control windows. Therefore, existing technologies suffer from low accuracy in root rot fungi monitoring and delayed control measures. Summary of the Invention

[0003] This invention addresses the technical problems of low accuracy in root rot fungus monitoring and outdated control measures in existing technologies by providing a plant cultivation method and system that combines root disease and pest monitoring.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0005] In a first aspect, the present invention provides a plant cultivation method incorporating root disease and pest monitoring, comprising:

[0006] The planting time and monitoring time interval of the target plant are obtained. Multiple monitoring time points are constructed based on the planting time and monitoring time interval of the target plant, wherein the planting time of the target plant is the first monitoring time point.

[0007] At each monitoring time point, soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area were collected, and the predicted root rot fungus quantity for the next monitoring time point was obtained.

[0008] The number of root rot fungi collected at the current monitoring time point is compared with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point.

[0009] The predicted number of root rot fungi at the next monitoring time point is corrected based on the cumulative prediction error to obtain the corrected number of root rot fungi.

[0010] When the number of root rot bacteria is greater than or equal to the preset root rot bacteria threshold, beneficial bacteria are introduced into the target area.

[0011] Secondly, the present invention provides a plant cultivation system incorporating root disease and pest monitoring, comprising:

[0012] The time point construction module is used to obtain the target plant seedling planting time point and the monitoring time interval, and construct multiple monitoring time points based on the target plant seedling planting time point and the monitoring time interval, wherein the target plant seedling planting time point is the first monitoring time point;

[0013] The data acquisition module is used to collect soil temperature, soil moisture, soil oxygen and root rot fungus quantity in the target area at each monitoring time point, and to obtain the predicted number of root rot fungus at the next monitoring time point.

[0014] The error analysis module is used to synchronously compare the number of root rot fungi collected at the current monitoring time point with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point.

[0015] The quantity correction module is used to correct the predicted number of root rot fungi at the next monitoring time point based on the cumulative prediction error, so as to obtain the corrected number of root rot fungi.

[0016] The decision output module is used to release beneficial bacteria into the target area when the number of corrected root rot bacteria is greater than or equal to a preset root rot bacteria threshold.

[0017] The beneficial effects of this invention are:

[0018] Compared to existing technologies, this application first obtains the planting time and monitoring time interval of the target plants, constructing multiple monitoring time points to establish a precise and flexible time framework for the entire root disease and pest monitoring and control method. Secondly, at each monitoring time point, soil temperature, soil moisture, soil oxygen, and root rot fungi quantity in the target area are collected, and the predicted root rot fungi quantity for the next monitoring time point is obtained, providing a reliable data foundation for subsequent error correction and beneficial bacteria application decisions. Thirdly, the root rot fungi quantity collected at the current monitoring time point is simultaneously compared with the predicted root rot fungi quantity at the previous monitoring time point to obtain the cumulative prediction error for the current monitoring time point. This eliminates the interference of random errors and reflects the cumulative prediction error of the model's long-term prediction bias, providing a reliable deviation benchmark for subsequent prediction correction. Furthermore, the predicted root rot fungi quantity for the next monitoring time point is corrected based on the cumulative prediction error to obtain the corrected root rot fungi quantity, eliminating systematic errors and ensuring that the prediction results are closer to the actual root rot fungi quantity, providing a reliable data foundation for subsequent precise intervention. Finally, when the number of root rot bacteria is greater than or equal to the preset root rot bacteria threshold, beneficial bacteria are introduced into the target area. This links accurate predictions with actual control measures, ensuring timely control before the number of root rot bacteria reaches the hazard threshold.

[0019] Through the above technical solution, this application collects multiple environmental parameters such as soil temperature, humidity, and oxygen levels, as well as root rot fungus quantity data, in the target area at multiple monitoring time points. Using a root rot fungus predictor bound to plant type identifiers, it predicts the initial number of root rot fungi for the next monitoring time point. Then, by averaging the single prediction errors from historical time points, it obtains the cumulative prediction error reflecting the overall deviation trend of the model. Based on this, dynamic error correction is performed to obtain the corrected root rot fungus quantity. When the corrected root rot fungus quantity reaches a preset root rot fungus threshold, beneficial bacteria are precisely introduced into the target area in advance. This improves the accuracy of root rot fungus monitoring and achieves precise pre-control from passive response to active prevention, effectively curbing the spread of root rot disease. Attached Figure Description

[0020] Figure 1 A flowchart illustrating a plant cultivation method incorporating root disease and pest monitoring, provided by this invention.

[0021] Figure 2 This is a schematic diagram of a plant cultivation system that incorporates root disease and pest monitoring, provided by the present invention.

[0022] In the attached diagram, the components represented by each number are as follows:

[0023] The module includes a time point construction module 11, a data acquisition module 12, an error analysis module 13, a quantity correction module 14, and a decision output module 15. Detailed Implementation

[0024] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0026] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0027] Example 1, as Figure 1 As shown, this embodiment of the invention provides a plant cultivation method that combines root disease and pest monitoring, including:

[0028] S10: Obtain the planting time of the target plant and the monitoring time interval, and construct multiple monitoring time points based on the planting time of the target plant and the monitoring time interval, wherein the planting time of the target plant is the first monitoring time point;

[0029] In traditional cultivation methods, monitoring of root diseases and pests relies on regular manual testing based on experience. This leads to problems such as inconsistent monitoring timing and limited parameter collection, often resulting in the spread of diseases due to missed opportunities for optimal prevention and control.

[0030] To address the aforementioned issues, this application obtains the planting time and monitoring time interval of the target plant, and constructs multiple monitoring time points based on the planting time and monitoring time interval, wherein the planting time of the target plant is the first monitoring time point.

[0031] Specifically, step S10 in the method includes:

[0032] The system receives the monitoring time interval uploaded by the user and records the time point at which the target plant is planted in the target area, which is taken as the planting time point of the target plant.

[0033] The planting time of the target plant is taken as the first monitoring time point, and the monitoring time points are sorted according to the monitoring time interval to obtain multiple monitoring time points.

[0034] In this embodiment, the monitoring time interval uploaded by the user terminal is first received, and the time point when the target plant is planted in the target area is recorded as the target plant planting time point. The monitoring time interval is a time interval between two consecutive monitoring sessions, customized by the user terminal based on factors such as plant type and growth stage, for example, 3 days, 7 days, 10 days, etc. The target plant planting time point is a time point accurate to the day, for example, March 20, 2025. Exemplarily, the monitoring time interval uploaded by the user terminal (such as a terminal device operated by the grower), such as 7 days, is received, and the time point when the target plant is planted in the target area, such as March 20, 2025, is recorded as the target plant planting time point. Monitoring of the target plant is initiated from the target plant planting time point.

[0035] Secondly, the planting date of the target plant is taken as the first monitoring date, and the monitoring dates are sorted according to the monitoring time interval to obtain multiple monitoring dates, with the planting date of the target plant being the first monitoring date. For example, if March 20, 2025 is the planting date of the target plant, then this is taken as the first monitoring date, and then the monitoring dates are sorted according to the monitoring time interval (e.g., 7 days) to obtain multiple monitoring dates: March 20, 2025, March 27, 2025, April 3, 2025, April 10, 2025, ... Through a clear time sequence, it is ensured that each monitoring is based on a fixed time interval, making the number of root rot fungi and environmental parameters at different time points comparable. Furthermore, starting monitoring from the planting date of the target plant can completely cover the entire growth cycle of the target plant from planting to maturity, avoiding missed detections due to inconsistent time points, and ensuring that the risk of pests and diseases is continuously tracked.

[0036] In summary, compared to existing technologies, this application obtains the planting time and monitoring time interval of the target plant, and constructs multiple monitoring time points based on the planting time and monitoring time interval. This establishes a precise and flexible time framework for the entire root system disease and pest monitoring and control method.

[0037] S20: At each monitoring time point, collect soil temperature, soil moisture, soil oxygen and root rot fungus quantity in the target area, and obtain the predicted root rot fungus quantity for the next monitoring time point;

[0038] Soil temperature affects metabolic activity and thus determines the reproduction rate. For example, 20-30℃ is the optimal temperature range for most root rot fungi. Growth slows down when the temperature deviates from this range. Soil moisture regulates spore germination and diffusion through water conditions. Growth is significantly accelerated when the moisture content is >80%. Soil oxygen is related to respiration mode. For example, a low-oxygen environment is more conducive to the reproduction of anaerobic root rot fungi. Therefore, the predicted number of root rot fungi at the next monitoring time point can be predicted based on these factors.

[0039] To address the aforementioned issues, this application collects soil temperature, soil moisture, soil oxygen, and root rot fungi quantity in the target area at each monitoring time point, and obtains the predicted root rot fungi quantity for the next monitoring time point.

[0040] Specifically, step S20 in the method includes:

[0041] Multiple integrated soil sensors are uniformly arranged in the target area. The integrated soil sensors include a temperature sensor, a humidity sensor, and an oxygen sensor.

[0042] At the current monitoring point, the data collected by each of the multiple integrated soil sensors are statistically analyzed to obtain the soil temperature, soil moisture, and soil oxygen in the target area.

[0043] At the current monitoring time point, multiple root rot fungus monitoring sites are randomly determined in the target area. Multiple root soil samples from target plant roots are collected at the multiple root rot fungus monitoring sites for root rot fungus detection, and multiple root rot fungus monitoring quantities are obtained.

[0044] The average number of root rot fungi monitored is calculated to obtain the number of root rot fungi in the target area at the current monitoring time.

[0045] Based on the soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area at the current monitoring time point, the predicted number of root rot fungi at the next monitoring time point is obtained.

[0046] In this embodiment, multiple integrated soil sensors are first uniformly arranged in the target area. These integrated soil sensors include a temperature sensor, a humidity sensor, and an oxygen sensor. For example, the target area is uniformly divided into grids, and multiple integrated soil sensors are arranged at nodes such as the four corners and the center of the grid. These integrated soil sensors accurately monitor several soil temperature, soil humidity, and soil oxygen data in the target area.

[0047] Secondly, at the current monitoring point, statistical analysis is performed on the data collected by each of the multiple integrated soil sensors to obtain the soil temperature, soil moisture, and soil oxygen in the target area. For example, at the current monitoring point, box plot analysis is performed on several data points collected by each of the multiple integrated soil sensors to remove outliers. The box plot displays the median, upper quartile (Q3), lower quartile (Q1), maximum, and minimum values ​​of the data. The interquartile range (IQR) is used to define a reasonable data range, where IQR = Q3 - Q1. Values ​​exceeding Q3 + 1.5 × IQR or below Q1 - 1.5 × IQR are marked as outliers. These outliers may originate from temporary sensor malfunctions, sudden changes in the local soil environment (such as instantaneous high humidity near irrigation inlets), or external interference (such as temperature deviations caused by sensor obstruction). The box plot analysis is used to remove outliers. After removing outliers, the median or mean of the remaining valid data is calculated as a representative result of this type of sensor. For example, after removing outliers through box plot analysis, the median of several data points from the temperature sensor, humidity sensor, and oxygen sensor is taken as the soil temperature (e.g., 25℃), soil humidity (e.g., 60%), and soil oxygen temperature (e.g., 5%) of the target area. In this way, compared with simple mean calculation, the interference of local abnormal environment on the overall parameters can be effectively reduced, ensuring that the final soil temperature, soil humidity, and soil oxygen are more in line with the real environmental characteristics of the target area, providing high-quality input parameters for the subsequent root rot fungus prediction model and avoiding prediction deviations caused by data noise.

[0048] Secondly, at the current monitoring time point, multiple root rot fungus monitoring sites are randomly selected in the target area. Soil samples from the root systems of multiple target plants are collected at these sites for root rot fungus detection, yielding multiple root rot fungus monitoring quantities. It is crucial that the random selection of multiple monitoring sites within the target area covers the soil around the roots of different plants within the target area to avoid sample bias caused by concentration near healthy or diseased plants. For example, at the current monitoring time point, five root rot fungus monitoring sites are randomly selected in the target area. Soil samples are collected from a depth of 10-20 cm around the roots at each monitoring site, and multiple root rot fungus monitoring quantities are obtained using molecular biology methods, culture counting methods, etc., such as 120 CFU / g, 150 CFU / g, 135 CFU / g, 160 CFU / g, and 115 CFU / g.

[0049] Furthermore, the average of the multiple root rot fungus monitoring quantities is performed to obtain the root rot fungus quantity in the target area at the current monitoring time. For example, if the collected multiple root rot fungus monitoring quantities are [value missing], then at the current monitoring time, the root rot fungus quantity in the target area = (120 + 150 + 135 + 160 + 115) / 5 = 136 CFU / g. By randomly determining multiple root rot fungus monitoring locations and averaging the collected root rot fungus monitoring quantities, the impact of uneven spatial distribution of root rot fungi in the soil can be effectively reduced, ensuring that the obtained root rot fungus quantity is more regionally representative.

[0050] Finally, based on the soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area at the current monitoring time point, the predicted number of root rot fungi at the next monitoring time point is obtained. This is because the reproduction rate of root rot fungi is strongly correlated with the current number of root rot fungi and environmental factors such as soil temperature, soil moisture, and soil oxygen. Soil temperature affects the metabolic rate of root rot fungi, soil moisture affects the survival water conditions of root rot fungi, and soil oxygen content affects the respiration mode of root rot fungi. The current number of root rot fungi is the basic number for subsequent reproduction. For example, root rot fungi reproduce faster in high humidity and high temperature environments. Therefore, the growth trend of root rot fungi can be scientifically inferred based on the number of root rot fungi at the current monitoring time point and soil temperature, soil moisture, and soil oxygen, so as to achieve accurate prediction of the number of root rot fungi at the next monitoring time point.

[0051] Specifically, the phrase "predicting the predicted number of root rot fungi at the next monitoring time point based on the soil temperature, soil moisture, soil oxygen, and root rot fungi quantity in the target area at the current monitoring time point" includes:

[0052] Obtain the plant identifier of the target plant and retrieve the root rot fungus predictor bound to the plant identifier;

[0053] The root rot fungus predictor will obtain the predicted number of root rot fungi for the next monitoring time point by inputting the soil temperature, soil moisture, soil oxygen, and root rot fungus quantity of the target area at the current monitoring time point.

[0054] In this embodiment, due to significant differences in the composition of root exudates, growth cycle length, and root microenvironment (such as pH value and microbial community structure) among different plants, the types and reproduction rates of root-rot fungi vary significantly in the root systems of different plants. For example, the organic acids secreted by tomato roots tend to make the rhizosphere soil slightly acidic, which accelerates the spore germination and hyphal expansion of root-rot fungi. Wheat roots are sensitive to soil temperature during the jointing stage; when the temperature remains between 15-20°C and the humidity exceeds 70%, the reproduction rate of root-rot fungi can increase by 2-3 times. Therefore, general prediction models are difficult to adapt to these plant-specific differences. It is necessary to call a dedicated predictor trained on historical data for each specific plant type to ensure the accuracy of root-rot fungi quantity prediction. Specifically:

[0055] First, the plant identifier of the target plant is obtained, and the root rot fungus predictor bound to the plant identifier is retrieved. The plant identifier includes the variety type of the target plant, such as tomato or wheat. For example, if the plant identifier of the target plant is tomato, the root rot fungus predictor bound to tomato is retrieved.

[0056] Secondly, the soil temperature, soil moisture, soil oxygen, and root rot fungus quantity of the target area at the current monitoring time are input into the root rot fungus predictor to obtain the predicted root rot fungus quantity for the next monitoring time. For example, the root rot fungus predictor is trained based on historical cultivation records of the target plant. By capturing the correlation between soil temperature, soil moisture, soil oxygen, the root rot fungus quantity at the current monitoring time, and the growth of the root rot fungus quantity, it predicts and outputs the predicted root rot fungus quantity for the next monitoring time. For instance, the soil temperature (e.g., 25℃), soil moisture (e.g., 60%), soil oxygen temperature (e.g., 5%), and root rot fungus quantity (e.g., 136 CFU / g) of the target area at the current monitoring time are input into the root rot fungus predictor bound to Tomato-Zhongza 9 to predict and output the predicted root rot fungus quantity for the next monitoring time (e.g., 250 CFU / g).

[0057] Furthermore, the construction steps of the "root rot fungus predictor" include:

[0058] A root rot fungus general predictor is obtained, and the root rot fungus general predictor is configured with the monitoring time interval to obtain a root rot fungus basic predictor.

[0059] Obtain the plant type identifier of the target plant, collect historical cultivation records based on the plant type identifier, and construct a soil parameter sample set and a root rot fungus quantity label set according to the monitoring time interval. Each soil parameter sample includes soil temperature sample, soil moisture sample, soil oxygen sample and root rot fungus quantity sample.

[0060] Using soil temperature, soil moisture, soil oxygen, and root rot fungus quantity samples as inputs, and root rot fungus quantity labels corresponding to the root rot fungus quantity label set as supervision signals, the root rot fungus basic predictor is adjusted and trained to obtain the root rot fungus predictor.

[0061] In this embodiment of the application, a general predictor of root rot fungi is first obtained, and the general predictor of root rot fungi is configured with the monitoring time interval to obtain a basic predictor of root rot fungi. The general predictor of root rot fungi is trained based on the common laws of growth of most plant root rot fungi.

[0062] For example, a general predictor of root rot fungi can be obtained through the following technical path: 1. Data preparation: Collect historical monitoring data of various plants such as tomatoes, wheat, and cucumbers. Each set of monitoring data includes soil temperature, soil moisture, soil oxygen, and root rot fungi quantity at multiple consecutive monitoring time points. The actual root rot fungi quantity at the next monitoring time point is used as the label to ensure that the data covers the root environment and growth cycle of different crops, laying the foundation for the model to extract common patterns. Then, the difference in dimensions is eliminated by standardization, and the data is divided into training set and validation set according to an 8:2 ratio. 2. Model construction: A two-layer LSTM design can be adopted. The first layer LSTM (64 neurons) retains the time sequence output and focuses on capturing short-term environmental fluctuation characteristics, such as key signals such as sudden temperature rise and sudden increase in humidity. The second layer LSTM (32 neurons) deepens the long-term trend learning and focuses on the cumulative effect of continuous time steps, such as the superimposed effect of continuous high humidity and increasing temperature on root rot fungi reproduction. At the same time, overfitting is suppressed by a 20% dropout rate. 3. Model training uses soil temperature, soil moisture, soil oxygen, and the current number of root rot fungi as input features, and the actual number of root rot fungi at the next monitoring time point as the supervision signal. The absolute difference between the model's predicted number of root rot fungi at the next monitoring time point and the actual number of root rot fungi is used as the loss function, such as the mean absolute error (MAE). The model parameters are optimized through multiple iterations to continuously learn the common patterns of soil temperature, soil moisture, soil oxygen, and root rot fungi reproduction. The change of the loss function is monitored in real time during the iteration process until the loss function value is stably lower than the preset threshold. For example, if the loss fluctuation is less than 10 CFU / g for 5 consecutive iterations and the prediction accuracy on the validation set reaches 95%, the model training is considered to have converged, and a general predictor of root rot fungi is obtained.

[0063] For example, the root rot fungus general predictor is invoked, and the time step parameter of the root rot fungus general predictor is adjusted according to the monitoring time interval (e.g., 7 days) uploaded by the user to obtain the root rot fungus basic predictor. For example, the default prediction period of the root rot fungus general predictor (e.g., 10 days) is adjusted to 7 days to ensure that the prediction period is consistent with the actual monitoring time interval and to avoid errors caused by time dimension mismatch.

[0064] Secondly, the plant type identifier of the target plant is obtained. Based on the plant type identifier, historical cultivation records are collected, and a soil parameter sample set and a root rot fungus quantity label set are constructed according to the monitoring time interval. Each soil parameter sample includes soil temperature, soil moisture, soil oxygen, and root rot fungus quantity samples. For example, based on the plant type identifier of the target plant, such as tomato, historical cultivation records of tomatoes are collected. According to the monitoring time interval uploaded by the user, several soil temperature, soil moisture, soil oxygen, and root rot fungus quantity samples at different monitoring time points are obtained, along with the root rot fungus quantity at the next monitoring time point. These are used as the root rot fungus quantity label set, forming a one-to-one corresponding training sample pair. During data collection, the starting point of the monitoring time is dynamically adjusted through a sliding window, such as using different monitoring time points as the starting benchmark to extract time-series segments, maximizing the utilization of limited historical data, mining more effective samples from existing data, and improving the richness and representativeness of the sample set. For example, in the historical cultivation records of tomatoes, the soil temperature of 28℃, soil moisture of 75%, soil oxygen of 6%, and root rot fungus count of 130 CFU / g on June 1, 2024, were obtained as soil parameter samples. The actual root rot fungus count of 150 CFU / g at the next monitoring point on June 8, 2024, was obtained as a root rot fungus count label. Following the same method, the soil temperature, soil moisture, soil oxygen, and root rot fungus count at different monitoring points for different plant types, as well as the actual root rot fungus count at the next monitoring point, were obtained to obtain a soil parameter sample set and a root rot fungus count label set.

[0065] Finally, using the soil temperature sample, soil moisture sample, soil oxygen sample, and root rot fungus quantity sample of each soil parameter sample as input, and the root rot fungus quantity label corresponding to the root rot fungus quantity label set as the supervision signal, the root rot fungus basic predictor is adjusted and trained to obtain the root rot fungus predictor. For example, the core structure of the root rot fungus basic predictor, such as the neuron weights and fully connected layer thresholds of the LSTM, has learned the common patterns of root rot fungus reproduction through general data. These parameters are further adjusted using soil parameter sample sets and root rot fungus quantity label sets for the target plant to enhance the adaptation to the plant-specific patterns. For instance, the soil parameter sample set and root rot fungus quantity label set are divided into training and validation sets in an 8:2 ratio. Soil temperature, soil moisture, soil oxygen, and root rot fungus quantity samples are used as inputs, and the corresponding root rot fungus quantity labels in the root rot fungus quantity label set are used as supervision signals. By iteratively optimizing the parameters, the adaptation to the specific patterns is enhanced, and the error between the predicted value and the actual label is continuously reduced. When the model's accuracy on the validation set reaches 95%, the model training is considered to have converged, and a root rot fungus predictor customized for the target plant is finally obtained. In this way, through personalized training, the root rot fungus predictor can accurately capture the unique root rot fungus growth patterns of the plant, making the prediction results more consistent with the actual cultivation scenario and providing a reliable basis for subsequent correction and intervention decisions.

[0066] In summary, compared to existing technologies, this application collects soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area at each monitoring time point, and obtains the predicted root rot fungus quantity for the next monitoring time point. Thus, based on the plant identifier of the target plant, the corresponding root rot fungus predictor is retrieved, and the predicted root rot fungus quantity for the next monitoring time point is predicted and output, providing a reliable data foundation for subsequent error correction and beneficial bacteria application decisions.

[0067] S30: Simultaneously compare the number of root rot fungi collected at the current monitoring time point with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point;

[0068] The prediction process may be affected by sudden fluctuations in environmental factors (such as sudden changes in soil temperature and humidity exceeding the training sample range), data collection bias (such as temporary sensor errors or insufficient representativeness of root soil sampling), and specific changes in plant growth stages (such as changes in root resistance with the growth period leading to deviations in reproductive patterns), resulting in prediction errors. Therefore, it is necessary to integrate historical prediction biases to quantify the long-term prediction accuracy of the model and provide a stable bias reference benchmark for the correction of subsequent prediction results.

[0069] To address the aforementioned issues, this application simultaneously compares the number of root rot fungi collected at the current monitoring time point with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point.

[0070] Specifically, step S30 in the method includes:

[0071] The prediction error between the number of root rot fungi collected at the current monitoring time point and the predicted number of root rot fungi at the previous monitoring time point is calculated as the single prediction error at the current monitoring time point.

[0072] Extract the single prediction error of all monitoring time points before the current monitoring time point to obtain multiple historical single prediction errors;

[0073] The cumulative prediction error at the current monitoring time is obtained by averaging the multiple historical single prediction errors with the single prediction error at the current monitoring time.

[0074] In this embodiment, the prediction error between the number of root-rot fungi collected at the current monitoring time point and the predicted number of root-rot fungi at the previous monitoring time point is first calculated as the single prediction error for the current monitoring time point. For example, let the current monitoring time point be T. n The previous monitoring point was T n-1 The previous monitoring time point T n-1 The predicted number of root rot fungi is P. n Current monitoring time point T n The number of root rot fungi collected is A n The single prediction error E at the current monitoring time point n =P n -A n For example, if the predicted number of root rot fungi at monitoring time T2 is 250 CFU / g, and the actual number of root rot fungi collected at monitoring time T3 is 280 CFU / g, then the single prediction error at monitoring time T3 is E3 = 250 - 280 = -30 CFU / g. In this case, the first monitoring time has no predicted number of root rot fungi from the previous monitoring time, therefore, the first monitoring time does not participate in the prediction error calculation.

[0075] Secondly, the single prediction errors of all monitoring time points before the current monitoring time point are extracted to obtain multiple historical single prediction errors. For example, the single prediction errors of all monitoring time points before the current monitoring time point are collected; for instance, if the current monitoring time point is T... n Extract T2 to T n-1 Single prediction error at each time point: E2, E3, ..., E n-1 Multiple historical single prediction errors are obtained. These historical errors record the prediction deviation of the model at different stages and under different environmental conditions. If the current monitoring time point is T3, the single error E2 of the monitoring time point T2 needs to be extracted, such as 10 CFU / g, to obtain the historical single prediction error {10 CFU / g}.

[0076] Finally, the average of the multiple historical single prediction errors and the single prediction error at the current monitoring time is calculated to obtain the cumulative prediction error at the current monitoring time. For example, the multiple historical single prediction errors E2, E3, ..., E... are... n-1 The single prediction error E at the current monitoring time point n Merge to obtain E2, E3, ..., E n-1 E n Then calculate the mean: (E2 + E3 + ... + E n-1 +E n The cumulative prediction error for the current monitoring time point is calculated as (n-1) / 2, where n is the current monitoring time point number and n≥2. This is because the first monitoring time point has no predicted number of root rot fungi from the previous monitoring time point, so both the historical single prediction error and the single prediction error for the current monitoring time point are 0, resulting in a cumulative prediction error of 0. For example, if the multiple historical single prediction errors for the current monitoring time point T3 are {10 CFU / g}, and the single prediction error for the current monitoring time point T3 is -30 CFU / g, then the cumulative prediction error for the current monitoring time point T3 is (10-30) / 2 = -10 CFU / g. The cumulative prediction error is less than 0, indicating that the historical prediction quantity is generally low. Thus, by calculating the average of multiple historical single prediction errors and the single prediction error for the current monitoring time point, the random fluctuations of single errors are smoothed out, and a cumulative prediction error that reflects the overall deviation trend of the model is obtained.

[0077] In summary, compared to existing technologies, this application simultaneously compares the number of root rot fungi collected at the current monitoring point with the predicted number at the previous monitoring point to obtain the cumulative prediction error at the current monitoring point. This eliminates the interference of random errors and reflects the cumulative prediction error of the model's long-term prediction bias, providing a reliable bias benchmark for subsequent prediction correction.

[0078] S40: Correct the predicted number of root rot fungi at the next monitoring time point based on the cumulative prediction error to obtain the corrected number of root rot fungi.

[0079] The predicted number of root rot fungi at the next monitoring time point, generated by the root rot fungi predictor, may deviate from the actual data due to systematic errors. The aforementioned steps obtain the cumulative prediction error at the current monitoring time point, which can reflect the overall deviation trend of the model over a long period of time. Therefore, the predicted number of root rot fungi at the next monitoring time point can be specifically corrected based on this error.

[0080] To address the aforementioned issues, this application corrects the predicted number of root rot fungi at the next monitoring time point based on the cumulative prediction error, thereby obtaining the corrected number of root rot fungi.

[0081] Specifically, step S40 in the method includes:

[0082] Based on the cumulative prediction error, the prediction correction coefficient is obtained;

[0083] The predicted number of root rot fungi at the next monitoring time point is corrected based on the prediction correction coefficient to obtain the corrected number of root rot fungi.

[0084] In this embodiment, the prediction correction coefficient is first obtained based on the cumulative prediction error. The prediction correction coefficient is calculated as: cumulative prediction error / predicted number of root rot fungi at the next monitoring time point. A positive cumulative prediction error indicates that the predicted number has generally been higher than the actual measured value in the past, i.e., the prediction is too high, and the prediction correction coefficient is positive. A negative cumulative prediction error indicates that the predicted number has generally been lower than the actual measured value in the past, i.e., the prediction is too low, and the prediction correction coefficient is negative. For example, if the cumulative prediction error at the current monitoring time point T3 is -10 CFU / g, and the predicted number of root rot fungi at the next monitoring time point is 250 CFU / g, then the prediction correction coefficient at this time is -10 / 250 = -0.04.

[0085] Secondly, the predicted number of root rot fungi at the next monitoring time point is corrected based on the prediction correction coefficient to obtain the corrected number of root rot fungi. The corrected number of root rot fungi is calculated as: Corrected number of root rot fungi = Predicted number of root rot fungi at the next monitoring time point * (1 - Prediction correction coefficient). If the prediction correction coefficient is positive, it indicates that the prediction is too high, and the corrected number of root rot fungi is less than the predicted number. If the prediction correction coefficient is negative, it indicates that the prediction is too low, and the corrected number of root rot fungi is greater than the predicted number. For example, if the prediction correction coefficient is -0.04 and the predicted number of root rot fungi at the next monitoring time point is 250 CFU / g, the corrected number of root rot fungi at the next monitoring time point is corrected based on the prediction correction coefficient, resulting in a corrected number of root rot fungi = 250 * (1 + 0.04) = 260 CFU / g. This is because a negative prediction correction coefficient indicates that the prediction is too low. Through correction, the predicted number of root rot fungi at the next monitoring time point is amplified and compensated, making the corrected number of root rot fungi closer to the actual value.

[0086] In summary, compared to existing technologies, this application corrects the predicted number of root rot fungi at the next monitoring time point based on the accumulated prediction error, obtaining a corrected number of root rot fungi. This eliminates systematic errors, ensuring that the prediction results are closer to the actual number of root rot fungi, and providing a reliable data foundation for subsequent precise interventions.

[0087] S50: When the number of corrected root rot bacteria is greater than or equal to the preset root rot bacteria threshold, beneficial bacteria are introduced into the target area.

[0088] In this embodiment, a preset root rot fungus threshold is set based on the biological characteristics of the target plant, the disease occurrence pattern of root rot, and historical cultivation data. This threshold serves as a red line for determining whether the number of root rot fungi is sufficient to cause disease risk. For example, the preset root rot fungus threshold for tomato roots against Fusarium may be set at 3000 CFU / g, and the preset root rot fungus threshold for wheat against root rot pathogens may be set at 2000 CFU / g. Those skilled in the art can comprehensively consider factors such as plant growth stage and soil type to set the preset root rot fungus threshold.

[0089] Furthermore, the number of corrected root rot bacteria is compared with the preset root rot bacteria threshold. If the number of corrected root rot bacteria is greater than or equal to the preset root rot bacteria threshold, it indicates that the number of root rot bacteria at the next monitoring point has approached or reached a harmful level. If no intervention is provided, it may lead to the spread of the disease. If the number of corrected root rot bacteria is less than the preset root rot bacteria threshold, the risk is low and no intervention is required for the time being. Only continuous monitoring is needed.

[0090] Finally, when the number of root rot fungi is greater than or equal to the preset root rot fungi threshold, beneficial bacteria are introduced into the target area. Beneficial bacteria (such as Bacillus subtilis and Trichoderma) compete for nutrients and living space in the rhizosphere soil, inhibiting the germination of root rot fungi spores and the growth of mycelia. Some beneficial bacteria can also secrete antibiotics, chitinases and other substances, which directly destroy the cell structure of root rot fungi or inhibit their metabolic activity, and regulate soil pH and improve aeration, creating an environment unfavorable to the reproduction of root rot fungi, thus achieving effective control.

[0091] In this way, accurate predictions are linked to actual prevention and control measures, ensuring timely control before the number of root rot fungi reaches the harmful threshold.

[0092] In summary, the embodiments of this application have at least the following technical effects:

[0093] Compared to existing technologies, this application first obtains the planting time and monitoring time interval of the target plant, and then constructs multiple monitoring time points based on the planting time and monitoring time interval. This establishes a precise and flexible time framework for the entire root system disease and pest monitoring and control method.

[0094] Secondly, this application collects soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area at each monitoring time point, and obtains the predicted root rot fungus quantity for the next monitoring time point. Thus, based on the plant identifier of the target plant, the corresponding root rot fungus predictor is retrieved, and the predicted root rot fungus quantity for the next monitoring time point is predicted and output, providing a reliable data foundation for subsequent error correction and beneficial bacteria application decisions.

[0095] Furthermore, this application simultaneously compares the number of root rot fungi collected at the current monitoring point with the predicted number at the previous monitoring point to obtain the cumulative prediction error at the current monitoring point. This eliminates the interference of random errors and reflects the cumulative prediction error of the model's long-term prediction bias, providing a reliable bias benchmark for subsequent prediction correction.

[0096] Furthermore, this application corrects the predicted number of root rot fungi at the next monitoring time point based on the cumulative prediction error to obtain the corrected number of root rot fungi. This eliminates systematic errors, ensuring that the prediction results are closer to the actual number of root rot fungi, and providing a reliable data foundation for subsequent precise interventions.

[0097] Finally, this application introduces beneficial bacteria into the target area when the number of corrected root rot bacteria is greater than or equal to a preset root rot bacteria threshold. This links accurate prediction with actual control measures, ensuring timely prevention and control before the number of root rot bacteria reaches the harmful threshold.

[0098] Through the above technical solution, this application collects multiple environmental parameters such as soil temperature, humidity, and oxygen levels, as well as root rot fungus quantity data, in the target area at multiple monitoring time points. Using a root rot fungus predictor bound to plant type identifiers, it predicts the initial number of root rot fungi for the next monitoring time point. Then, by averaging the single prediction errors from historical time points, it obtains the cumulative prediction error reflecting the overall deviation trend of the model. Based on this, dynamic error correction is performed to obtain the corrected root rot fungus quantity. When the corrected root rot fungus quantity reaches a preset root rot fungus threshold, beneficial bacteria are precisely introduced into the target area in advance. This improves the accuracy of root rot fungus monitoring and achieves precise pre-control from passive response to active prevention, effectively curbing the spread of root rot disease.

[0099] Example 2, as Figure 2 As shown, based on the same inventive concept as the plant cultivation method combining root disease and pest monitoring provided in Embodiment 1, this embodiment of the invention also provides a plant cultivation system combining root disease and pest monitoring, comprising:

[0100] The time point construction module 11 is used to obtain the target plant seedling planting time point and the monitoring time interval, and construct multiple monitoring time points based on the target plant seedling planting time point and the monitoring time interval, wherein the target plant seedling planting time point is the first monitoring time point;

[0101] The data acquisition module 12 is used to collect soil temperature, soil moisture, soil oxygen and root rot fungus quantity in the target area at each monitoring time point, and to obtain the predicted number of root rot fungus at the next monitoring time point.

[0102] Error analysis module 13 is used to synchronously compare the number of root rot fungi collected at the current monitoring time point with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point.

[0103] The quantity correction module 14 is used to correct the predicted number of root rot fungi at the next monitoring time point based on the cumulative prediction error, so as to obtain the corrected number of root rot fungi.

[0104] The decision output module 15 is used to release beneficial bacteria into the target area when the number of corrected root rot bacteria is greater than or equal to a preset root rot bacteria threshold.

[0105] Specifically, the time point construction module 11 is used for:

[0106] The system receives the monitoring time interval uploaded by the user and records the time point at which the target plant is planted in the target area, which is taken as the planting time point of the target plant.

[0107] The planting time of the target plant is taken as the first monitoring time point, and the monitoring time points are sorted according to the monitoring time interval to obtain multiple monitoring time points.

[0108] Specifically, the data acquisition module 12 is used for:

[0109] Multiple integrated soil sensors are uniformly arranged in the target area. The integrated soil sensors include a temperature sensor, a humidity sensor, and an oxygen sensor.

[0110] At the current monitoring point, the data collected by each of the multiple integrated soil sensors are statistically analyzed to obtain the soil temperature, soil moisture, and soil oxygen in the target area.

[0111] At the current monitoring time point, multiple root rot fungus monitoring sites are randomly determined in the target area. Multiple root soil samples from target plant roots are collected at the multiple root rot fungus monitoring sites for root rot fungus detection, and multiple root rot fungus monitoring quantities are obtained.

[0112] The average number of root rot fungi monitored is calculated to obtain the number of root rot fungi in the target area at the current monitoring time.

[0113] Based on the soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area at the current monitoring time point, the predicted number of root rot fungi at the next monitoring time point is obtained.

[0114] Specifically, the phrase "predicting the predicted number of root rot fungi at the next monitoring time point based on the soil temperature, soil moisture, soil oxygen, and root rot fungi quantity in the target area at the current monitoring time point" includes:

[0115] Obtain the plant identifier of the target plant and retrieve the root rot fungus predictor bound to the plant identifier;

[0116] The root rot fungus predictor will obtain the predicted number of root rot fungi for the next monitoring time point by inputting the soil temperature, soil moisture, soil oxygen, and root rot fungus quantity of the target area at the current monitoring time point.

[0117] Furthermore, the construction steps of the "root rot fungus predictor" include:

[0118] A root rot fungus general predictor is obtained, and the root rot fungus general predictor is configured with the monitoring time interval to obtain a root rot fungus basic predictor.

[0119] Obtain the plant type identifier of the target plant, collect historical cultivation records based on the plant type identifier, and construct a soil parameter sample set and a root rot fungus quantity label set according to the monitoring time interval. Each soil parameter sample includes soil temperature sample, soil moisture sample, soil oxygen sample and root rot fungus quantity sample.

[0120] Using soil temperature, soil moisture, soil oxygen, and root rot fungus quantity samples as inputs, and root rot fungus quantity labels corresponding to the root rot fungus quantity label set as supervision signals, the root rot fungus basic predictor is adjusted and trained to obtain the root rot fungus predictor.

[0121] Specifically, the error analysis module 13 is used for:

[0122] The prediction error between the number of root rot fungi collected at the current monitoring time point and the predicted number of root rot fungi at the previous monitoring time point is calculated as the single prediction error at the current monitoring time point.

[0123] Extract the single prediction error of all monitoring time points before the current monitoring time point to obtain multiple historical single prediction errors;

[0124] The cumulative prediction error at the current monitoring time is obtained by averaging the multiple historical single prediction errors with the single prediction error at the current monitoring time.

[0125] The quantity correction module 14 is specifically used for:

[0126] Based on the cumulative prediction error, the prediction correction coefficient is obtained;

[0127] The predicted number of root rot fungi at the next monitoring time point is corrected based on the prediction correction coefficient to obtain the corrected number of root rot fungi.

[0128] The decision output module 15 is specifically used for:

[0129] When the number of root rot bacteria is greater than or equal to the preset root rot bacteria threshold, beneficial bacteria are introduced into the target area.

[0130] In summary, the embodiments of this application have at least the following technical effects:

[0131] Compared to existing technologies, this application firstly uses a time-point construction module to obtain the planting time of the target plant and the monitoring time interval, constructing multiple monitoring time points to establish a precise and flexible time framework for the entire root disease and pest monitoring and control method. Secondly, through a data acquisition module, soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area are collected at each monitoring time point, and the predicted root rot fungus quantity for the next monitoring time point is obtained, providing a reliable data foundation for subsequent error correction and beneficial bacteria application decisions. Thirdly, through an error analysis module, the root rot fungus quantity collected at the current monitoring time point is compared with the predicted root rot fungus quantity at the previous monitoring time point to obtain the cumulative prediction error for the current monitoring time point, eliminating the interference of random errors and reflecting the cumulative prediction error of the model's long-term prediction bias, providing a reliable deviation benchmark for subsequent prediction correction. Furthermore, through a quantity correction module, the predicted root rot fungus quantity for the next monitoring time point is corrected based on the cumulative prediction error to obtain the corrected root rot fungus quantity, eliminating systematic errors and ensuring that the prediction results are closer to the actual root rot fungus quantity, providing a reliable data foundation for subsequent precise intervention. Finally, through the decision output module, when the number of root rot fungi is greater than or equal to the preset root rot fungi threshold, beneficial bacteria are released into the target area. This links accurate predictions with actual control measures, ensuring timely control before the number of root rot fungi reaches the harmful threshold. In this way, the accuracy of root rot fungi monitoring is improved, and precise pre-control is achieved, moving from passive response to proactive prevention.

[0132] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0133] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0134] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0135] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0136] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0137] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0138] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A plant cultivation method incorporating root disease and pest monitoring, characterized in that, The method includes: The planting time and monitoring time interval of the target plant are obtained, and multiple monitoring time points are constructed based on the planting time and monitoring time interval, wherein the planting time of the target plant is the first monitoring time point; At each monitoring time point, soil temperature, soil moisture, soil oxygen, and root rot fungi quantity in the target area were collected, and the predicted root rot fungi quantity for the next monitoring time point was obtained, including: Multiple integrated soil sensors are uniformly arranged in the target area. The integrated soil sensors include a temperature sensor, a humidity sensor, and an oxygen sensor. At the current monitoring point, the data collected by each of the multiple integrated soil sensors are statistically analyzed to obtain the soil temperature, soil moisture, and soil oxygen in the target area. At the current monitoring time point, multiple root rot fungus monitoring sites are randomly determined in the target area. Multiple root soil samples from target plant roots are collected at the multiple root rot fungus monitoring sites for root rot fungus detection, and multiple root rot fungus monitoring quantities are obtained. The average number of root rot fungi monitored is calculated to obtain the number of root rot fungi in the target area at the current monitoring time. Based on the soil temperature, soil moisture, soil oxygen, and root rot fungus quantity in the target area at the current monitoring time point, predict the predicted root rot fungus quantity for the next monitoring time point, including: Obtain the plant identifier of the target plant and retrieve the root rot fungus predictor bound to the plant identifier; The soil temperature, soil moisture, soil oxygen, and number of root rot fungi in the target area at the current monitoring time point are input into the root rot fungi predictor to obtain the predicted number of root rot fungi at the next monitoring time point. The construction steps of the root rot fungus predictor include: A root rot fungus general predictor is obtained, and the root rot fungus general predictor is configured with the monitoring time interval to obtain a root rot fungus basic predictor. Obtain the plant type identifier of the target plant, collect historical cultivation records based on the plant type identifier, and construct a soil parameter sample set and a root rot fungus quantity label set according to the monitoring time interval. Each soil parameter sample includes soil temperature sample, soil moisture sample, soil oxygen sample and root rot fungus quantity sample. Using soil temperature, soil moisture, soil oxygen, and root rot fungus quantity samples as inputs, and root rot fungus quantity labels corresponding to the root rot fungus quantity label set as supervision signals, the root rot fungus basic predictor is adjusted and trained to obtain the root rot fungus predictor. The number of root rot fungi collected at the current monitoring time point is compared with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point. The predicted number of root rot fungi at the next monitoring time point is corrected based on the cumulative prediction error to obtain the corrected number of root rot fungi. When the number of root rot bacteria is greater than or equal to the preset root rot bacteria threshold, beneficial bacteria are introduced into the target area.

2. The method according to claim 1, characterized in that, The planting time and monitoring time interval of the target plant are obtained. Multiple monitoring time points are constructed based on the planting time and monitoring time interval, wherein the planting time is the first monitoring time point, including: The system receives the monitoring time interval uploaded by the user and records the time point at which the target plant is planted in the target area, which is taken as the planting time point of the target plant. The planting time of the target plant is taken as the first monitoring time point, and the monitoring time points are sorted according to the monitoring time interval to obtain multiple monitoring time points.

3. The method according to claim 1, characterized in that, The number of root rot fungi collected at the current monitoring time point is compared with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point, including: The prediction error between the number of root rot fungi collected at the current monitoring time point and the predicted number of root rot fungi at the previous monitoring time point is calculated as the single prediction error at the current monitoring time point. Extract the single prediction error of all monitoring time points before the current monitoring time point to obtain multiple historical single prediction errors; The cumulative prediction error at the current monitoring time is obtained by averaging the multiple historical single prediction errors with the single prediction error at the current monitoring time.

4. The method according to claim 1, characterized in that, The predicted number of root rot fungi at the next monitoring time point is corrected based on the cumulative prediction error to obtain the corrected number of root rot fungi, including: Based on the cumulative prediction error, the prediction correction coefficient is obtained; The predicted number of root rot fungi at the next monitoring time point is corrected based on the prediction correction coefficient to obtain the corrected number of root rot fungi.

5. A plant cultivation system incorporating root disease and pest monitoring, characterized in that, For performing the method according to any one of claims 1-4, comprising: The time point construction module is used to obtain the target plant seedling planting time point and the monitoring time interval, and construct multiple monitoring time points based on the target plant seedling planting time point and the monitoring time interval, wherein the target plant seedling planting time point is the first monitoring time point; The data acquisition module is used to collect soil temperature, soil moisture, soil oxygen and root rot fungus quantity in the target area at each monitoring time point, and to obtain the predicted number of root rot fungus at the next monitoring time point. The error analysis module is used to synchronously compare the number of root rot fungi collected at the current monitoring time point with the predicted number of root rot fungi at the previous monitoring time point to obtain the cumulative prediction error at the current monitoring time point. The quantity correction module is used to correct the predicted number of root rot fungi at the next monitoring time point based on the cumulative prediction error, so as to obtain the corrected number of root rot fungi. The decision output module is used to release beneficial bacteria into the target area when the number of corrected root rot bacteria is greater than or equal to a preset root rot bacteria threshold.

Images