Crop diagnosis regulation and control model and method based on multi-source data analysis

Abstract

The invention discloses a crop diagnosis regulation and control model and method based on multi-source data analysis, and belongs to the technical field of crop diagnos.The crop diagnosis regulation and control model comprises a multi-source data acquisition layer for acquiring data information related to crop growth from a plurality of data sources; the data processing fusion layer is used for processing and fusing the acquired data information to obtain standard data information; the diagnostic analysis layer performs analysis based on the standard data information and judges whether crop growth is abnormal or not; and the regulation and control processing layer is used for performing analysis based on the acquired standard data information and performing corresponding regulation and control on the crop region with abnormal growth. According to the invention, a multi-source data fusion technology is adopted to carry out water shortage and pest diagnosis on crops, so that the detection accuracy can be improved; and meanwhile, the potential water shortage condition of the crops and the pest outbreak condition can be judged in advance under the condition of no water shortage or slight pest, so that the intervention is performed in advance before the abnormal condition of the crops occurs, and the influence of judgment hysteresis on the crops can be reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of crop diagnosis, and in particular relates to a crop diagnosis and control model and method based on multi-source data analysis. Background Art

[0002] With the growing global population, dwindling arable land resources, and the numerous challenges posed by climate change, increasing crop yields, improving crop quality, and achieving precision agricultural management have become critical issues that need to be addressed in the current agricultural sector. Crop diagnostics and regulation, a core component of precision agriculture, aims to optimize the crop growth environment and improve resource utilization efficiency by accurately identifying the physiological, ecological, and nutritional status of crops during growth and implementing targeted regulatory measures in a timely manner. Ultimately, this approach results in high-yield, high-quality, and efficient crop production.

[0003] Currently, when diagnosing crops, inspectors usually conduct inspections in crop planting areas and then judge the crop growth status based on their experience. This method can only diagnose crops that have already shown obvious abnormalities, so that corresponding treatment measures can be taken, resulting in a certain lag in the implementation of treatment measures. In addition, it is difficult for inspectors to accurately capture subtle changes in crops in the early stages, and thus it is impossible to treat them at the optimal time, which in turn affects crop yield and quality. For example, when watering and fertilizing, most of the time, watering is implemented when the crops are already in a water shortage state, which will cause a lag. In addition, different inspectors have different levels and judgment standards, which makes it difficult to ensure the accuracy and consistency of the diagnostic results. Summary of the Invention

[0004] The purpose of the present invention is to provide a crop diagnosis and control model and method based on multi-source data analysis to solve the problems faced by the above-mentioned background technology.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A crop diagnosis and control model based on multi-source data analysis, the model comprising:

[0007] A multi-source data acquisition layer, wherein the multi-source data acquisition layer is used to obtain data information related to crop growth from multiple data sources;

[0008] A data processing and fusion layer processes and fuses the acquired data information to obtain standard data information;

[0009] A diagnostic analysis layer, which performs analysis based on standard data information to determine whether crop growth is abnormal;

[0010] The regulation and processing layer performs analysis based on the acquired standard data information and performs corresponding regulation on the crop areas with abnormal growth.

[0011] Furthermore, the data information includes weather information, soil information, crop physiological information and crop morphological information.

[0012] Furthermore, the diagnostic analysis layer works as follows:

[0013] Based on standard data information, a crop growth status judgment model is constructed, and the model expression is: Among them, A so is the soil state coefficient, A psy is the physiological index coefficient, A fom is the morphological index coefficient;

[0014] By obtaining the current data information of the crop and inputting it into the constructed crop growth status judgment model A, the current growth status value AR of the crop is obtained. When AR>AR t When the crop needs watering, AR t is the judgment threshold.

[0015] Furthermore, the soil state coefficient A so , physiological index coefficient A psy , morphological index coefficient A fom The acquisition method is:

[0016] The image information of crops is captured by drones, and the leaf curling degree PM, wilting degree PC, and leaf area index PLAI of crops are obtained based on the image information. Obtain the morphological index coefficient A fom , PM0 is the standard value of leaf curling, PC0 is the standard value of wilting, and PLAI0 is the standard value of leaf area index;

[0017] By installing multiple humidity sensors at different depths in the soil, the soil moisture content can be obtained based on the specific gravity at different depths and the humidity sensor values. wr , through the formula Obtain soil state coefficient A so , is the standard value of soil moisture;

[0018] Obtain multiple physiological indicators of crops, including relative water content RWC, normalized vegetation index NDV, stomatal conductance GS, chlorophyll content CHI, through the formula Get the physiological index coefficient A psy, NDV0 is the normalized difference vegetation index comparison value, RWC0 is the relative water content comparison value, GS0 is the stomatal conductance comparison value, CHI0 is the chlorophyll content comparison value, w1, w2, w3 and w4 are their respective weight coefficients.

[0019] Furthermore, the diagnostic analysis layer working method further includes:

[0020] When AR≤AR t When n monitoring and collection time points are set, the growth state value AR collected at each monitoring and collection time point is obtained. i , and formulate the curve function AR(x) of the growth state value changing with time point, according to the weather information at the monitoring and collection time point, search the standard growth state value of the crop in the current growth stage and weather conditions from the historical database Thus, the deviation value Q of each monitoring collection time point is obtained i ,

[0021] By formula The growth difference coefficient is obtained when SF>SF t When the temperature is low, it is judged that the crops need watering;

[0022] Among them, m is the number of monitoring and collection time points whose deviation value is higher than the average deviation value, x1 is the first monitoring and collection time point, and x n The last monitoring collection time point, AR t (x) is the curve function of the preset growth status judgment threshold changing with time points, SF t is the threshold value for determining the growth difference coefficient.

[0023] Furthermore, the working method of the control processing layer is:

[0024] When it is determined that crops need watering, the weather information for the next b hours is obtained from the Meteorological Bureau to obtain the weather influence coefficient μ sr ,

[0025] By the formula QV=γ1*(AR-AR t )*μ sr Or the formula QV=γ2*(SF-SF t )*μ sr Determine the watering volume QV;

[0026] Among them, ET L is the light intensity, ET T is the temperature value, ET W is the wind speed level, ET R is the rainfall, ET S is the humidity value, are light intensity comparison value, temperature comparison value, wind speed comparison value, rainfall comparison value and humidity comparison value respectively; ∈1, ∈2, ∈3, ∈4 and ∈5 are their respective adjustment coefficients; γ1 and γ2 are watering amount conversion coefficients.

[0027] Furthermore, the diagnostic analysis layer is also used to judge the insect pest situation of crops, and the judgment method is:

[0028] Use drones to obtain crop images, use image recognition technology to identify pest-infested areas in the images, and obtain the area ratio S of the pest-infested areas. s and the pest species u and the number NU of each pest in the infested area k , assign a certain weight β to each pest according to its impact on crops k , and thus calculate the degree of pest damage

[0029] When 0<Lv<Lv1, it is judged as micro pest;

[0030] When Lv1≤Lv≤Lv2, it is judged as a mild insect infestation;

[0031] When Lv2<Lv<Lv3, it is judged as moderate pest;

[0032] When Lv≥Lv3, it is judged as a severe insect infestation;

[0033] At the same time, in the case of micro-pests, the pest outbreak situation is judged based on the continuous change rate of the daily pest degree and the daily morphological index coefficient. When the daily pest degree change rate gradually increases as a whole and the corresponding daily morphological index coefficient gradually increases, it means that there is a pest outbreak in the crop, and intervention should be carried out in advance.

[0034] A crop diagnosis and control method based on multi-source data analysis, wherein the control method is controlled and implemented by the crop diagnosis and control model based on multi-source data analysis, and the method comprises:

[0035] Step 1: Obtain data information related to crop growth from multiple data sources, and process and fuse them to obtain standard data information;

[0036] Step 2: Based on the standard data information, a crop growth status judgment model is constructed, and the crop growth status value is calculated according to the model. According to the growth status value, whether the crop is short of water is judged;

[0037] Step 3: Regulate watering in water-scarce crop areas based on weather information;

[0038] Step 4: Based on the crop image information, determine the crop pest situation and pest outbreak situation to intervene in advance.

[0039] Beneficial effects of the present invention:

[0040] The present invention uses multi-source data fusion technology to diagnose water shortages and insect pests in crops. It does not require manual detection, and the information obtained from multiple data terminals is comprehensively analyzed, which can greatly improve the accuracy of judgment. At the same time, when there is no water shortage or the insect pest is mild, the potential water shortage and insect pest outbreak of crops can be judged in advance. In this way, early intervention can be carried out before abnormal conditions occur in crops, which can avoid the impact of judgment lag on crops and ensure crop yield.

[0041] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 It is a structural diagram of the model of the present invention;

[0044] Figure 2 This is a flow chart of the method for crop diagnosis and regulation of the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0046] A crop diagnosis and regulation model based on multi-source data analysis, such as Figure 1 As shown, the diagnostic regulation model includes:

[0047] Multi-source data collection layer, which is used to obtain data information related to crop growth from multiple data sources. The data information includes weather information, soil information, crop physiological information, and crop morphological information;

[0048] Data processing and fusion layer: The data processing and fusion layer processes and fuses the acquired data information to obtain standard data information;

[0049] The diagnostic analysis layer analyzes standard data information to determine whether crop growth is abnormal;

[0050] The control and processing layer analyzes the acquired standard data information and makes corresponding adjustments to the crop areas with abnormal growth.

[0051] Through the above technical solution, the present application first obtains crop-related data information from multiple data sources, including weather information, soil information, crop physiological information, crop image information, crop morphology, etc., and after data processing, obtains standard data information. Data processing includes removing noise and outliers, identifying and eliminating sensor errors, data transmission errors or extreme outliers, and standardizing and normalizing the data, converting data of different dimensions into a unified scale for subsequent analysis. After the processing is completed, based on the obtained standard data information, a comprehensive analysis is performed to determine whether the crop growth is abnormal, including whether the crop is water-deficient, whether there are insect pests, and insect pest outbreaks, etc., and corresponding regulation is carried out based on these conditions to reduce crop losses and ensure crop yields. In this way, no manual detection is required, and the information obtained from multiple data sources is comprehensively analyzed, which can greatly improve the accuracy of judgment. At the same time, when there is no water shortage or the insect pest is mild, the potential water shortage and insect pest outbreak of the crop can be judged in advance. In this way, early intervention can be carried out before the crop abnormality occurs, which can avoid the impact of the lag in judgment on the crop and ensure crop yields.

[0052] The working method of the diagnostic analysis layer is to build a crop growth status judgment model based on standard data information. The model expression is: Among them, A so is the soil state coefficient, A psy is the physiological index coefficient, A fom is the morphological index coefficient;

[0053] By obtaining the current data information of the crop and inputting it into the constructed crop growth status judgment model A, the current growth status value AR of the crop is obtained. When AR>AR t When the crop needs watering, AR t is the judgment threshold, where soil state coefficient A so , physiological index coefficient A psy , morphological index coefficient A fom The acquisition method is:

[0054] The image information of crops is captured by drones, and the leaf curling degree PM, wilting degree PC, and leaf area index PLAI of crops are obtained based on the image information. Obtain the morphological index coefficient A fom , PM0 is the standard value of leaf curling, PC0 is the standard value of wilting, and PLAI0 is the standard value of leaf area index;

[0055] By installing multiple humidity sensors at different depths in the soil, the soil moisture content can be obtained based on the specific gravity at different depths and the humidity sensor values. wr , through the formula Obtain soil state coefficient A so , is the standard value of soil moisture;

[0056] Obtain multiple physiological indicators of crops, including relative water content RWC, normalized vegetation index NDV, stomatal conductance GS, chlorophyll content CHI, through the formula Get the physiological index coefficient A psy , NDV0 is the normalized difference vegetation index comparison value, RWC0 is the relative water content comparison value, GS0 is the stomatal conductance comparison value, CHI0 is the chlorophyll content comparison value, w1, w2, w3 and w4 are their respective weight coefficients.

[0057] The above embodiment provides a specific case for judging the water shortage of crops. When crops are short of water, they are generally characterized by low soil moisture, wilting of crop leaves, reduced stomatal conductance, and a significant decrease in chlorophyll content. Therefore, based on this information, the morphological index coefficient is derived from the morphological condition of the crop, the physiological index coefficient is derived from the physiological condition of the crop, and the soil state coefficient is derived from the moisture content of the soil in which the crop is located. The image information of the crop is captured by a drone, and the leaf curl PM, wilting PC, and leaf area index PLAI of the crop are obtained based on hyperspectral remote sensing technology, and then the formula is used to calculate the leaf curl PM, wilting PC, and leaf area index PLAI of the crop. Obtain the morphological index coefficient A fom , PM0 is the standard value of leaf curling, PC0 is the standard value of wilting, PLAI0 is the standard value of leaf area index. It can be seen that when the leaf curling is greater, the wilting degree is higher or the leaf area index is smaller, the crop water shortage phenomenon is more serious; then by installing multiple humidity sensors at different depths in the soil, according to the specific gravity of different depths combined with the humidity sensor value, the soil moisture content is obtained. wr , through the formula Obtain soil state coefficient A so , is the standard value of soil moisture. It can be seen that the less the soil moisture content, the more serious the corresponding crop water shortage phenomenon. Then obtain multiple physiological indicators of crops, including relative water content RWC, normalized vegetation index NDV, stomatal conductance GS, chlorophyll content CHI, through the formula Get the physiological index coefficient A psy , NDV0 is the normalized vegetation index comparison value, RWC0 is the relative water content comparison value, GS0 is the stomatal conductance comparison value, CHI0 is the chlorophyll content comparison value, w1, w2, w3 and w4 are their respective weight coefficients. When the crop is watered, its water content RWC, normalized vegetation index NDV, stomatal conductance GS, and chlorophyll content CHI will all decrease. Therefore, different proportions of weight coefficients are given according to experience, and then these parameters are combined for comprehensive analysis to obtain the physiological index coefficient A. psy , when A psy The smaller the value, the more serious the corresponding crop water shortage phenomenon. The comparison values and standard values of the above data can be formulated based on historical data and empirical data; finally, based on the soil state coefficient A so , physiological index coefficient A psy , morphological index coefficient A fom A crop growth status judgment model is constructed, and the model expression is: It can be seen from the formula that when the value of the morphological index coefficient is larger, or the soil state coefficient A so And physiological index coefficient A psy The smaller the value, the worse the growth state of the crop. Therefore, the current data information of the crop is obtained and input into the constructed crop growth state judgment model A to obtain the current growth state value AR of the crop. When AR>AR t When, AR t In order to judge the threshold value, it is formulated according to experience, indicating that the crop growth state is poor, and it is judged that the crop needs to be watered. In this way, the morphological index coefficient can be obtained according to the morphological condition of the crop, the physiological index coefficient can be obtained according to the physiological condition of the crop, and the soil state coefficient can be obtained according to the moisture content of the soil where the crop is located. A comprehensive analysis can be conducted based on the morphology, soil condition and physiological condition of the crop, which can more accurately judge the water shortage of the crop and improve the accuracy of the judgment.

[0058] The diagnostic analysis layer working method also includes: when AR≤AR t When n monitoring and collection time points are set, the growth state value AR collected at each monitoring and collection time point is obtained. i , and formulate the curve function AR(x) of the growth state value changing with time point, according to the weather information at the monitoring and collection time point, search the standard growth state value of the crop in the current growth stage and weather conditions from the historical database Thus, the deviation value Q of each monitoring collection time point is obtained i ,

[0059] By formula The growth difference coefficient is obtained when SF>SF t When the temperature is low, it is judged that the crops need watering;

[0060] Among them, m is the number of monitoring and collection time points whose deviation value is higher than the average deviation value, x1 is the first monitoring and collection time point, and x n The last monitoring collection time point, AR t (x) is the curve function of the preset growth status judgment threshold changing with time points, SF t is the growth difference coefficient judgment threshold.

[0061] The above embodiment provides another method for judging the water shortage of crops. First, when AR≤AR t When there is no water shortage, set n monitoring and collection time points, and obtain the growth status value AR collected at each monitoring and collection time point i , and formulate the curve function AR(x) of the growth state value changing with time point, and at the same time, search the standard growth state value of the crop under the current growth stage and weather conditions from the historical database according to the weather information at the monitoring and collection time point Thus, the deviation value Q of each monitoring collection time point is obtained i , The larger the deviation value, the greater the difference between the crop growth situation at the current monitoring and collection time point and the standard situation, and the greater the possibility of true existence. Then, through the formula The growth difference coefficient is obtained, and the formula It indicates the cumulative difference between the change of the acquired growth status value and the growth status judgment threshold. If the value is larger, it means that the change of the acquired growth status value is closer to the growth status judgment threshold, then the possibility of potential water shortage of the crop is greater. Formula It represents the average deviation of the deviation value, and It indicates an out-of-tolerance situation of the deviation value relative to the average deviation. Obviously, the larger the value of the two, the greater the gap between the crop growth and the standard. Therefore, the larger the value, the greater the possibility of potential water shortage of the crop. Therefore, after comprehensive analysis, the growth difference coefficient SF is obtained, and then compared with the growth difference coefficient judgment threshold SF set according to experience. t Compare, when SF>SF tWhen the water level is too low, it indicates that the crop is potentially water-deficient, and watering is necessary. This approach allows for analysis and judgment of potential water shortages before the crop is clearly water-deficient, allowing for early warning intervention. This prevents the impact of delayed judgment on the crop and ensures crop yields.

[0062] The working method of the control processing layer is: when it is determined that the crops need to be watered, the weather information for the next b hours is obtained from the meteorological bureau, and the weather influence coefficient μ is obtained. sr ,

[0063] By the formula QV=γ1*(AR-AR t )*μ sr Or the formula QV=γ2*(SF-SF t )*μ sr Determine the watering volume QV;

[0064] Among them, ET L is the light intensity, ET T is the temperature value, ET W is the wind speed level, ET R is the rainfall, ET S is the humidity value, are light intensity comparison value, temperature comparison value, wind speed comparison value, rainfall comparison value and humidity comparison value respectively; ∈1, ∈2, ∈3, ∈4 and ∈5 are their respective adjustment coefficients; γ1 and γ2 are watering amount conversion coefficients.

[0065] The above embodiment provides a specific method for regulating the amount of watering for crops when it is determined that the crops are short of water. First, based on weather information, weather information for the next b hours is obtained from the meteorological bureau to obtain the weather influence coefficient μ sr , Among them, ET L is the light intensity, ET T is the temperature value, ET W is the wind speed level, ET R is the rainfall, ET S is the humidity value, They are light intensity comparison value, temperature comparison value, wind speed comparison value, rainfall comparison value, and humidity comparison value, and ∈1, ∈2, ∈3, ∈4, and ∈5 can be determined as their respective adjustment coefficients based on historical data and empirical data. They are determined based on actual conditions. For example, when the light intensity is very high, the proportion of ∈1 can be increased. For example, when the rainfall on that day is high, the proportion of ∈4 can be increased. Due to the high light intensity, high temperature, high wind speed level, or low humidity value and low rainfall, the corresponding amount of watering is also more. Therefore, the weather influence coefficient μ is obtained by combining weather information. sr, when it is substituted into the formula QV=γ1*(AR-AR t )*μ sr Or the formula QV=γ2*(SF-SF t )*μ sr Determine the watering amount QV. In the formula, γ1 and γ2 are watering amount conversion coefficients, which can be determined based on historical data combined with comparison data. In this way, the watering amount can be accurately determined based on future weather conditions and the actual water shortage of crops, thus ensuring crop yield while saving resources.

[0066] The diagnostic analysis layer is also used to judge the pest situation of crops. The judgment method is: obtain crop images through drones, obtain the pest area in the image through image recognition technology, and obtain the area ratio S of the pest area. s and the pest species u and the number NU of each pest in the infested area k , assign a certain weight β to each pest according to its impact on crops k , and thus calculate the degree of pest damage

[0067] When 0<Lv<Lv1, it is judged as micro pest;

[0068] When Lv1≤Lv≤Lv2, it is judged as a mild insect infestation;

[0069] When Lv2<Lv<Lv3, it is judged as moderate pest;

[0070] When Lv≥Lv3, it is judged as severe pest infestation; Lv1, Lv2, and Lv3 are the thresholds for judging the degree of pest infestation;

[0071] At the same time, in the case of micro-pests, the pest outbreak situation is judged based on the continuous change rate of the daily pest degree and the daily morphological index coefficient. When the daily pest degree change rate gradually increases as a whole and the corresponding daily morphological index coefficient gradually increases, it means that there is a pest outbreak in the crop, and intervention should be carried out in advance.

[0072] The above embodiment provides a specific method for judging the situation of crop pests. First, the image of the crop is obtained by the drone, and the pest area in the image is obtained by image recognition technology. The area ratio S of the pest area is obtained. s and the pest species u and the number NU of each pest in the infested area k , assign a certain weight β to each pest according to its impact on crops k , and thus calculate the degree of pest damage As can be seen, the greater the pest severity value, the more severe the corresponding pest. Therefore, when 0 < Lv < Lv1, it is judged as a minor pest; when Lv1 ≤ Lv ≤ Lv2, it is judged as a mild pest; when Lv2 < Lv < Lv3, it is judged as a moderate pest; and when Lv ≥ Lv3, it is judged as a severe pest. This method allows real-time monitoring of pest conditions, allowing for timely understanding of crop damage and appropriate treatment to minimize losses. Furthermore, in the case of minor pests, the continuous daily rate of change in pest severity and the daily morphological index coefficient are used to determine the severity of pest outbreaks. Since the morphological index coefficient reflects the growth form of the crop, a larger value indicates poorer growth. If the daily rate of change in pest severity gradually increases, and the corresponding daily morphological index coefficient also gradually increases, then the likelihood of a crop pest outbreak is greater, and early intervention is necessary. In this way, pest outbreaks can be judged in advance when the pest situation is very small, so that early intervention can be carried out before abnormal conditions occur in crops, which can reduce the impact of judgment lags on crops and ensure crop yields.

[0073] In another embodiment, a crop diagnosis and control method based on multi-source data analysis is disclosed. The control method is controlled and implemented by the above-mentioned crop diagnosis and control model based on multi-source data analysis, such as Figure 2 As shown, the control method includes:

[0074] Step 1: Acquire data information related to crop growth from multiple data sources, and process and fuse them to obtain standard data information;

[0075] Step 2: Based on the standard data information, a crop growth status judgment model is constructed, and the crop growth status value is calculated according to the model. According to the growth status value, whether the crop is short of water is judged;

[0076] Step 3: Regulate watering in water-scarce crop areas based on weather information;

[0077] Step 4: Based on the crop image information, determine the crop pest situation and pest outbreak situation to intervene in advance.

[0078] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A crop diagnosis and control model based on multi-source data analysis, characterized in that: The model includes: A multi-source data acquisition layer, wherein the multi-source data acquisition layer is used to obtain data information related to crop growth from multiple data sources; A data processing and fusion layer processes and fuses the acquired data information to obtain standard data information; A diagnostic analysis layer, which performs analysis based on standard data information to determine whether crop growth is abnormal; The regulation and processing layer performs analysis based on the acquired standard data information and performs corresponding regulation on the crop areas with abnormal growth.

2. A crop diagnosis and control model based on multi-source data analysis according to claim 1, characterized in that: The data information includes weather information, soil information, crop physiological information and crop morphological information.

3. The crop diagnosis and control model based on multi-source data analysis according to claim 2, characterized in that: The diagnostic analysis layer works as follows: Based on standard data information, a crop growth status judgment model is constructed, and the model expression is: Among them, A so is the soil state coefficient, A psy is the physiological index coefficient, A fom is the morphological index coefficient; By obtaining the current data information of the crop and inputting it into the constructed crop growth status judgment model A, the current growth status value AR of the crop is obtained. When AR>AR t When the crop needs watering, AR t is the judgment threshold.

4. A crop diagnosis and control model based on multi-source data analysis according to claim 3, characterized in that: The soil state coefficient A so , physiological index coefficient A psy , morphological index coefficient A fom The acquisition method is: The image information of crops is captured by drones, and the leaf curling degree PM, wilting degree PC, and leaf area index PLAI of crops are obtained based on the image information. Obtain the morphological index coefficient A fom , PM0 is the standard value of leaf curling, PC0 is the standard value of wilting, and PLAI0 is the standard value of leaf area index; By installing multiple humidity sensors at different depths in the soil, the soil moisture content can be obtained based on the specific gravity at different depths and the humidity sensor values. wr , through the formula Obtain soil state coefficient A so , so wr0 is the standard value of soil moisture; Obtain multiple physiological indicators of crops, including relative water content RWC, normalized vegetation index NDV, stomatal conductance GS, chlorophyll content CHI, through formula A psy =w1* Get the physiological index coefficient A psy , NDV0 is the normalized difference vegetation index comparison value, RWC0 is the relative water content comparison value, GS0 is the stomatal conductance comparison value, CHI0 is the chlorophyll content comparison value, w1, w2, w3 and w4 are their respective weight coefficients.

5. The crop diagnosis and control model based on multi-source data analysis according to claim 3, characterized in that: The diagnostic analysis layer working method further includes: When AR≤AR t When n monitoring and collection time points are set, the growth state value AR collected at each monitoring and collection time point is obtained. i , and formulate the curve function AR(x) of the growth state value changing with time point, according to the weather information at the monitoring collection time point, search the standard growth state value AR of the crop under the current growth stage and weather conditions from the historical database zi , thus obtaining the deviation value Q of each monitoring collection time point i , By formula The growth difference coefficient is obtained when SF>SF t When the temperature is low, it is judged that the crops need watering; Among them, m is the number of monitoring and collection time points whose deviation value is higher than the average deviation value, x1 is the first monitoring and collection time point, and x n The last monitoring collection time point, AR t (x) is the curve function of the preset growth status judgment threshold changing with time points, SF t is the growth difference coefficient judgment threshold.

6. A crop diagnosis and control model based on multi-source data analysis according to claim 5, characterized in that: The working method of the control processing layer is: When it is determined that crops need watering, the weather information for the next b hours is obtained from the Meteorological Bureau to obtain the weather influence coefficient μ sr , By the formula QV=γ1*(AR-AR t )*μ sr Or the formula QV=γ2*(SF-SF t )*μ sr Determine the watering volume QV; Among them, ET L is the light intensity, ET T is the temperature value, ET W is the wind speed level, ET R is the rainfall, ET S is the humidity value, are light intensity comparison value, temperature comparison value, wind speed comparison value, rainfall comparison value and humidity comparison value respectively; ∈1, ∈2, ∈3, ∈4 and ∈5 are their respective adjustment coefficients; γ1 and γ2 are watering amount conversion coefficients.

7. The crop diagnosis and control model based on multi-source data analysis according to claim 6, characterized in that: The diagnostic analysis layer is also used to judge the insect pest situation of crops. The judgment method is as follows: Use drones to obtain crop images, use image recognition technology to identify pest-infested areas in the images, and obtain the area ratio S of the pest-infested areas. s and the pest species u and the number NU of each pest in the infested area k , assign a certain weight β to each pest according to its impact on crops k , and thus calculate the degree of pest damage When 0<Lv<Lv1, it is judged as micro pest; When Lv1≤Lv≤Lv2, it is judged as a mild insect infestation; When Lv2<Lv<Lv3, it is judged as moderate pest; When Lv≥Lv3, it is judged as a severe insect infestation; At the same time, in the case of micro-pests, the pest outbreak situation is judged based on the continuous change rate of the daily pest degree and the daily morphological index coefficient. When the daily pest degree change rate gradually increases as a whole and the corresponding daily morphological index coefficient gradually increases, it means that there is a pest outbreak in the crop, and intervention should be carried out in advance.

8. A crop diagnosis and control method based on multi-source data analysis, wherein the control method is controlled and implemented by the crop diagnosis and control model based on multi-source data analysis according to any one of claims 1 to 7, characterized in that: The method comprises: Step 1: Obtain data information related to crop growth from multiple data sources, and process and fuse them to obtain standard data information; Step 2: Based on the standard data information, a crop growth status judgment model is constructed, and the crop growth status value is calculated according to the model. According to the growth status value, whether the crop is short of water is judged; Step 3: Regulate watering in water-scarce crop areas based on weather information; Step 4: Based on the crop image information, determine the crop pest situation and pest outbreak situation to intervene in advance.

Images