Mulberry field monitoring method and system based on Internet of Things

Abstract

The invention relates to the technical field of intelligent agriculture, in particular to a mulberry field monitoring method and system based on the Internet of Things. The method comprises the following steps: acquiring environment and soil data of a mulberry field through a sensor network to obtain original sensor data; performing time sequence data integration on the original sensor data to obtain an environment parameter data set; the method comprises the following steps: performing multispectral image acquisition on mulberries in a mulberry field through a multispectral camera, and performing image preprocessing to obtain preprocessed image data; estimating the chlorophyll content according to the preprocessed image data to obtain chlorophyll content data; and performing growth health parameter analysis according to the preprocessed image data and the chlorophyll content data to obtain growth health parameters. By combining the sensor technology, the multispectral imaging technology and the intelligent algorithm, comprehensive, accurate and real-time monitoring of the mulberry field is achieved, and an effective technical means is provided for fine management of the mulberry field.

Description

Technical Field

[0001] The present invention relates to the technical field of smart agriculture, and in particular to a mulberry garden monitoring method and system based on the Internet of Things. Background Art

[0002] Traditional mulberry garden management mainly relies on manual experience, relying on manual observation of the growth status of mulberry trees, soil moisture, etc., and fertilization, irrigation, and pest control based on experience. This method has problems such as strong subjectivity, low efficiency, and poor accuracy, and it is difficult to meet the needs of modern agricultural refined management. With the development of Internet of Things technology, some mulberry garden monitoring systems based on the Internet of Things have begun to appear. These systems usually include components such as sensors, wireless networks, and data processing platforms, which can realize real-time collection and remote monitoring of mulberry garden environment and mulberry tree growth information. However, the existing mulberry garden monitoring methods based on the Internet of Things still have the problems of single monitoring means and low efficiency.

[0003] The problems of single monitoring method and low efficiency are as follows: The monitoring indicators are single: they mainly focus on the monitoring of environmental parameters (such as temperature, humidity, light, etc.), and lack effective monitoring of the growth status of mulberry trees themselves (such as chlorophyll content, diseases and pests, etc.).

[0004] Insufficient data analysis capabilities: Only simple data collection and display can be achieved, and there is a lack of in-depth analysis and mining of data, and it is unable to provide effective decision-making support.

[0005] Fertilization decisions lack scientific basis: Fertilization plans still rely mainly on manual experience and lack comprehensive consideration of mulberry tree growth stages, nutrient requirements and soil fertility conditions, resulting in low fertilization efficiency. Summary of the invention

[0006] Based on this, it is necessary to provide a mulberry garden monitoring method and system based on the Internet of Things to solve at least one of the above technical problems.

[0007] To achieve the above purpose, a mulberry garden monitoring method based on the Internet of Things includes the following steps: Step S1: Collect environmental and soil data of the mulberry garden through the sensor network to obtain original sensor data; perform time-sequential data integration on the original sensor data to obtain an environmental parameter data set; Step S2: collecting multispectral images of mulberry trees in the mulberry garden through a multispectral camera, and performing image preprocessing to obtain preprocessed image data; estimating chlorophyll content based on the preprocessed image data to obtain chlorophyll content data; analyzing growth health parameters based on the preprocessed image data and the chlorophyll content data to obtain growth health parameters; Step S3: extracting growth stage features from the environmental parameter data set and the growth health parameters to obtain key features of the growth stage; performing growth stage discrimination based on the key features of the growth stage to obtain a growth stage discrimination result; Step S4: performing initial nutrient demand calculation according to the growth stage determination result and the preset basic nutrient demand model to obtain initial nutrient demand data; performing environmental factor data extraction and soil data extraction on the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; performing environmental factor impact correction analysis according to the relevant environmental factor data to obtain environmental factor impact data; performing soil fertility index extraction according to the relevant soil data to obtain soil fertility index data; performing soil condition impact analysis according to the soil fertility index data to obtain soil condition impact data; dynamically adjusting the nutrient demand of the initial nutrient demand data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient demand; generating a nutrient demand formula according to the growth stage determination result and the basic nutrient demand to obtain the nutrient demand formula data; Step S5: Acquire the data of mulberry garden fertilization equipment; perform fertilization planning according to the nutrient requirement formula and the data of mulberry garden fertilization equipment to realize the mulberry tree growth monitoring operation.

[0008] The present invention collects and integrates environmental and soil data in real time through a sensor network, can comprehensively and dynamically monitor the growth environment of a mulberry garden, provide a reliable data basis for subsequent analysis and decision-making, and realize accurate perception of the mulberry garden environment. Utilizing multispectral imaging technology, the phenotypic characteristics, chlorophyll content and pest and disease information of mulberry trees can be quickly and non-destructively obtained, the growth health status of mulberry trees can be comprehensively evaluated, and an important reference basis can be provided for accurate fertilization. By analyzing environmental data and mulberry growth health parameters, the growth stage of mulberry trees can be accurately judged, key guidance can be provided for formulating targeted fertilization plans, and intelligent management of the mulberry growth process can be realized. Taking into account factors such as the growth stage of mulberry trees, environmental factors and soil conditions, the nutrient requirements of mulberry trees can be accurately calculated, and personalized fertilization formulas can be generated to realize accurate fertilization, improve fertilizer utilization efficiency, and reduce pollution to the environment. Combining fertilization equipment data and nutrient requirement formulas for fertilization planning can realize automated and accurate fertilization operations, improve fertilization efficiency, reduce labor costs, and realize continuous monitoring of mulberry growth, ultimately improving the yield and quality of mulberry gardens. Therefore, the present invention provides a mulberry garden monitoring method based on the Internet of Things. By combining sensor technology, multispectral imaging technology and intelligent algorithms, comprehensive, accurate and real-time monitoring of the mulberry garden is achieved, which solves the problems of single monitoring means and low efficiency of existing methods, and provides effective technical means for the refined management of mulberry gardens.

[0009] Preferably, step S1 comprises the following steps: Step S11: deploying a sensor network according to the terrain and area of ​​the mulberry garden, and positioning the sensors to obtain sensor location information; Step S12: collecting environment and soil data according to the sensor location information and a preset collection time interval to obtain raw sensor data; Step S13: performing data preprocessing on the original sensor data to obtain preprocessed sensor data; Step S14: Performing time-sequential data integration on the pre-processed sensor data according to the sensor location information to obtain an environmental parameter data set.

[0010] The present invention can ensure that sensor nodes are evenly distributed and avoid data collection blind spots by deploying and accurately locating the sensor network according to the topography and area of ​​the mulberry garden, thereby comprehensively reflecting the environment and soil conditions in the mulberry garden, improving the representativeness and accuracy of data collection, and providing a reliable data basis for subsequent accurate analysis and decision-making. According to the sensor location information and the preset collection time interval, data collection can be carried out to achieve real-time and continuous monitoring of the mulberry garden environment and soil conditions, obtain rich dynamic data, help to timely discover environmental anomalies and soil changes, and provide necessary data support for subsequent growth stage judgment and nutrient demand calculation. Pre-processing operations such as cleaning, interpolation and normalization of the original sensor data can effectively remove noise and abnormal data, supplement missing information, and convert data of different dimensions into unified dimensionless data, improve the quality and consistency of the data, and lay the foundation for subsequent data analysis and model training. By integrating the preprocessed data according to the sensor location information and time sequence, a complete and ordered environmental parameter data set can be constructed. This data set contains the environmental and soil information at each time point and each location, which can clearly reflect the temporal and spatial changes in the mulberry garden environment and soil conditions, and provide complete and reliable data support for subsequent growth stage judgment and nutrient demand calculation.

[0011] Preferably, step S2 comprises the following steps: Step S21: collecting multispectral images of mulberry trees in the mulberry garden using a multispectral camera to obtain a multispectral image data set; Step S22: performing image preprocessing on the multispectral image data set to obtain preprocessed image data; performing mulberry tree image feature extraction on the preprocessed image data to obtain mulberry tree phenotypic feature data; Step S23: Estimating the chlorophyll content according to the pre-processed image data and the mulberry tree phenotypic characteristic data to obtain chlorophyll content data; Step S24: performing pest and disease identification on the pre-processed image data to obtain a pest and disease identification result; Step S25: integrating the mulberry tree phenotypic characteristic data, chlorophyll content data and pest and disease identification results into growth health parameters to obtain growth health parameters.

[0012] The present invention can obtain more abundant mulberry tree spectrum information than traditional RGB images by using a multi-spectral camera to collect multi-band image data, including near-infrared and red-edge bands invisible to the human eye. This information is crucial for evaluating the growth status, chlorophyll content, and pests and diseases of mulberry trees, and provides a data basis for subsequent in-depth analysis. Through image preprocessing, including operations such as image stitching, radiation correction, and geometric correction, the effects of noise, distortion, and uneven illumination in the image can be eliminated to improve image quality. On this basis, mulberry tree image features, such as canopy projection area, height, and width, can be extracted to quantitatively describe the growth status of mulberry trees, providing important phenotypic data for subsequent growth health parameter analysis. Chlorophyll content estimation based on preprocessed image data and mulberry tree phenotypic feature data can quickly and non-destructively obtain chlorophyll content information of mulberry leaves, which reflects the photosynthesis capacity and nutritional status of mulberry trees and is an important indicator for evaluating the growth and health of mulberry trees. Using pre-processed image data for pest and disease identification can automatically and quickly detect whether mulberry trees are infected with pests and diseases, and identify the types and severity of pests and diseases, which helps to take timely prevention and control measures, reduce losses, and ensure the healthy growth of mulberry trees. Integrating mulberry tree phenotypic characteristic data, chlorophyll content data, and pest and disease identification results to construct growth health parameters can comprehensively reflect the growth status, nutritional level, and health status of mulberry trees, providing a more comprehensive and reliable basis for subsequent growth stage identification and nutrient demand calculation, and ultimately achieving refined management of mulberry trees.

[0013] Preferably, step S23 includes the following steps: Step S231: extracting the region of interest from the pre-processed image data to obtain mulberry tree canopy image data; extracting the specific band reflectance from the mulberry tree canopy image data to obtain multi-band reflectance data; Step S232: Calculate vegetation index according to multi-band reflectance data to obtain vegetation index data; Step S233: extracting red edge parameters using spectral data near the red light band and near infrared band in the mulberry tree canopy image data to obtain red edge parameter data; Step S234: constructing a spectral feature data set according to the multi-band reflectance data, the vegetation index data, and the red edge parameter data to obtain a spectral feature data set; Step S235: constructing a chlorophyll content estimation model using the spectral feature data set and the mulberry tree phenotypic feature data to obtain a chlorophyll content estimation model; Step S236: input the mulberry tree canopy image data into the chlorophyll content estimation model to perform chlorophyll content estimation to obtain chlorophyll content data.

[0014] The present invention can remove the interference of background information on chlorophyll content estimation and improve the estimation accuracy by extracting mulberry canopy image data. By extracting the reflectance data of a specific band, the band information closely related to the chlorophyll content can be obtained, providing a data basis for subsequent vegetation index calculation and red edge parameter extraction. Calculating vegetation indices, such as NDVI, EVI, etc., can enhance vegetation information, suppress the influence of soil background and other noises, and further improve the accuracy and stability of chlorophyll content estimation. Extracting red edge parameters, such as red edge position, red edge amplitude and red edge width, can capture subtle spectral changes caused by changes in chlorophyll content, provide more sensitive chlorophyll content information, thereby improving the accuracy and reliability of the estimation model. By integrating multi-band reflectance data, vegetation index data and red edge parameter data into a spectral feature data set, a feature set containing rich spectral information can be constructed, providing more comprehensive input data for the chlorophyll content estimation model, thereby improving the prediction ability of the model. By using the spectral feature data set and the phenotypic feature data of mulberry trees to construct a chlorophyll content estimation model, we can combine spectral information and phenotypic information to establish a more accurate and robust estimation model, thereby improving the accuracy and reliability of chlorophyll content estimation. By inputting the mulberry canopy image data into the constructed chlorophyll content estimation model, the chlorophyll content of mulberry leaves can be estimated quickly and efficiently, providing important data support for mulberry growth monitoring and nutrient management.

[0015] Preferably, step S3 comprises the following steps: Step S31: performing data feature preprocessing on the environmental parameter data set and the growth health parameter to obtain preprocessed growth environment data; Step S32: extracting growth stage features from the pre-processed growth environment data to obtain key features of the growth stage; Step S33: acquiring historical growth data and expert knowledge; constructing a growth stage model based on the historical growth data and expert knowledge to obtain a mulberry tree growth stage model; Step S34: inputting the key features of the growth stage into the mulberry tree growth stage discrimination model, performing growth stage discrimination, and obtaining a growth stage discrimination result.

[0016] The present invention can improve the quality and consistency of data by preprocessing environmental parameter data sets and growth health parameters, including operations such as data cleaning, outlier processing and standardization, eliminate the influence of dimensions between different features, and lay a good data foundation for subsequent feature extraction and model training, thereby improving the accuracy of growth stage discrimination. Extracting key features of the growth stage from the preprocessed growth environment data, such as cumulative temperature, average chlorophyll content, canopy growth rate and incidence of pests and diseases, can effectively capture environmental and physiological indicators closely related to the growth stage of mulberry trees, and provide more representative and distinguishing feature inputs for growth stage discrimination. Combining historical growth data and expert knowledge to construct a mulberry tree growth stage model, it is possible to make full use of existing experience and data, establish a discrimination model that is more in line with the growth law of mulberry trees, improve the generalization ability and prediction accuracy of the model, and thus more accurately judge the growth stage of mulberry trees. The extracted key features of the growth stage are input into the trained mulberry tree growth stage discrimination model, and the growth stage of the mulberry tree can be automatically and quickly judged, providing a key basis for subsequent precision fertilization, and realizing intelligent management of the mulberry tree growth process.

[0017] Preferably, step S4 comprises the following steps: Step S41: Calculate the initial nutrient requirement according to the growth stage determination result and the preset basic nutrient requirement model to obtain initial nutrient requirement data; Step S42: extracting environmental factor data and soil data from the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; Step S43: performing environmental factor impact correction analysis based on relevant environmental factor data to obtain environmental factor impact data; Step S44: extracting soil fertility indexes according to relevant soil data to obtain soil fertility index data; performing soil condition impact analysis according to the soil fertility index data to obtain soil condition impact data; Step S45: dynamically adjusting the initial nutrient requirement data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient requirement; Step S46: Generate a nutrient requirement formula according to the growth stage determination result and the basic nutrient requirement to obtain nutrient requirement formula data.

[0018] The present invention can quickly calculate the initial demand of mulberry trees for various nutrient elements in the current growth stage according to the determined growth stage of mulberry trees and the preset nutrient demand model, provide basic data for subsequent nutrient demand adjustment, and ensure that the formulation of fertilization scheme is based on scientific nutrient demand assessment. Extracting relevant environmental factor data and soil data from the environmental parameter data set can provide necessary data support for subsequent analysis of the impact of environmental and soil conditions on nutrient demand, and ensure that the formulation of fertilization scheme fully considers the actual growth environment. By analyzing the impact of environmental factors (such as temperature, light, rainfall, etc.) on nutrient demand and making corrections, the nutrient demand calculation can be made closer to the actual growth environment of mulberry trees, the accuracy and effectiveness of fertilization schemes can be improved, and overfertilization or nutrient deficiency can be avoided. By analyzing the impact of soil fertility indicators (such as soil organic matter content, effective nutrient content, etc.) on nutrient demand, repeated fertilization can be avoided, fertilizer costs can be saved, and fertilization schemes can be adjusted according to the actual conditions of the soil, fertilizer utilization efficiency can be improved, and pollution to the environment can be reduced. By comprehensively considering the impact of environmental factors and soil conditions on nutrient requirements and dynamically adjusting the initial nutrient requirements, a more accurate basic nutrient requirement can be obtained, so that the fertilization plan is more in line with the actual needs of mulberry trees and avoids insufficient or excessive fertilization due to environmental and soil factors. According to the growth stage of the mulberry tree and the final determined basic nutrient requirement, a specific nutrient requirement formula is generated to guide precise fertilization and provide detailed information including fertilizer type, application amount, application time and application method, so as to achieve scientific and efficient nutrient management of mulberry trees.

[0019] Preferably, step S43 includes the following steps: Step S431: cleaning the relevant environmental factor data to obtain cleaned environmental factor data, wherein the cleaned environmental factor data includes temperature data, humidity data, light data and rainfall data; Step S432: determining the daily effective temperature of the temperature data to obtain the daily effective temperature data; calculating the effective accumulated temperature of the daily effective temperature data to obtain the effective accumulated temperature data; Step S433: performing photosynthetically active radiation conversion on the illumination data to obtain photosynthetically active radiation data; performing illumination effective radiation accumulation calculation on the photosynthetically active radiation data to obtain illumination effective radiation accumulation data; Step S434: Calculate the evapotranspiration of the mulberry tree according to the temperature data and the humidity data to obtain evapotranspiration data; calculate the water deficit index according to the rainfall data and the evapotranspiration data to obtain water deficit index data; Step S435: performing grade classification processing on the effective accumulated temperature data, the light effective radiation accumulation data and the water deficit index data respectively to obtain effective accumulated temperature grade data, light effective radiation grade data and water deficit index grade data; Step S436: querying a preset correction coefficient table according to the effective accumulated temperature level data, the light effective radiation level data, and the water deficit index level data to obtain a preliminary nutrient demand correction coefficient; Step S437: Correcting the preliminary nutrient requirement correction coefficient to obtain nutrient requirement correction coefficient data; generating environmental factor impact data based on the nutrient requirement correction coefficient data to obtain environmental factor impact data.

[0020] The present invention can improve the accuracy and reliability of data by cleaning environmental factor data, removing abnormal values ​​and missing values, ensuring that subsequent analysis is performed based on valid data, thereby more accurately evaluating the impact of environmental factors on nutrient demand. By calculating the daily effective temperature and effective accumulated temperature, the impact of temperature on the growth and development of mulberry trees can be quantified, providing a scientific basis for adjusting nutrient demand according to temperature conditions in the future, and making the fertilization scheme more adaptable to the growth rhythm of mulberry trees. By converting the illumination data into photosynthetically active radiation and calculating the cumulative value, the impact of illumination on photosynthesis and nutrient absorption of mulberry trees can be more accurately quantified, thereby more accurately adjusting nutrient demand and improving fertilizer utilization efficiency. By calculating the evapotranspiration of mulberry trees and the water deficit index, the water status of mulberry trees can be quantitatively evaluated, providing a reference for adjusting nutrient demand according to water conditions, and avoiding nutrient absorption disorders caused by water stress. By classifying the environmental factor data, the subsequent correction coefficient query process can be simplified, and the impact of environmental factors can be more easily quantified and applied to nutrient demand adjustment. By consulting the preset correction coefficient table, you can quickly obtain the nutrient demand correction coefficient corresponding to the current environmental conditions, simplify the calculation process, and make the nutrient demand adjustment more operational. Correcting the preliminary correction coefficient can further improve the accuracy of the correction coefficient and make it more in line with the actual situation of specific mulberry varieties and soil types. The precise environmental factor impact data finally generated can more effectively guide the dynamic adjustment of nutrient demand and achieve precision fertilization.

[0021] Preferably, step S44 includes the following steps: Step S441: performing soil data cleaning on relevant soil data to obtain cleaned soil data; extracting main nutrient content from the cleaned soil data to obtain main nutrient content data; Step S442: Calculate the nutrient ratio according to the main nutrient content data to obtain nutrient ratio data; Step S443: Calculating the nutrient balance index according to the nutrient ratio data and the preset ideal nutrient ratio range value to obtain the nutrient balance index; Step S444: obtaining soil standard data; determining soil index weights for the cleaned soil data according to the soil standard data to obtain soil index weight data; Step S445: Calculate the soil fertility comprehensive score according to the cleaned soil data and the soil index weight data to obtain the soil fertility comprehensive score data; Step S446: integrating the nutrient balance index and the soil fertility comprehensive score data into soil fertility index data to obtain soil fertility index data; Step S447: Evaluate the nutrient supply capacity according to the cleaned soil data and the soil fertility index data to obtain soil nutrient supply capacity data; Step S448: Calculate the nutrient demand correction coefficient according to the soil nutrient supply capacity data to obtain soil nutrient demand correction coefficient data; generate soil condition impact data according to the soil nutrient demand correction coefficient data to obtain soil condition impact data.

[0022] By cleaning soil data and extracting the main nutrient content, outliers and noise can be removed to ensure that subsequent analysis is based on reliable data, thereby accurately evaluating soil nutrient status and providing a basis for precision fertilization. Calculating the proportion of main nutrients in the soil can understand the balance of soil nutrients and provide a basis for subsequent adjustment of fertilization plans to achieve balanced nutrient supply. Calculating the nutrient balance index can quantify the balance of soil nutrients and provide an important reference for evaluating soil fertility status and formulating balanced fertilization plans. Obtaining soil standard data and determining soil indicator weights can make soil fertility assessment more scientific and targeted, ensuring that important soil indicators play a greater role in fertility assessment. Calculating the comprehensive score of soil fertility can comprehensively reflect the fertility level of the soil and provide an important basis for evaluating soil nutrient supply capacity and formulating fertilization plans. Integrating the nutrient balance index and the comprehensive score of soil fertility can more comprehensively reflect the soil fertility status, avoid the limitations of a single indicator, and improve the scientific nature of the fertilization plan. By evaluating the soil nutrient supply capacity, we can understand the amount of nutrients that the soil can provide, and provide key information for formulating more accurate fertilization plans to avoid over-fertilization or under-fertilization. By calculating the nutrient demand correction coefficient and soil condition impact data based on the soil nutrient supply capacity, we can adjust the nutrient demand according to the actual soil conditions, avoid repeated fertilization, improve fertilizer utilization efficiency, and ultimately achieve accurate fertilization to promote the healthy growth of mulberry trees.

[0023] Preferably, step S46 includes the following steps: Step S461: querying the basic nutrient prescription corresponding to the growth stage from the preset mulberry tree nutrient requirement database according to the growth stage determination result to obtain the basic nutrient prescription data; Step S462: Calculate the nutrient requirement ratio according to the basic nutrient requirement and the basic nutrient formula data to obtain target nutrient ratio data; Step S463: adjusting the target nutrient ratio data based on the soil fertility index data to obtain adjusted nutrient ratio data; Step S464: using the environmental factor impact data to finely adjust the adjusted nutrient ratio data based on the environmental factors to obtain refined nutrient ratio data; Step S465: Nutrient form selection is performed according to the soil fertility index data, the growth stage determination result and the refined nutrient ratio data to obtain nutrient requirement formula data.

[0024] The present invention can quickly obtain the nutrient ratio recommended in this stage by querying the basic nutrient formula according to the growth stage of mulberry trees, provide a basis for formulating personalized fertilization schemes, and ensure that the fertilization scheme conforms to the growth law of mulberry trees. According to the basic nutrient requirement and the basic nutrient formula, the target nutrient ratio is calculated, and while meeting the total nutrient requirement, a reasonable nutrient ratio can be maintained, nutrient imbalance can be avoided, and the healthy growth of mulberry trees can be promoted. According to the soil fertility index data, the target nutrient ratio is adjusted, and the interaction between nutrients can be considered, certain nutrients can be avoided to be excessive or insufficient, fertilizer utilization efficiency can be improved, and the adverse effects of nutrient antagonism on mulberry growth can be prevented. The nutrient ratio is finely adjusted using environmental factor influence data, so that the fertilization scheme can be more adapted to the current environmental conditions, the absorption and utilization rate of nutrients by mulberry trees can be improved, and the influence of environmental factors on nutrient effectiveness can be reduced. Comprehensively considering the soil fertility index, the growth stage of mulberry trees and the refined nutrient ratio, selecting a suitable nutrient form, the fertilizer utilization rate can be maximized, nutrient loss can be avoided, and the most suitable nutrient form can be provided according to the specific needs of mulberry trees, thereby achieving the best fertilization effect.

[0025] Preferably, the present invention further provides a mulberry garden monitoring system based on the Internet of Things, which is used to execute the mulberry garden monitoring method based on the Internet of Things as described above. The mulberry garden monitoring system based on the Internet of Things comprises: The environment and soil data acquisition module is used to collect environment and soil data of the mulberry garden through the sensor network to obtain the original sensor data; the original sensor data is integrated in time sequence to obtain the environmental parameter data set; The mulberry tree growth status analysis module is used to collect multispectral images of mulberry trees in the mulberry garden through a multispectral camera, and perform image preprocessing to obtain preprocessed image data; estimate the chlorophyll content based on the preprocessed image data to obtain chlorophyll content data; perform growth health parameter analysis based on the preprocessed image data and chlorophyll content data to obtain growth health parameters; The growth stage discrimination module is used to extract growth stage features from the environmental parameter data set and the growth health parameters to obtain key features of the growth stage; the growth stage is discriminated based on the key features of the growth stage to obtain the growth stage judgment result; The nutrient requirement calculation module is used to calculate the initial nutrient requirement according to the growth stage judgment result and the preset basic nutrient requirement model to obtain the initial nutrient requirement data; extract the environmental factor data and soil data of the environmental parameter data set to obtain the relevant environmental factor data and the relevant soil data; perform environmental factor impact correction analysis according to the relevant environmental factor data to obtain the environmental factor impact data; extract the soil fertility index according to the relevant soil data to obtain the soil fertility index data; perform soil condition impact analysis according to the soil fertility index data to obtain the soil condition impact data; dynamically adjust the nutrient requirement of the initial nutrient requirement data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient requirement; generate the nutrient requirement formula according to the growth stage judgment result and the basic nutrient requirement to obtain the nutrient requirement formula data; The fertilization planning module is used to obtain the data of mulberry garden fertilization equipment; fertilization planning is processed according to the nutrient demand formula and the data of mulberry garden fertilization equipment to realize the mulberry tree growth monitoring operation.

[0026] The present invention collects and integrates environmental and soil data in real time through a sensor network, can comprehensively and dynamically monitor the growth environment of a mulberry garden, provide a reliable data basis for subsequent analysis and decision-making, and realize accurate perception of the mulberry garden environment. Utilizing multispectral imaging technology, the phenotypic characteristics, chlorophyll content and pest and disease information of mulberry trees can be quickly and non-destructively obtained, the growth health status of mulberry trees can be comprehensively evaluated, and an important reference basis can be provided for accurate fertilization. By analyzing environmental data and mulberry growth health parameters, the growth stage of mulberry trees can be accurately judged, key guidance can be provided for formulating targeted fertilization plans, and intelligent management of the mulberry growth process can be realized. Taking into account factors such as the growth stage of mulberry trees, environmental factors and soil conditions, the nutrient requirements of mulberry trees can be accurately calculated, and personalized fertilization formulas can be generated to realize accurate fertilization, improve fertilizer utilization efficiency, and reduce pollution to the environment. Combining fertilization equipment data and nutrient requirement formulas for fertilization planning can realize automated and accurate fertilization operations, improve fertilization efficiency, reduce labor costs, and realize continuous monitoring of mulberry growth, ultimately improving the yield and quality of mulberry gardens. Therefore, the present invention provides a mulberry garden monitoring method based on the Internet of Things. By combining sensor technology, multispectral imaging technology and intelligent algorithms, comprehensive, accurate and real-time monitoring of the mulberry garden is achieved, which solves the problems of single monitoring means and low efficiency of existing methods, and provides effective technical means for the refined management of mulberry gardens. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of the steps of a mulberry garden monitoring method based on the Internet of Things; Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart; Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0028] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0029] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0030] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0031] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0032] To achieve this, please refer to Figures 1 to 3 , a mulberry garden monitoring method based on the Internet of Things, comprising the following steps: Step S1: Collect environmental and soil data of the mulberry garden through the sensor network to obtain original sensor data; perform time-sequential data integration on the original sensor data to obtain an environmental parameter data set; Step S2: collecting multispectral images of mulberry trees in the mulberry garden through a multispectral camera, and performing image preprocessing to obtain preprocessed image data; estimating chlorophyll content based on the preprocessed image data to obtain chlorophyll content data; analyzing growth health parameters based on the preprocessed image data and the chlorophyll content data to obtain growth health parameters; Step S3: extracting growth stage features from the environmental parameter data set and the growth health parameters to obtain key features of the growth stage; performing growth stage discrimination based on the key features of the growth stage to obtain a growth stage discrimination result; Step S4: performing initial nutrient demand calculation according to the growth stage determination result and the preset basic nutrient demand model to obtain initial nutrient demand data; performing environmental factor data extraction and soil data extraction on the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; performing environmental factor impact correction analysis according to the relevant environmental factor data to obtain environmental factor impact data; performing soil fertility index extraction according to the relevant soil data to obtain soil fertility index data; performing soil condition impact analysis according to the soil fertility index data to obtain soil condition impact data; dynamically adjusting the nutrient demand of the initial nutrient demand data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient demand; generating a nutrient demand formula according to the growth stage determination result and the basic nutrient demand to obtain the nutrient demand formula data; Step S5: Acquire the data of mulberry garden fertilization equipment; perform fertilization planning according to the nutrient requirement formula and the data of mulberry garden fertilization equipment to realize the mulberry tree growth monitoring operation.

[0033] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of the mulberry garden monitoring method based on the Internet of Things of the present invention. In this example, the mulberry garden monitoring method based on the Internet of Things includes the following steps: Step S1: Collect environmental and soil data of the mulberry garden through the sensor network to obtain original sensor data; perform time-sequential data integration on the original sensor data to obtain an environmental parameter data set; In the embodiment of the present invention, the sensor network deployed in the mulberry garden collects environmental and soil data such as air temperature, humidity, light intensity, soil temperature and humidity, pH value, and conductivity in real time. The data collected by the sensor nodes are accompanied by timestamps and location information and uploaded to the cloud server via a wireless network. On the cloud server, the raw data is pre-processed by cleaning, interpolation, and normalization, and then integrated according to the time sequence and location information to form an environmental parameter data set.

[0034] Step S2: collecting multispectral images of mulberry trees in the mulberry garden through a multispectral camera, and performing image preprocessing to obtain preprocessed image data; estimating chlorophyll content based on the preprocessed image data to obtain chlorophyll content data; analyzing growth health parameters based on the preprocessed image data and the chlorophyll content data to obtain growth health parameters; In an embodiment of the present invention, a multispectral camera is used to carry out aerial photography of a mulberry garden to obtain multispectral image data. The image is preprocessed by splicing, radiation correction, geometric correction, etc., and then the canopy area of ​​the mulberry tree is extracted to calculate the phenotypic characteristics such as canopy area and height. The chlorophyll content is calculated using the preprocessed image data and the established estimation model. Finally, the phenotypic characteristic data, chlorophyll content data, and pest and disease identification results are integrated to form growth health parameters.

[0035] Step S3: extracting growth stage features from the environmental parameter data set and the growth health parameters to obtain key features of the growth stage; performing growth stage discrimination based on the key features of the growth stage to obtain a growth stage discrimination result; In the embodiment of the present invention, the environmental parameter data set and the growth health parameter are preprocessed, including data cleaning, standardization, etc. Then, key features related to the growth stage are extracted, such as cumulative temperature, average chlorophyll content, canopy growth rate, etc. The pre-trained mulberry tree growth stage discrimination model is used to judge the current growth stage of the mulberry tree (e.g., budding stage, young leaf stage, mature stage, dormant stage) according to the extracted key features.

[0036] Step S4: performing initial nutrient demand calculation according to the growth stage determination result and the preset basic nutrient demand model to obtain initial nutrient demand data; performing environmental factor data extraction and soil data extraction on the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; performing environmental factor impact correction analysis according to the relevant environmental factor data to obtain environmental factor impact data; performing soil fertility index extraction according to the relevant soil data to obtain soil fertility index data; performing soil condition impact analysis according to the soil fertility index data to obtain soil condition impact data; dynamically adjusting the nutrient demand of the initial nutrient demand data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient demand; generating a nutrient demand formula according to the growth stage determination result and the basic nutrient demand to obtain the nutrient demand formula data; In the embodiment of the present invention, according to the mulberry tree growth stage, the basic nutrient requirement model is queried from the database to calculate the initial nutrient requirement. Then, environmental factor data and soil data such as temperature, illumination, rainfall are extracted from the environmental parameter data set, the influence of environmental factors and soil conditions on the nutrient requirement is analyzed respectively, and the initial nutrient requirement is corrected. Finally, according to the corrected nutrient requirement and the mulberry tree growth stage, the nutrient requirement formula data including information such as fertilizer type, application amount, and application time are generated.

[0037] Step S5: obtaining data of mulberry garden fertilization equipment; performing fertilization planning according to the nutrient requirement formula and the data of mulberry garden fertilization equipment to realize the mulberry tree growth monitoring operation; In the embodiment of the present invention, data of mulberry garden fertilization equipment is obtained, such as the model of the fertilizer machine, fertilization speed, fertilization width, fertilizer box capacity, etc. Combined with the generated nutrient requirement formula data, fertilization planning processing is performed, such as planning fertilization routes, calculating fertilization amounts, controlling operating parameters of fertilization equipment, etc., ultimately achieving precise fertilization and continuous monitoring of mulberry tree growth.

[0038] Preferably, step S1 comprises the following steps: Step S11: deploying a sensor network according to the terrain and area of ​​the mulberry garden, and positioning the sensors to obtain sensor location information; Step S12: collecting environment and soil data according to the sensor location information and a preset collection time interval to obtain raw sensor data; Step S13: performing data preprocessing on the original sensor data to obtain preprocessed sensor data; Step S14: Performing time-sequential data integration on the pre-processed sensor data according to the sensor location information to obtain an environmental parameter data set.

[0039] In an embodiment of the present invention, a drone equipped with an RTK (Real-Time Kinematic) GPS module is first used to perform low-altitude aerial photography and mapping of the mulberry garden to obtain high-precision terrain data, including information such as the mulberry garden boundary, area, slope, and elevation. Then, according to the mapping data, the mulberry garden is divided into square grids with a side length of 10 meters. A sensor node is deployed at the center point of each grid, and each node integrates an air temperature and humidity sensor, a light sensor, a soil temperature and humidity sensor, a soil PH sensor, and a soil EC sensor (conductivity). Each sensor node has a built-in GPS module for obtaining accurate geographic location information (latitude and longitude). The location information of the sensor node will be uploaded to the cloud server to establish a sensor location information database. If the terrain of the mulberry garden is complex, such as if there is a significant slope change, the grid size and sensor node density will be adjusted according to the actual situation to ensure the representativeness of data collection, such as increasing the sensor node density in areas with larger slopes.

[0040] Each sensor node collects data at a preset time interval (for example, once every hour). For example, the collection time is set to every hour. After the preset collection time point is reached, the sensors of each sensor node start working at the same time to collect the corresponding environment and soil data. The air temperature and humidity sensor collects the current air temperature and humidity; the light sensor collects the current light intensity; the soil temperature and humidity sensor collects the temperature and humidity of the soil; the soil PH sensor collects the acidity and alkalinity of the soil; the soil EC sensor collects the electrical conductivity of the soil. Each collected data is accompanied by a timestamp and the corresponding sensor node ID. The collected data is transmitted to the cloud server in real time via a wireless network (such as NB-IoT narrowband Internet of Things) and stored in the database to form raw sensor data.

[0041] After the raw sensor data is uploaded to the cloud server, it needs to be preprocessed to improve the data quality. First, the data is cleaned and outliers are removed. For example, a reasonable numerical range is set for each sensor, and values ​​outside the range will be identified as outliers and removed. At the same time, the Kalman filter algorithm is used to remove noise data caused by sensor fluctuations or environmental interference. Secondly, missing data is processed. If the data of a sensor node is missing at a certain time point, linear interpolation is used to supplement it based on the data of adjacent time points. Finally, data normalization is performed. Since different sensors have different measurement units and data ranges, in order to facilitate subsequent data analysis and modeling, the data needs to be normalized, such as using the Min-Max normalization method to scale all data to a range of 0 to 1.

[0042] The preprocessed sensor data is integrated according to the sensor location information and time sequence. First, the data collected by each sensor node is grouped according to the sensor node ID. Then, within each group, the data is sorted according to the timestamp. Finally, the data of all sensor nodes are merged into a complete data set in time sequence to form an environmental parameter data set. This data set contains the environmental and soil data collected by all sensor nodes at each time point, such as air temperature, air humidity, light intensity, soil temperature, soil humidity, soil pH, soil conductivity, and records the time and location information corresponding to each data point.

[0043] Preferably, step S2 comprises the following steps: Step S21: collecting multispectral images of mulberry trees in the mulberry garden using a multispectral camera to obtain a multispectral image data set; Step S22: performing image preprocessing on the multispectral image data set to obtain preprocessed image data; performing mulberry tree image feature extraction on the preprocessed image data to obtain mulberry tree phenotypic feature data; Step S23: Estimating the chlorophyll content according to the pre-processed image data and the mulberry tree phenotypic characteristic data to obtain chlorophyll content data; Step S24: performing pest and disease identification on the pre-processed image data to obtain a pest and disease identification result; Step S25: integrating the mulberry tree phenotypic characteristic data, chlorophyll content data and pest and disease identification results into growth health parameters to obtain growth health parameters.

[0044] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes: Step S21: collecting multispectral images of mulberry trees in the mulberry garden using a multispectral camera to obtain a multispectral image data set; In an embodiment of the present invention, a multispectral camera is mounted on a drone equipped with a GPS module to take aerial photos of a mulberry garden. The multispectral camera is set to collect spectral information of five bands: red light (R), green light (G), blue light (B), near infrared (NIR), and red edge (RE). The drone flies according to a pre-planned route to ensure coverage of the entire mulberry garden, and takes images with an overlap of 20% for subsequent image stitching. Each time a shot is taken, the GPS module records the position and attitude information of the drone, and stores it in association with the corresponding multispectral image. All collected multispectral images, together with their corresponding GPS information, constitute a multispectral image dataset.

[0045] Step S22: performing image preprocessing on the multispectral image data set to obtain preprocessed image data; performing mulberry tree image feature extraction on the preprocessed image data to obtain mulberry tree phenotypic feature data; In the embodiment of the present invention, first, the GPS information of the collected image is used to splice the overlapping multispectral images into a complete multispectral image of the mulberry garden through an image stitching algorithm. Then, radiation correction is performed to eliminate the influence of sensor error and uneven illumination. The DN value of the image is converted into a reflectance value using a reflectivity correction method based on a reference plate. Next, geometric correction is performed to correct the geometric distortion of the image using ground control points, and the image is projected into a unified coordinate system. After completing the above preprocessing, preprocessed image data is obtained. Next, the normalized difference vegetation index (NDVI) is used to perform threshold segmentation on the preprocessed image data to extract the mulberry canopy area. Then, based on the extracted mulberry canopy area, the phenotypic characteristics such as the canopy projection area, canopy height, and canopy width of the mulberry tree are calculated, and these characteristic data are stored to obtain the mulberry phenotypic characteristic data.

[0046] Step S23: Estimating the chlorophyll content according to the pre-processed image data and the mulberry tree phenotypic characteristic data to obtain chlorophyll content data; In an embodiment of the present invention, based on the pre-processed image data, the reflectance values ​​of the red light (R) and near infrared (NIR) bands of the mulberry tree canopy area are extracted. Then, the chlorophyll content estimation model established (for example, a chlorophyll content estimation model based on the red edge position: Chlorophyll content=a*Red Edge Position+b, where a and b are model parameters, obtained by fitting the measured data) is used to calculate the chlorophyll content of the mulberry tree in combination with the extracted red light and near infrared band reflectance values. The calculated chlorophyll content data of each mulberry tree is stored to obtain chlorophyll content data.

[0047] Step S24: performing pest and disease identification on the pre-processed image data to obtain a pest and disease identification result; In an embodiment of the present invention, based on the pre-processed image data, a trained mulberry pest and disease recognition model (for example, a pest and disease recognition model based on a convolutional neural network) is used to identify pests and diseases on mulberry images. The model can identify a variety of common mulberry pests and diseases, such as mulberry blight, mulberry brown spot, mulberry borer, etc. The model outputs information such as the type and severity of the pest and disease of each mulberry tree, and stores this information to obtain a pest and disease recognition result.

[0048] Step S25: integrating the mulberry tree phenotypic characteristic data, chlorophyll content data and pest and disease identification results into growth health parameters to obtain growth health parameters; In the embodiment of the present invention, mulberry tree phenotypic characteristic data (such as canopy projection area, canopy height, canopy width), chlorophyll content data and pest and disease identification results are integrated to construct a comprehensive mulberry tree growth health parameter data set. This data set contains various growth indicators and health information of each mulberry tree, which is used for subsequent growth stage discrimination and nutrient requirement calculation. For example, chlorophyll content, canopy area, pest and disease severity and other indicators can be weighted summed to obtain a comprehensive growth health score as a mulberry tree growth health parameter.

[0049] Preferably, step S23 includes the following steps: Step S231: extracting the region of interest from the pre-processed image data to obtain mulberry tree canopy image data; extracting the specific band reflectance from the mulberry tree canopy image data to obtain multi-band reflectance data; Step S232: Calculate vegetation index according to multi-band reflectance data to obtain vegetation index data; Step S233: extracting red edge parameters using spectral data near the red light band and near infrared band in the mulberry tree canopy image data to obtain red edge parameter data; Step S234: constructing a spectral feature data set according to the multi-band reflectance data, the vegetation index data, and the red edge parameter data to obtain a spectral feature data set; Step S235: constructing a chlorophyll content estimation model using the spectral feature data set and the mulberry tree phenotypic feature data to obtain a chlorophyll content estimation model; Step S236: input the mulberry tree canopy image data into the chlorophyll content estimation model to perform chlorophyll content estimation to obtain chlorophyll content data.

[0050] In an embodiment of the present invention, an existing mulberry canopy segmentation model (for example, a U-Net model based on deep learning) is used to perform pixel-level classification on the preprocessed multispectral image, and each pixel in the image is marked as "mulberry canopy" or "background". Then, according to the classification result, all pixels marked as "mulberry canopy" are extracted to generate mulberry canopy image data. From the segmented mulberry canopy image data, the reflectance data of a pre-set specific band is extracted. For example, the reflectance values ​​of five bands of red light (R), green light (G), blue light (B), near infrared (NIR) and red edge (RE) are extracted. The multiple band reflectance values ​​of each pixel are stored as a vector, and the reflectance vectors of all pixels constitute multi-band reflectance data.

[0051] Based on the extracted multi-band reflectance data, a series of vegetation indices are calculated, such as the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Greenness Index (GI), etc. For example, the calculation formula of NDVI is: (NIR-R) / (NIR+R). The vegetation index values ​​calculated for each pixel are stored as a vector, and the vegetation index vectors of all pixels constitute the vegetation index data.

[0052] The spectral reflectance data near the red band (R) and the near infrared band (NIR) in the mulberry canopy image data are curve fitted, for example, using a Gaussian function or an inverse Gaussian function for fitting. Then, according to the fitting curve, red edge parameters such as red edge position (REP), red edge amplitude (REA) and red edge width (REW) are extracted. The red edge parameters of each pixel are stored as a vector, and the red edge parameter vectors of all pixels constitute the red edge parameter data.

[0053] The multi-band reflectance data, vegetation index data, and red edge parameter data of each pixel are combined into a feature vector. The feature vectors of all pixels form a spectral feature dataset. This dataset contains rich spectral information of the mulberry canopy and is used to build a subsequent chlorophyll content estimation model.

[0054] Collect the measured chlorophyll content data of a certain number of mulberry samples, as well as the corresponding spectral feature data and phenotypic feature data. Use these data to build a chlorophyll content estimation model. For example, machine learning algorithms such as multivariate linear regression, support vector machine (SVM), and random forest can be used for model training. The input of the model is spectral features and phenotypic features, and the output is chlorophyll content. Through training, a chlorophyll content estimation model is obtained.

[0055] The mulberry canopy image data is input into the trained chlorophyll content estimation model, and the model predicts the chlorophyll content of each pixel based on the spectral characteristics and phenotypic characteristics of the image. The chlorophyll content prediction values ​​of all pixels are stored to obtain the chlorophyll content data.

[0056] Preferably, step S3 comprises the following steps: Step S31: performing data feature preprocessing on the environmental parameter data set and the growth health parameter to obtain preprocessed growth environment data; Step S32: extracting growth stage features from the pre-processed growth environment data to obtain key features of the growth stage; Step S33: acquiring historical growth data and expert knowledge; constructing a growth stage model based on the historical growth data and expert knowledge to obtain a mulberry tree growth stage model; Step S34: inputting the key features of the growth stage into the mulberry tree growth stage discrimination model, performing growth stage discrimination, and obtaining a growth stage discrimination result.

[0057] In an embodiment of the present invention, environmental parameter data sets (including air temperature, humidity, light, soil temperature and humidity, PH, EC, etc.) and growth health parameters (such as chlorophyll content, canopy area, severity of pests and diseases, etc.) are integrated. First, the data is cleaned to remove outliers and missing values. For example, a box plot method is used to identify and remove outliers in environmental parameters and growth health parameters; and a linear interpolation method is used to fill missing values. Then, the data is standardized, for example, using a Z-score standardization method to convert the data into a distribution with a mean of 0 and a standard deviation of 1, eliminating the influence of dimensions between different features. Finally, the preprocessed environmental parameter data and growth health parameters are merged to form preprocessed growth environment data.

[0058] Extract key features related to the growth stage of mulberry trees from the preprocessed growth environment data. These features can include: Cumulative temperature: Calculate the cumulative value of the daily average temperature from the mulberry budding period to the current time.

[0059] Average Chlorophyll Content: Calculate the average chlorophyll content of the current mulberry trees.

[0060] Canopy growth rate: Calculate the rate of change of mulberry canopy area over a period of time.

[0061] Incidence of pests and diseases: Statistics on the current occurrence of mulberry pests and diseases.

[0062] The extracted key features are combined into a feature vector as the key features in the growth stage.

[0063] Collect environmental parameter data, growth health parameters and corresponding growth stage records (such as budding period, young leaf period, maturity period, dormancy period, etc.) of different mulberry varieties at different growth stages over the years. Combine the experience and knowledge of mulberry planting experts to determine the key feature range of each growth stage. For example, experts can judge based on experience that when the cumulative temperature reaches a certain value and the chlorophyll content is within a specific range, the mulberry tree is in the young leaf stage. Use these historical data and expert knowledge to build a mulberry growth stage model. For example, a decision tree, support vector machine (SVM) or other classification algorithms can be used, with key features of the growth stage as input and the mulberry growth stage as output, to train a mulberry growth stage discrimination model.

[0064] The application of expert knowledge is as follows: Determination of key characteristic range: Experts determine the key characteristic range of different growth stages based on experience. For example, experts believe that "when the cumulative temperature reaches 300-500℃ and the average chlorophyll content is between 35-50 SPAD, the mulberry tree is in the young leaf stage."

[0065] Standards for dividing growth stages: Experts divide the growth stages of mulberry trees (such as budding stage, young leaf stage, maturity stage, dormancy stage, etc.) based on the tree's morphology, physiological characteristics and production needs.

[0066] Feature weight allocation: Experts judge the relative importance of different features for growth stage discrimination and assign corresponding weights.

[0067] Expert knowledge is quantified through rule base: expert experience is summarized into "if-then" rules. For example, "if the cumulative temperature is greater than 300℃ and the average chlorophyll content is greater than 35 SPAD, then the mulberry tree is in the young leaf stage." By consulting books, journals and research reports related to mulberry planting, collecting the experience and rules summarized by experts, the following expert knowledge is finally obtained: Germination period: When the daily average temperature is stable at above 10℃, the buds begin to sprout and swell.

[0068] Young leaf stage: The cumulative temperature reaches 300-500℃, the average chlorophyll content is between 35-50 SPAD, and the canopy growth rate is relatively fast.

[0069] Maturity period: The cumulative temperature reaches 800-1200℃, the average chlorophyll content reaches its peak, and the canopy growth rate slows down.

[0070] Dormant period: When the average daily temperature is below 5℃, the leaves begin to fall off.

[0071] Then, a decision tree model was used to transform the above expert knowledge into rules, and the rules were trained with historical growth data to obtain a mulberry tree growth stage discrimination model.

[0072] The key features of the growth stage extracted in step S32 are input into the trained mulberry tree growth stage discrimination model. The model predicts the current growth stage of the mulberry tree, such as the budding stage, young leaf stage, mature stage or dormant stage, based on the input feature vector. The output of the model is the growth stage judgment result.

[0073] Preferably, step S4 comprises the following steps: Step S41: Calculate the initial nutrient requirement according to the growth stage determination result and the preset basic nutrient requirement model to obtain initial nutrient requirement data; Step S42: extracting environmental factor data and soil data from the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; Step S43: performing environmental factor impact correction analysis based on relevant environmental factor data to obtain environmental factor impact data; Step S44: extracting soil fertility indexes according to relevant soil data to obtain soil fertility index data; performing soil condition impact analysis according to the soil fertility index data to obtain soil condition impact data; Step S45: dynamically adjusting the initial nutrient requirement data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient requirement; Step S46: Generate a nutrient requirement formula according to the growth stage determination result and the basic nutrient requirement to obtain nutrient requirement formula data.

[0074] As an example of the present invention, refer to Figure 3 As shown, in this example, step S4 includes: Step S41: Calculate the initial nutrient requirement according to the growth stage determination result and the preset basic nutrient requirement model to obtain initial nutrient requirement data; In an embodiment of the present invention, according to the growth stage of the mulberry tree determined in step S3 (e.g., budding stage, young leaf stage, mature stage, dormant stage), a preset basic nutrient requirement model database is queried. The database stores the requirements of different mulberry varieties for major nutrients such as nitrogen (N), phosphorus (P), and potassium (K) at different growth stages. The basic nutrient requirement model can be a regression model fitted according to historical data. For example, assuming that the mulberry tree is in the young leaf stage, the nitrogen, phosphorus, and potassium requirements corresponding to the young leaf stage are retrieved from the database, such as N: 10kg / mu, P: 5kg / mu, K: 8kg / mu. These values ​​are the initial nutrient requirement data. Step S42: Extract environmental factor data and soil data from the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; In the embodiment of the present invention, environmental factor data related to nutrient requirements are extracted from the environmental parameter data set, such as temperature, light, rainfall, air humidity, and soil temperature and humidity data. At the same time, soil data are extracted, including soil pH value, electrical conductivity (EC), and nitrogen, phosphorus, and potassium content in the soil.

[0075] Step S43: performing environmental factor impact correction analysis based on relevant environmental factor data to obtain environmental factor impact data; In an embodiment of the present invention, the influence of environmental factors on nutrient demand is analyzed. For example, high temperature will accelerate nutrient consumption, and it is necessary to increase nutrient supply; sufficient light is conducive to photosynthesis, thereby improving the efficiency of nutrient absorption; rainfall will affect the leaching loss of soil nutrients. According to the influence of these environmental factors, the initial nutrient demand data is corrected. The correction method can be a regression model established based on historical data. For example, if the rainfall is large, the initial nutrient demand data is adjusted upward according to the preset leaching loss rate. The correction amount of environmental factors on nutrient demand is recorded as environmental factor impact data. Step S44: Extract soil fertility indicators based on relevant soil data to obtain soil fertility index data; perform soil condition impact analysis based on the soil fertility index data to obtain soil condition impact data; In an embodiment of the present invention, soil fertility indicators, such as soil organic matter content, effective nitrogen, phosphorus and potassium content, cation exchange capacity (CEC), etc., are extracted based on soil data. Then, the impact of soil fertility indicators on nutrient demand is analyzed. For example, if the effective nitrogen content in the soil is high, the amount of nitrogen fertilizer applied can be reduced. Based on the soil fertility indicators, the initial nutrient demand data is corrected. The correction method can be based on the rules of the soil fertility evaluation system or a regression model established based on historical data. The correction amount of soil conditions on nutrient demand is recorded as soil condition impact data.

[0076] Step S45: dynamically adjusting the initial nutrient requirement data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient requirement; In the embodiment of the present invention, the environmental factor impact data and the soil condition impact data are superimposed on the initial nutrient requirement data to obtain the final basic nutrient requirement. For example, if the initial nitrogen requirement is 10 kg / mu, the environmental factor impact data is +1 kg / mu, and the soil condition impact data is -2 kg / mu, then the final basic nitrogen requirement is 10+1-2 = 9 kg / mu.

[0077] Step S46: generating a nutrient requirement formula according to the growth stage determination result and the basic nutrient requirement to obtain nutrient requirement formula data; In the embodiment of the present invention, specific nutrient requirement formula data is generated according to the growth stage of the mulberry tree and the basic nutrient requirement. The formula data includes information such as the type, application amount and application time of the required fertilizer. For example, according to the basic nutrient requirement, the specific application amount of fertilizers such as urea and potassium dihydrogen phosphate required is calculated, and the fertilization time is determined according to the growth stage of the mulberry tree.

[0078] Preferably, step S43 includes the following steps: Step S431: cleaning the relevant environmental factor data to obtain cleaned environmental factor data, wherein the cleaned environmental factor data includes temperature data, humidity data, light data and rainfall data; Step S432: determining the daily effective temperature of the temperature data to obtain the daily effective temperature data; calculating the effective accumulated temperature of the daily effective temperature data to obtain the effective accumulated temperature data; Step S433: performing photosynthetically active radiation conversion on the illumination data to obtain photosynthetically active radiation data; performing illumination effective radiation accumulation calculation on the photosynthetically active radiation data to obtain illumination effective radiation accumulation data; Step S434: Calculate the evapotranspiration of the mulberry tree according to the temperature data and the humidity data to obtain evapotranspiration data; calculate the water deficit index according to the rainfall data and the evapotranspiration data to obtain water deficit index data; Step S435: performing grade classification processing on the effective accumulated temperature data, the light effective radiation accumulation data and the water deficit index data respectively to obtain effective accumulated temperature grade data, light effective radiation grade data and water deficit index grade data; Step S436: querying a preset correction coefficient table according to the effective accumulated temperature level data, the light effective radiation level data, and the water deficit index level data to obtain a preliminary nutrient demand correction coefficient; Step S437: Correcting the preliminary nutrient requirement correction coefficient to obtain nutrient requirement correction coefficient data; generating environmental factor impact data based on the nutrient requirement correction coefficient data to obtain environmental factor impact data.

[0079] In an embodiment of the present invention, the temperature, light and rainfall data extracted from the environmental parameter data set are cleaned. The box plot method is used to identify and remove outliers. For example, the reasonable range of temperature data is set to -10°C to 50°C, and data outside this range is regarded as outliers and removed. Missing data is filled using linear interpolation. The cleaned data includes hourly temperature, light intensity and rainfall data.

[0080] The biological zero of mulberry trees is set to 10°C. The daily effective temperature is calculated by adding up the positive values ​​of the difference between the daily hourly temperature and the biological zero to get the effective temperature of the day. For example, if the hourly temperature value of a certain day is [12,15,18,...,8], then the effective temperature of the day is (12-10)+(15-10) +(18-10)+... +0 = the effective temperature value of the day. The effective temperature of each day is added up to get the effective accumulated temperature. For example, starting from the budding period of mulberry trees, the effective temperature of each day is added up to get the effective accumulated temperature from the budding period to the present.

[0081] Convert the light intensity data into photosynthetically active radiation (PAR) data. The conversion method uses an empirical formula or a table lookup method to convert the light intensity value into the corresponding PAR value. For example, a linear relationship can be used for conversion: PAR=a*light intensity+b, where a and b are coefficients obtained through experimental measurement. Add up the PAR values ​​every hour of the day to get the daily cumulative amount of light effective radiation. Then add up the daily cumulative amount of light effective radiation to get the cumulative data of light effective radiation from the budding period of the mulberry tree to the present.

[0082] Use the Penman-Monteith formula or other evapotranspiration calculation models, combined with temperature, humidity, wind speed and radiation data, to calculate the evapotranspiration of mulberry trees. Compare the daily evapotranspiration with the daily rainfall to calculate the water deficit index. Water deficit index = (evapotranspiration - rainfall) / evapotranspiration. This index reflects the degree of water deficit of mulberry trees.

[0083] According to the preset thresholds, the effective accumulated temperature, the accumulated effective radiation of light and the water deficit index are divided into different levels. For example, the effective accumulated temperature is divided into three levels: low, medium and high; the accumulated effective radiation of light is divided into three levels: low, medium and high; and the water deficit index is divided into three levels: mild deficit, moderate deficit and severe deficit.

[0084] According to the classification level, the corresponding nutrient demand correction coefficient is queried from the preset correction coefficient table. The correction coefficient table is a three-dimensional table, with the effective accumulated temperature level, light effective radiation level and water deficit index level as indexes, and each cell in the table stores the corresponding nutrient demand correction coefficient. For example, if the effective accumulated temperature level is "medium", the light effective radiation level is "high", and the water deficit index level is "mild deficit", the corresponding correction coefficient is queried from the table.

[0085] Correct the preliminary nutrient requirement correction coefficient found in the correction coefficient table. The correction method can take into account the influence of factors such as mulberry variety and soil type. For example, different mulberry varieties have different sensitivities to environmental factors, and the correction coefficient needs to be fine-tuned. The correction process can be based on expert experience, or a machine learning algorithm can be used to learn and adjust based on historical data. Finally, accurate nutrient requirement correction coefficient data is obtained.

[0086] Apply the nutrient demand correction coefficient data to the initial nutrient demand data to calculate the impact of environmental factors on nutrient demand, that is, the environmental factor impact data. The calculation method is: environmental factor impact data = initial nutrient demand data * nutrient demand correction coefficient. For example, assuming that the initial nitrogen demand of mulberry trees is 10kg / mu, and the corrected nitrogen demand correction coefficient is 0.1 (indicating that environmental factors increase nitrogen demand by 10%), the environmental factor impact data is 10kg / mu*0.1=1kg / mu. This means that due to the influence of environmental factors, an additional 1 kg / mu of nitrogen fertilizer is required. Calculate the environmental factor impact data of nutrients such as nitrogen, phosphorus, and potassium separately.

[0087] Preferably, step S44 includes the following steps: Step S441: performing soil data cleaning on relevant soil data to obtain cleaned soil data; extracting main nutrient content from the cleaned soil data to obtain main nutrient content data; Step S442: Calculate the nutrient ratio according to the main nutrient content data to obtain nutrient ratio data; Step S443: Calculating the nutrient balance index according to the nutrient ratio data and the preset ideal nutrient ratio range value to obtain the nutrient balance index; Step S444: obtaining soil standard data; determining soil index weights for the cleaned soil data according to the soil standard data to obtain soil index weight data; Step S445: Calculate the soil fertility comprehensive score according to the cleaned soil data and the soil index weight data to obtain the soil fertility comprehensive score data; Step S446: integrating the nutrient balance index and the soil fertility comprehensive score data into soil fertility index data to obtain soil fertility index data; Step S447: Evaluate the nutrient supply capacity according to the cleaned soil data and the soil fertility index data to obtain soil nutrient supply capacity data; Step S448: Calculate the nutrient demand correction coefficient according to the soil nutrient supply capacity data to obtain soil nutrient demand correction coefficient data; generate soil condition impact data according to the soil nutrient demand correction coefficient data to obtain soil condition impact data.

[0088] In an embodiment of the present invention, the collected soil data (e.g., soil pH value, electrical conductivity, nitrogen, phosphorus, potassium content, organic matter content, etc.) is cleaned. Outliers are removed using a preset reasonable range. For example, the reasonable range of soil pH value is usually between 4 and 9, and values ​​outside this range are considered outliers and removed. Missing values ​​are filled using linear interpolation. The cleaned data is the cleaned soil data. The main nutrient content data, such as the content of nitrogen (N), phosphorus (P), and potassium (K), are extracted from the cleaned soil data.

[0089] According to the extracted nitrogen (N), phosphorus (P), and potassium (K) content, the ratio of the three is calculated, for example, N:P:K=1:0.5:0.8. This ratio data reflects the relative content of the three main nutrients in the soil.

[0090] An ideal nutrient ratio range is set in advance, such as N:P:K = 1:0.4-0.6:0.6-0.9. The nutrient balance index is calculated based on the degree of deviation between the calculated nutrient ratio data and the ideal nutrient ratio range. For example, the degree of deviation can be calculated using vector distance or other similarity measurement methods, and the degree of deviation can be converted into a nutrient balance index between 0 and 1, where 1 represents complete balance and 0 represents extreme imbalance.

[0091] Obtain soil standard data, such as national or local soil nutrient classification standards. The standards specify the threshold ranges of various indicators for different grades of soil. Determine the weights of different soil indicators based on the soil standard data. For example, expert scoring or analytic hierarchy process (AHP) can be used to determine the weights, with more important indicators given higher weights.

[0092] The soil fertility comprehensive score is calculated based on the cleaned soil data and soil index weights. The calculation method can adopt the weighted summation method, which multiplies the value of each soil index by its corresponding weight and then sums them to obtain the soil fertility comprehensive score.

[0093] The nutrient balance index and the soil fertility comprehensive score are integrated to form the final soil fertility index data. For example, the two indicators can be weighted averaged to obtain a comprehensive soil fertility index.

[0094] Based on the cleaned soil data and soil fertility index data, the soil's nutrient supply capacity is evaluated. The evaluation method can use a machine learning model to make predictions based on historical data. For example, the amount of nitrogen, phosphorus and potassium that the soil can provide, i.e. the soil nutrient supply capacity data, can be predicted based on indicators such as the soil's effective nitrogen, phosphorus and potassium content, organic matter content, and cation exchange capacity.

[0095] According to the soil nutrient supply capacity data, calculate the nutrient demand correction coefficient. For example, if the nitrogen that the soil can provide can meet 80% of the mulberry tree's needs, the nitrogen demand correction coefficient is -0.2 (indicating a 20% reduction in nitrogen fertilizer application). Apply the nutrient demand correction coefficient to the initial nutrient demand data to calculate the impact of soil conditions on nutrient demand, that is, the soil condition impact data. The calculation method is: soil condition impact data = initial nutrient demand data * nutrient demand correction coefficient. For example, assuming that the initial nitrogen demand of mulberry trees is 10kg / mu and the nitrogen demand correction coefficient is -0.2, the soil condition impact data is 10kg / mu*-0.2=-2kg / mu. This means that since the soil can already provide some nitrogen, the nitrogen fertilizer application needs to be reduced by 2kg / mu. Calculate the soil condition impact data of nutrients such as nitrogen, phosphorus, and potassium separately. These data will be used for subsequent dynamic adjustment of nutrient demand, and ultimately generate an accurate fertilization plan.

[0096] Preferably, step S46 includes the following steps: Step S461: querying the basic nutrient prescription corresponding to the growth stage from the preset mulberry tree nutrient requirement database according to the growth stage determination result to obtain the basic nutrient prescription data; Step S462: Calculate the nutrient requirement ratio according to the basic nutrient requirement and the basic nutrient formula data to obtain target nutrient ratio data; Step S463: adjusting the target nutrient ratio data based on the soil fertility index data to obtain adjusted nutrient ratio data; Step S464: using the environmental factor impact data to finely adjust the adjusted nutrient ratio data based on the environmental factors to obtain refined nutrient ratio data; Step S465: Nutrient form selection is performed according to the soil fertility index data, the growth stage determination result and the refined nutrient ratio data to obtain nutrient requirement formula data.

[0097] In the embodiment of the present invention, according to the growth stage of the mulberry tree determined in step S3 (e.g., budding stage, young leaf stage, mature stage, dormant stage), the basic nutrient formula of the corresponding growth stage is queried from the preset mulberry tree nutrient requirement database. The database stores the ratios of the main nutrient elements such as nitrogen (N), phosphorus (P), potassium (K) recommended for different growth stages. For example, assuming that the mulberry tree is in the young leaf stage, the basic nutrient formula of the young leaf stage obtained by querying the database is N:P:K=2:1:1.5.

[0098] According to the basic nutrient requirement calculated in step S45 and the basic nutrient formula queried in step S461, the target nutrient ratio data is calculated. For example, assuming that the basic nutrient requirement is N: 9kg / mu, P: 4kg / mu, K: 7kg / mu, and the basic nutrient formula is N: P: K = 2: 1: 1.5, the target nutrient ratio data is calculated to be N: P: K = 9 / 4.5: 4 / 4: 7 / 4.67 ≈ 2: 1: 1.5. This means that under the premise of meeting the total nutrient requirement, the nutrient ratio is kept as close to the basic nutrient formula as possible.

[0099] According to the soil fertility index data obtained in step S44, the target nutrient ratio data is adjusted to take into account the interaction between nutrients. For example, too high potassium content in the soil will inhibit the absorption of magnesium. Therefore, if the soil potassium content is high, it is necessary to appropriately reduce the proportion of potassium fertilizer and increase the application of magnesium fertilizer. Assuming that the soil test results show that the potassium content is high, the potassium ratio in the target nutrient ratio data is adjusted downward, for example, adjusted to N:P:K=2:1:1.2.

[0100] According to the environmental factor impact data obtained in step S43, the adjusted nutrient ratio data is finely adjusted. For example, high temperature will increase nitrogen loss, so under high temperature conditions, the ratio of nitrogen fertilizer needs to be appropriately increased. Assuming that the environmental factor impact data shows that high temperature will cause increased nitrogen loss, the nitrogen ratio in the adjusted nutrient ratio data is increased, for example, adjusted to N:P:K=2.2:1:1.2.

[0101] According to the soil fertility index data, mulberry growth stage and refined nutrient ratio data, the appropriate nutrient form is selected, and the nutrient requirement formula data is finally generated. For example, if the soil pH value is low, nitrate nitrogen fertilizer is selected instead of ammonium nitrogen fertilizer, because ammonium nitrogen fertilizer is easily volatilized and lost in acidic soil. At the same time, the appropriate type of fertilizer is selected according to the growth stage of the mulberry tree. For example, in the young leaf stage, in order to promote rapid growth, quick-acting nitrogen fertilizer can be selected. The final nutrient requirement formula data contains detailed information such as fertilizer type, application amount, application time and application method. For example, the formula data can be: "Apply 15kg / mu of urea, 10kg / mu of potassium dihydrogen phosphate, and 5kg / mu of magnesium sulfate, one month after the mulberry trees sprout, and use broadcasting method."

[0102] Preferably, the present invention further provides a mulberry garden monitoring system based on the Internet of Things, which is used to execute the mulberry garden monitoring method based on the Internet of Things as described above. The mulberry garden monitoring system based on the Internet of Things comprises: The environment and soil data acquisition module is used to collect environment and soil data of the mulberry garden through the sensor network to obtain the original sensor data; the original sensor data is integrated in time sequence to obtain the environmental parameter data set; The mulberry tree growth status analysis module is used to collect multispectral images of mulberry trees in the mulberry garden through a multispectral camera, and perform image preprocessing to obtain preprocessed image data; estimate the chlorophyll content based on the preprocessed image data to obtain chlorophyll content data; perform growth health parameter analysis based on the preprocessed image data and chlorophyll content data to obtain growth health parameters; The growth stage discrimination module is used to extract growth stage features from the environmental parameter data set and the growth health parameters to obtain key features of the growth stage; the growth stage is discriminated based on the key features of the growth stage to obtain the growth stage judgment result; The nutrient requirement calculation module is used to calculate the initial nutrient requirement according to the growth stage judgment result and the preset basic nutrient requirement model to obtain the initial nutrient requirement data; extract the environmental factor data and soil data of the environmental parameter data set to obtain the relevant environmental factor data and the relevant soil data; perform environmental factor impact correction analysis according to the relevant environmental factor data to obtain the environmental factor impact data; extract the soil fertility index according to the relevant soil data to obtain the soil fertility index data; perform soil condition impact analysis according to the soil fertility index data to obtain the soil condition impact data; dynamically adjust the nutrient requirement of the initial nutrient requirement data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient requirement; generate the nutrient requirement formula according to the growth stage judgment result and the basic nutrient requirement to obtain the nutrient requirement formula data; The fertilization planning module is used to obtain the data of mulberry garden fertilization equipment; fertilization planning is processed according to the nutrient demand formula and the data of mulberry garden fertilization equipment to realize the mulberry tree growth monitoring operation.

[0103] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0104] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A mulberry garden monitoring method based on the Internet of Things, characterized in that: The following steps are involved: Step S1: Collect environmental and soil data of the mulberry garden through the sensor network to obtain original sensor data; perform time-sequential data integration on the original sensor data to obtain an environmental parameter data set; Step S2: collecting multispectral images of mulberry trees in the mulberry garden through a multispectral camera, and performing image preprocessing to obtain preprocessed image data; estimating chlorophyll content based on the preprocessed image data to obtain chlorophyll content data; Perform growth health parameter analysis based on pre-processed image data and chlorophyll content data to obtain growth health parameters; Step S3: extracting growth stage features from the environmental parameter data set and growth health parameters to obtain key features of the growth stage; The growth stage is judged according to the key characteristics of the growth stage to obtain the growth stage judgment result; Step S4: performing initial nutrient demand calculation according to the growth stage determination result and the preset basic nutrient demand model to obtain initial nutrient demand data; performing environmental factor data extraction and soil data extraction on the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; Perform environmental factor impact correction analysis based on relevant environmental factor data to obtain environmental factor impact data; perform soil fertility index extraction based on relevant soil data to obtain soil fertility index data; perform soil condition impact analysis based on soil fertility index data to obtain soil condition impact data; dynamically adjust the initial nutrient demand data based on environmental factor impact data and soil condition impact data to obtain basic nutrient demand; generate nutrient demand formula based on growth stage determination results and basic nutrient demand to obtain nutrient demand formula data; Step S5: Acquire the data of mulberry garden fertilization equipment; perform fertilization planning according to the nutrient requirement formula and the data of mulberry garden fertilization equipment to realize the mulberry tree growth monitoring operation.

2. The mulberry garden monitoring method based on the Internet of Things according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: deploying a sensor network according to the terrain and area of ​​the mulberry garden, and positioning the sensors to obtain sensor location information; Step S12: collecting environment and soil data according to the sensor location information and a preset collection time interval to obtain raw sensor data; Step S13: performing data preprocessing on the original sensor data to obtain preprocessed sensor data; Step S14: Performing time-sequential data integration on the pre-processed sensor data according to the sensor location information to obtain an environmental parameter data set.

3. The mulberry garden monitoring method based on the Internet of Things according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: collecting multispectral images of mulberry trees in the mulberry garden using a multispectral camera to obtain a multispectral image data set; Step S22: performing image preprocessing on the multispectral image data set to obtain preprocessed image data; performing mulberry tree image feature extraction on the preprocessed image data to obtain mulberry tree phenotypic feature data; Step S23: Estimating the chlorophyll content according to the pre-processed image data and the mulberry tree phenotypic characteristic data to obtain chlorophyll content data; Step S24: performing pest and disease identification on the pre-processed image data to obtain a pest and disease identification result; Step S25: integrating the mulberry tree phenotypic characteristic data, chlorophyll content data and pest and disease identification results into growth health parameters to obtain growth health parameters.

4. The mulberry garden monitoring method based on the Internet of Things according to claim 3 is characterized in that: Step S23 includes the following steps: Step S231: extracting the region of interest from the pre-processed image data to obtain mulberry tree canopy image data; extracting the specific band reflectance from the mulberry tree canopy image data to obtain multi-band reflectance data; Step S232: Calculate vegetation index according to multi-band reflectance data to obtain vegetation index data; Step S233: extracting red edge parameters using spectral data near the red light band and near infrared band in the mulberry tree canopy image data to obtain red edge parameter data; Step S234: constructing a spectral feature data set according to the multi-band reflectance data, the vegetation index data, and the red edge parameter data to obtain a spectral feature data set; Step S235: constructing a chlorophyll content estimation model using the spectral feature data set and the mulberry tree phenotypic feature data to obtain a chlorophyll content estimation model; Step S236: input the mulberry tree canopy image data into the chlorophyll content estimation model to perform chlorophyll content estimation to obtain chlorophyll content data.

5. The mulberry garden monitoring method based on the Internet of Things according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: performing data feature preprocessing on the environmental parameter data set and the growth health parameter to obtain preprocessed growth environment data; Step S32: extracting growth stage features from the pre-processed growth environment data to obtain key features of the growth stage; Step S33: acquiring historical growth data and expert knowledge; constructing a growth stage model based on the historical growth data and expert knowledge to obtain a mulberry tree growth stage model; Step S34: inputting the key features of the growth stage into the mulberry tree growth stage discrimination model, performing growth stage discrimination, and obtaining a growth stage discrimination result.

6. The mulberry garden monitoring method based on the Internet of Things according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: Calculate the initial nutrient requirement according to the growth stage determination result and the preset basic nutrient requirement model to obtain initial nutrient requirement data; Step S42: extracting environmental factor data and soil data from the environmental parameter data set to obtain relevant environmental factor data and relevant soil data; Step S43: performing environmental factor impact correction analysis based on relevant environmental factor data to obtain environmental factor impact data; Step S44: extracting soil fertility indexes according to relevant soil data to obtain soil fertility index data; performing soil condition impact analysis according to the soil fertility index data to obtain soil condition impact data; Step S45: dynamically adjusting the initial nutrient requirement data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient requirement; Step S46: Generate a nutrient requirement formula according to the growth stage determination result and the basic nutrient requirement to obtain nutrient requirement formula data.

7. The mulberry garden monitoring method based on the Internet of Things according to claim 6 is characterized in that: Step S43 includes the following steps: Step S431: cleaning the relevant environmental factor data to obtain cleaned environmental factor data, wherein the cleaned environmental factor data includes temperature data, humidity data, light data and rainfall data; Step S432: determining the daily effective temperature of the temperature data to obtain the daily effective temperature data; calculating the effective accumulated temperature of the daily effective temperature data to obtain the effective accumulated temperature data; Step S433: performing photosynthetically active radiation conversion on the illumination data to obtain photosynthetically active radiation data; performing illumination effective radiation accumulation calculation on the photosynthetically active radiation data to obtain illumination effective radiation accumulation data; Step S434: Calculate the evapotranspiration of the mulberry tree according to the temperature data and the humidity data to obtain evapotranspiration data; calculate the water deficit index according to the rainfall data and the evapotranspiration data to obtain water deficit index data; Step S435: performing grade classification processing on the effective accumulated temperature data, the light effective radiation accumulation data and the water deficit index data respectively to obtain effective accumulated temperature grade data, light effective radiation grade data and water deficit index grade data; Step S436: querying a preset correction coefficient table according to the effective accumulated temperature level data, the light effective radiation level data, and the water deficit index level data to obtain a preliminary nutrient demand correction coefficient; Step S437: Correcting the preliminary nutrient requirement correction coefficient to obtain nutrient requirement correction coefficient data; generating environmental factor impact data based on the nutrient requirement correction coefficient data to obtain environmental factor impact data.

8. The mulberry garden monitoring method based on the Internet of Things according to claim 6 is characterized in that: Step S44 includes the following steps: Step S441: performing soil data cleaning on relevant soil data to obtain cleaned soil data; extracting main nutrient content from the cleaned soil data to obtain main nutrient content data; Step S442: Calculate the nutrient ratio according to the main nutrient content data to obtain nutrient ratio data; Step S443: Calculating the nutrient balance index according to the nutrient ratio data and the preset ideal nutrient ratio range value to obtain the nutrient balance index; Step S444: obtaining soil standard data; determining soil index weights for the cleaned soil data according to the soil standard data to obtain soil index weight data; Step S445: Calculate the soil fertility comprehensive score according to the cleaned soil data and the soil index weight data to obtain the soil fertility comprehensive score data; Step S446: integrating the nutrient balance index and the soil fertility comprehensive score data into soil fertility index data to obtain soil fertility index data; Step S447: Evaluate the nutrient supply capacity according to the cleaned soil data and the soil fertility index data to obtain soil nutrient supply capacity data; Step S448: Calculate the nutrient demand correction coefficient according to the soil nutrient supply capacity data to obtain soil nutrient demand correction coefficient data; generate soil condition impact data according to the soil nutrient demand correction coefficient data to obtain soil condition impact data.

9. The mulberry garden monitoring method based on the Internet of Things according to claim 6 is characterized in that: Step S46 includes the following steps: Step S461: querying the basic nutrient prescription corresponding to the growth stage from the preset mulberry tree nutrient requirement database according to the growth stage determination result to obtain the basic nutrient prescription data; Step S462: Calculate the nutrient requirement ratio according to the basic nutrient requirement and the basic nutrient formula data to obtain target nutrient ratio data; Step S463: adjusting the target nutrient ratio data based on the soil fertility index data to obtain adjusted nutrient ratio data; Step S464: using the environmental factor impact data to finely adjust the adjusted nutrient ratio data based on the environmental factors to obtain refined nutrient ratio data; Step S465: Nutrient form selection is performed according to the soil fertility index data, the growth stage determination result and the refined nutrient ratio data to obtain nutrient requirement formula data.

10. A mulberry garden monitoring system based on the Internet of Things, characterized in that: Used to execute the mulberry garden monitoring method based on the Internet of Things as claimed in claim 1, the mulberry garden monitoring system based on the Internet of Things comprises: The environment and soil data acquisition module is used to collect environment and soil data of the mulberry garden through the sensor network to obtain the original sensor data; the original sensor data is integrated in time sequence to obtain the environmental parameter data set; The mulberry tree growth status analysis module is used to collect multispectral images of mulberry trees in the mulberry garden through a multispectral camera, and perform image preprocessing to obtain preprocessed image data; estimate the chlorophyll content based on the preprocessed image data to obtain chlorophyll content data; perform growth health parameter analysis based on the preprocessed image data and chlorophyll content data to obtain growth health parameters; The growth stage discrimination module is used to extract growth stage features from the environmental parameter data set and the growth health parameters to obtain key features of the growth stage; the growth stage is discriminated based on the key features of the growth stage to obtain the growth stage judgment result; The nutrient requirement calculation module is used to calculate the initial nutrient requirement according to the growth stage judgment result and the preset basic nutrient requirement model to obtain the initial nutrient requirement data; extract the environmental factor data and soil data of the environmental parameter data set to obtain the relevant environmental factor data and the relevant soil data; perform environmental factor impact correction analysis according to the relevant environmental factor data to obtain the environmental factor impact data; extract the soil fertility index according to the relevant soil data to obtain the soil fertility index data; perform soil condition impact analysis according to the soil fertility index data to obtain the soil condition impact data; dynamically adjust the nutrient requirement of the initial nutrient requirement data according to the environmental factor impact data and the soil condition impact data to obtain the basic nutrient requirement; generate the nutrient requirement formula according to the growth stage judgment result and the basic nutrient requirement to obtain the nutrient requirement formula data; The fertilization planning module is used to obtain the data of mulberry garden fertilization equipment; fertilization planning is processed according to the nutrient demand formula and the data of mulberry garden fertilization equipment to realize the mulberry tree growth monitoring operation.

Images