Big data based smart agriculture monitoring system

Abstract

The application relates to the technical field of agricultural monitoring, and discloses a smart agricultural monitoring system based on big data, which comprises a division unit, an integration unit, an analysis unit, a correlation unit and an evaluation unit. The system takes a single plant as a core, divides grids according to the characteristics of the aboveground and root systems, is different from traditional large-area monitoring, can realize accurate focusing on crops, fuses various data such as images, environments and soils, combines big data and smart agriculture, breaks the limitation of single data, correlates the aboveground and underground conditions through the aboveground physiological advanced value and the root depth activity coefficient, breaks the monitoring lag situation of only monitoring the aboveground part, understands the growth condition of the crops in advance, accurately evaluates the health of the crops, and achieves the effect of accurate detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural monitoring, in particular to a smart agricultural monitoring system based on big data. BACKGROUND

[0002] In the field of smart agriculture, crop monitoring mainly relies on satellite remote sensing images, unmanned aerial vehicle shooting images, and environmental and soil data. Deep learning algorithms are usually used to comprehensively analyze these multi-element data to accurately capture surface conditions such as leaf color changes, so as to timely and comprehensively grasp the growth state of crops.

[0003] However, in the greenhouse planting environment, the data collected at present mostly focuses on the above-ground part of the crops. Although data analysis can help understand the growth trend of crops to some extent, physiological changes of crops often occur before surface morphological changes. For example, after 3-5 days of oxygen deficiency in the root system of crops, the leaf wilting phenomenon will appear, which will lead to poor monitoring effect and obvious lag if only the surface morphology of crops is monitored, and the real growth status of crops cannot be understood in time. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a smart agricultural monitoring system based on big data, which solves the above problems.

[0005] The above technical purpose of the present application is realized by the following technical scheme:

[0006] The smart agricultural monitoring system based on big data comprises:

[0007] A division unit is used to obtain the maximum diameter above ground and the maximum extension radius of the root system of each crop in the greenhouse, determine the dynamic grid size of a single crop, and divide the planting area into a plurality of dynamic grids with the dynamic grid size, each grid covering a single crop and its root system extension range;

[0008] An integration unit is used to collect image data, environmental data and soil data at different depths of a single crop in each grid in real time, and integrate the three preprocessed data to obtain a big data fusion matrix of a single crop;

[0009] An analysis unit is used to analyze the big data fusion matrix to obtain an above-ground physiological leading value and a root depth activity coefficient, respectively;

[0010] An association unit is used to perform correlation analysis on the above-ground physiological leading value and the root depth activity coefficient to obtain a crop health correlation index;

[0011] An evaluation unit is used to determine the health status of crops according to the crop health correlation index.

[0012] Further, the dynamic grid size of the single crop is determined, comprising:

[0013] The maximum diameter above ground and the maximum extension radius of the root system of each crop are calculated to obtain a proportionality coefficient;

[0014] Based on the proportionality coefficient and the maximum diameter above ground of each crop, the root system redundancy coverage radius is determined;

[0015] Taking the plant of the single crop as the center, the planting area is divided into a square grid with a side length of 2 times the root system redundancy coverage radius to generate the dynamic grid size of the single crop.

[0016] Further, the three pre-processed data are integrated to obtain a big data fusion matrix of the single crop, comprising:

[0017] The image data is extracted to obtain an image feature vector;

[0018] The correlation of each data in the environmental data is analyzed, and the correlation is fused to obtain a comprehensive environmental impact index;

[0019] The soil data at different depths is analyzed to obtain a soil feature tensor;

[0020] The image feature vector, the comprehensive environmental impact index and the soil feature tensor are fused to generate the big data fusion matrix of the single crop.

[0021] Further, the big data fusion matrix is analyzed to obtain an above-ground physiological advance value, comprising:

[0022] The big data fusion matrix is image extracted to obtain a time series image change rate;

[0023] According to the time series image change rate and the comprehensive environmental impact index, an environmental physiological response coefficient is determined;

[0024] The soil feature tensor is screened to generate a shallow soil influence weight;

[0025] The historical data of the crop is obtained, and the historical data is analyzed to obtain a historical mean value of the image feature vector at the same growth stage;

[0026] Based on the synchronous comprehensive environmental impact index, the historical mean value is corrected to obtain a dynamic physiological reference vector.

[0027] Further, the big data fusion matrix is analyzed to obtain the above-ground physiological advance value, further comprising:

[0028] The time series image change rate, the environmental physiological response coefficient and the shallow soil influence weight are fused to obtain a current physiological feature prediction value;

[0029] Compare the current physiological characteristic prediction value with the dynamic physiological reference vector to obtain the advance prediction deviation;

[0030] The lead prediction deviation is calculated to generate the above-ground physiological lead value.

[0031] Furthermore, the big data fusion matrix was analyzed to obtain the root depth activity coefficient, including:

[0032] The soil characteristic tensor is parsed to extract the nutrient availability and water conductivity of the soil at each depth, and the depth soil activity weight is obtained;

[0033] Based on the dynamic changes of temperature and light duration in the comprehensive environmental impact index, the depth soil activity weight is corrected to generate a depth correction coefficient;

[0034] Extract the image feature vector to obtain the aboveground growth rate characteristics, and fuse the aboveground growth rate characteristics with the depth correction coefficient to obtain the preliminary root activity coefficient;

[0035] Determine the root activity baseline value for the same period in history based on historical data, and calculate the deviation rate between the preliminary root activity coefficient and the root activity baseline value;

[0036] The preliminary root activity coefficient was calibrated based on the deviation rate to obtain the root depth activity coefficient.

[0037] Furthermore, the root activity baseline values ​​for the same period in history are determined based on historical data, including:

[0038] Get the current date and the crop planting date, and determine the growth stage matching factor based on the two dates;

[0039] Screening growth stage matching factors from historical data All healthy crop samples within 5% are used to construct a temporally and spatially similar sample set;

[0040] Cluster analysis is performed on the spatiotemporally similar sample sets to obtain the basic activity distribution interval;

[0041] Extract environmental data and construct environmental correction tensor;

[0042] Adjust the basic activity distribution interval according to the environmental correction tensor to generate the environmental adaptability benchmark interval;

[0043] The environmental adaptability benchmark interval was disassembled to separate the trend part and the cycle part, and the root activity benchmark value for the same historical period was calculated.

[0044] Further, the aboveground physiological advance value and the root depth activity coefficient are subjected to correlation analysis to obtain a crop health correlation index, including:

[0045] The aboveground physiological advance value and the root depth activity coefficient are decomposed to generate a multi-dimensional growth coordination tensor;

[0046] The matching degree of the aboveground and underground characteristic components in the growth coordination tensor is calculated to obtain a dynamic correlation degree of the growth cycle and the spatial distribution;

[0047] Based on historical data, the growth correlation of the healthy crops in the same period is determined, and a health fluctuation feature library is established;

[0048] The difference value between the current dynamic correlation degree and the health fluctuation feature library is calculated to generate a health feature deviation degree;

[0049] The dynamic correlation degree and the health feature deviation degree are analyzed to obtain the crop health correlation index.

[0050] Further, based on historical data, the growth correlation of the healthy crops in the same period is determined, and a health fluctuation feature library is established, including:

[0051] Based on historical data, the health crop data in different growth stages in the same period is determined;

[0052] The growth stage matching degree and the environmental factor similarity of the health crop data and the current crop are calculated to generate a health sample screening index;

[0053] Based on the health sample screening index, the target historical health crop sample is screened out to obtain a growth correlation feature matrix;

[0054] The growth correlation feature matrix is analyzed and a health fluctuation feature library is established.

[0055] Further, according to the crop health correlation index, the crop health status is determined, including:

[0056] Based on the aboveground physiological advance value, the root depth activity coefficient and the historical data, a health threshold interval is determined;

[0057] The health threshold interval and the crop health correlation index are compared to obtain the crop health status.

[0058] In summary, the present application has the following advantages:

[0059] The precision of single plant crop monitoring is realized by dynamic grid division, the traditional greenhouse monitoring usually adopts fixed area sampling, it is difficult to take into account the growth difference of different crops, and the grid size is dynamically calculated based on the maximum diameter on the ground and the root system expansion radius, the root system redundant coverage radius is determined by combining the proportion coefficient, so that each grid accurately covers single plant and the influence range of root system, this dynamic adjustment mechanism not only avoids the interference of adjacent crops, but also ensures the integrity of the data of the environment around the root system.

[0060] The fusion matrix is generated by integrating image data, environmental data and soil data, and then the big data is combined with intelligent agriculture to improve the monitoring effect, the time sequence image change rate, environmental physiological response coefficient and other characteristics are extracted, the dynamic physiological benchmark is obtained by combining the historical data correction, the above-ground physiological advanced value is calculated, and the soil feature tensor is analyzed, the root depth activity coefficient is generated by combining the environmental dynamic correction, the correlation analysis of above-ground morphological changes and root activity is carried out, the physiological abnormalities can be captured before the symptoms of crops appear, for example, the physiological fluctuation caused by root hypoxia can be identified 3-5 days in advance, the time limitation of traditional surface morphology monitoring is broken, and a key window period is provided for timely intervention, so that the lag of monitoring is avoided.

[0061] The matching degree of above-ground and underground features is calculated by the growth coordination tensor, and the health correlation index is generated by combining the health fluctuation feature library, the physiological linkage of the above-ground part and the root system of crops is considered, and the influence of environmental factors and historical data is also taken into account, so that the health evaluation is upgraded from single index judgment to multi-dimensional dynamic correlation analysis, compared with the traditional evaluation method which only depends on the surface characteristics such as leaf color, the health condition determined by the health correlation index can more comprehensively reflect the real physiological state of crops, and the monitoring effect is improved. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 It is a schematic diagram of the intelligent agricultural monitoring system based on big data of the application. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the application.

[0064] Reference Figure 1 The intelligent agricultural monitoring system based on big data comprises:

[0065] The division unit is configured to obtain the maximum above-ground diameter and the maximum root system expansion radius of each crop in the greenhouse, determine a dynamic grid size of the single crop, and divide the planting area into a plurality of dynamic grids according to the dynamic grid size, each of which covers the single crop and the root system expansion range thereof;

[0066] The integration unit is configured to collect image data, environmental data and soil data at different depths of the single crop in each grid in real time, integrate the three data after preprocessing, and obtain a big data fusion matrix of the single crop;

[0067] The analysis unit is configured to analyze the big data fusion matrix to obtain an above-ground physiological leading value and a root system depth activity coefficient, respectively.

[0068] The correlation unit is configured to perform correlation analysis on the above-ground physiological leading value and the root system depth activity coefficient to obtain a crop health correlation index.

[0069] The evaluation unit is configured to determine the health status of the crop according to the crop health correlation index.

[0070] Through dynamic grid division, the single crop and the root system range are accurately covered, the above-ground image, environmental data and soil data at different depths are integrated, the above-ground and underground data fusion is realized, the health index is obtained by analyzing the above-ground physiological leading value and the root system depth activity coefficient, the physiological change can be captured in advance, the lag of only monitoring the surface morphology is overcome, the real growth state of the crop is more timely and accurately reflected, and the method is suitable for monitoring and planting of crops in the greenhouse.

[0071] In one case of the embodiment, the dynamic grid size of the single crop is determined, including:

[0072] The maximum above-ground diameter and the maximum root system expansion radius of each crop are calculated to obtain a proportion coefficient, specifically including: the maximum above-ground diameter of the single crop is divided by the maximum root system expansion radius to obtain the proportion coefficient.

[0073] Based on the proportion coefficient and the maximum above-ground diameter of each crop, a root system redundant coverage radius is determined, specifically including: the proportion coefficient is multiplied by 0.3, the maximum above-ground diameter is multiplied by 0.7, and the two products are added, and the obtained result is the root system redundant coverage radius.

[0074] With the single crop plant as the center, the planting area is divided into a square grid with a side length of 2 times the root system redundancy coverage radius, to generate the dynamic grid size of the single crop plant, specifically including: establishing a plane rectangular coordinate system with the single crop plant as the origin, obtaining the plant center point coordinates, multiplying the root system redundancy coverage radius by 2 to obtain the square side length, taking the plant of the single crop as the center point as the reference to extend the length of half of the side length in the east, west, south and north directions respectively, determining the coordinate values of the four vertices of the square, forming a closed square area through the coordinate point connecting line, which is the dynamic grid size of the single crop plant, ensuring that the distance from the grid boundary to the plant center is equal everywhere and covers the root system expansion range.

[0075] Through dynamic grid division, accurate coverage of the crop root system range is achieved. In greenhouse planting, the limitations of traditional monitoring of only the aboveground part are solved. With the calculation of the root system redundancy coverage radius, the grid can meet the root system expansion demand, which makes the monitoring extend to the underground root system area. Combined with soil data, etc., physiological changes such as root system hypoxia can be captured in advance to avoid lag and help growers to grasp the real growth status of crops in a timely manner.

[0076] By taking the plant as the center of the dynamic grid and ensuring that the distance from the boundary to the center is equal, the balance and integrity of the monitoring range are guaranteed. The clear coordinate definition of the square grid facilitates the integration of various big data related data. In intelligent agricultural monitoring, this structured grid can more efficiently analyze data and improve the accuracy of crop growth state judgment.

[0077] In one case of the embodiment, the three preprocessed data are integrated to obtain a big data fusion matrix of the single crop plant, including:

[0078] The image data is extracted to obtain an image feature vector, specifically including: the preprocessed image data is subjected to grayscale processing to convert the color image into a grayscale image, the Sobel operator is used to calculate the gradient values in the horizontal and vertical directions, the leaf edge contour is identified by the gradient value size, the Harris corner point detection algorithm is used to calculate the gray value change of the pixel points in the image, the leaf vertex position is determined and the coordinates are recorded, and the number of leaves is counted, the pixel point counting method is used to calculate the leaf area (100 pixel points correspond to 1 square centimeter in reality), and then the plant height can be converted by the difference value of the upper and lower edge pixel points (200 pixel points correspond to 1 meter in reality), the number of leaves, area, plant height and edge contour features are arranged in a fixed order to form an image feature vector;

[0079] The correlation degrees of various data in the environmental data are analyzed, and the correlation degrees are fused to obtain a comprehensive environmental influence index, specifically including: standardizing parameters such as temperature, humidity, light intensity, and carbon dioxide concentration of the environmental data, calculating linear correlation coefficients between two parameters by using a Pearson correlation coefficient algorithm to form a correlation matrix, comparing the importance of each parameter by using an analytic hierarchy process to construct a judgment matrix and calculate a weight vector by using a square root method, and performing weighted summation on the correlation matrix and the weight vector to obtain the comprehensive environmental influence index, wherein the elements of the correlation matrix range from -1 to 1, and the sum of the elements of the weight vector is 1;

[0080] The soil data at different depths are analyzed to obtain a soil feature tensor, specifically including: dividing the soil data at different depths into three layers according to 0 20 centimeters, 20 40 centimeters, and 40 60 centimeters, collecting indicators such as pH value, nitrogen content, phosphorus content, potassium content, and organic matter content in each layer, calculating variance contribution rates of the indicators by using a principal component analysis method, selecting the first three principal components with a cumulative contribution rate exceeding 85%, arranging the three principal components of each layer in order of depth from shallow to deep, with each principal component corresponding to a specific numerical value, and forming a three-dimensional array containing depth layering, principal component sequence number, and feature value, which is the soil feature tensor.

[0081] The image feature vector, the comprehensive environmental influence index, and the soil feature tensor are fused to generate a big data fusion matrix of a single crop, specifically including: converting the image feature vector into a row vector, taking the comprehensive environmental influence index as a single column vector, and expanding the soil feature tensor into a two-dimensional matrix, sequentially concatenating the image feature vector, the comprehensive environmental influence index, and the expanded soil feature tensor matrix in this order by using a serial fusion algorithm, determining the total number of rows of the matrix as the dimension value of the image feature vector plus 1 plus the number of rows of the expanded soil feature tensor, and determining the total number of columns of the matrix as the maximum value among the number of columns of the image feature vector, the number of columns of the comprehensive environmental influence index, and the number of columns of the expanded soil feature tensor, so as to determine the number of rows and the number of columns, and form the matrix, which is the big data fusion matrix.

[0082] The big data fusion matrix is generated by integrating image, environmental, and soil data, breaking through the limitation of focusing only on the aboveground part in greenhouse planting, the image feature vector captures aboveground information such as leaf number and plant height, the soil feature tensor covers key indicators at different depths, and the comprehensive environmental influence index fuses multiple environmental parameters, so that the linkage analysis of crops above and below ground and biological and environmental data is realized, comprehensive data support is provided for accurately judging the growth state, the one-sidedness of single aboveground data is avoided, and the purpose of monitoring crops based on big data is achieved.

[0083] Through the combination of soil data and image data, the root system problem can be warned in advance through soil index abnormalities before the wilting of the leaves, effectively solving the lag of conventional monitoring. The standardization of the matrix can enable efficient correlation of multi-source data, making it easier to capture the correlation rules of physiological and morphological changes and timely grasp the real growth status of crops.

[0084] In one case of the embodiment, the big data fusion matrix is analyzed to obtain an above-ground physiological advanced value, including:

[0085] The big data fusion matrix is subjected to image extraction to obtain a time series image change rate, specifically including: extracting column data corresponding to the image feature vector from the big data fusion matrix, which contains leaf number, area, plant height and other characteristics, sorting them according to the collection time to form a continuous time series, using a sliding window method, setting the window length to 24 hours and the window step to 6 hours, for each window, calculating the average value of leaf area and the average value of plant height in the window, and then calculating the mean difference of the adjacent two windows, specifically: the average value of leaf area of the latter window minus the average value of leaf area of the former window to obtain the mean difference of leaf area, and through the same calculation method, the mean difference of plant height is obtained. The mean difference of leaf area is divided by 24 hours to obtain the leaf area change per unit time, and the mean difference of plant height is divided by 24 hours to obtain the plant height change per unit time. Linear regression algorithm is used to perform trend fitting on the leaf area change per unit time and the plant height change per unit time, and the slope of the straight line obtained by fitting is the corresponding time series image change rate;

[0086] According to the time series image change rate and the comprehensive environmental influence index, the environmental physiological response coefficient is determined, specifically including: setting the leaf area change rate weight to 0.6, the plant height change rate weight to 0.2, and the comprehensive environmental influence index weight to 0.2, multiplying the leaf area change rate and the plant height change rate in the time series image change rate and the comprehensive environmental influence index by the corresponding weight, and then summing them up, the sum is the environmental physiological response coefficient;

[0087] The soil feature tensor is screened to generate a shallow soil influence weight, specifically including: from the soil feature tensor, 3 principal component data corresponding to the 0 20 cm shallow layer are screened out, the eigenvalues of the three principal components are normalized, the numerical values are converted to the 0-1 interval through linear mapping, the information entropy of each principal component is calculated, the information entropy of each principal component is subtracted by 1 to obtain the preliminary entropy weight, the three preliminary entropy weights are normalized so that their sum is 1, and the three normalized preliminary entropy weights are multiplied by the preset adjustment coefficient 1.2 to obtain the shallow soil influence weight. The number of shallow soil influence weights is three, and the sum of the three shallow soil influence weights is 1;

[0088] The historical data of the crops is obtained, and the historical data is analyzed to obtain historical mean values of image feature vectors in the same growth stage. Specifically, historical data of the crops in the last three complete growth cycles is obtained from the historical data, and is divided into growth stages such as the seedling stage (1-30 days after sowing) and the jointing stage (31-60 days after sowing) according to the number of days after sowing. For the current stage of the crops, all image feature vectors in the same stage in the historical data are extracted, and the features include the number of leaves, the area, the plant height, and the edge contour. The values of each feature in the same stage are summed up respectively by using an arithmetic mean algorithm, and then divided by the number of samples (the number of samples in each stage is not less than 50 groups). The result is the historical mean value of each feature in the image feature vector in the same growth stage.

[0089] The historical mean values are corrected based on a comprehensive environmental influence index in the same period to obtain a dynamic physiological benchmark vector. Specifically, when the comprehensive environmental influence index is 0, the corresponding correction coefficient is 0.8, and when the index is 2, the corresponding correction coefficient is 1.2. When the index is between 0 and 2, the corresponding coefficient is obtained by proportional linear conversion. The historical mean value of each feature in the historical mean vector is multiplied by the converted correction coefficient, and the four corrected values are arranged in the original order to form a vector, which is the dynamic physiological benchmark vector. The proportional linear conversion includes: dividing the comprehensive environmental influence index into 20 interval segments between 0 and 2, each interval segment corresponds to an index change of 0.1. For each increase of 0.1 in the index value, the correction coefficient increases by 0.02. For example, when the comprehensive environmental influence index is 0.1, the corresponding correction coefficient is 0.82, and when the index is 0.2, the corresponding correction coefficient is 0.84. This process continues until the index reaches 2, and the correction coefficient reaches 1.2.

[0090] Through the combination of the dynamic physiological benchmark vector and the time series image change rate, the physiological state of the crops is perceived in advance. The dynamic physiological benchmark vector is corrected based on historical data and real-time environment, breaking the limitation of fixed threshold, and more closely following the real growth law of crops. The time series image change rate captures the subtle growth trend through a sliding window, and after comparison with the benchmark vector, it can identify physiological deviation in advance when the leaf morphology is not abnormal. For example, in the early stage of root hypoxia, a small deviation in the plant height change rate from the benchmark can be used as an early warning, greatly shortening the response time difference between physiological changes and apparent manifestations.

[0091] The accuracy of physiological state evaluation is improved through the quantitative fusion of environmental and soil factors. The environmental physiological response coefficient integrates the image change and the environmental influence by weighting, and the shallow soil influence weight focuses on the key indicators of the root active zone. The two are analyzed in conjunction with the aboveground features, reflecting the correlation between soil, environment, and crops. This makes the physiological advance value not only reflect the surface growth trend, but also be related to the soil environment state of the root, avoiding misjudgment by single aboveground data.

[0092] In one case of the embodiment, the big data fusion matrix is analyzed to obtain the aboveground physiological leading value, and the method further comprises:

[0093] The time series image change rate, the environmental physiological response coefficient, and the shallow soil influence weight are fused to obtain a current physiological characteristic prediction value, and the method specifically comprises: setting the time series image change rate weight as 0.5, the environmental physiological response coefficient weight as 0.25, and the shallow soil influence weight as 0.25, multiplying the time series image change rate by 0.5, the environmental physiological response coefficient by 0.25, and the shallow soil influence weight by 0.25, adding the products of the three, and obtaining the result as the current physiological characteristic prediction value;

[0094] The time series image change rate directly reflects the real-time growth dynamics of the leaf area, plant height and other aboveground parts, is the most intuitive quantitative index of the physiological state of the crop, plays a leading role in the current characteristic prediction, and is therefore given the highest weight of 0.5. The environmental physiological response coefficient is the coupling result of the external environment and the growth dynamics, belongs to an indirect influencing factor, and is therefore given a weight of 0.25. The shallow soil influence weight corresponds to the 0-20 cm soil layer in which the root activity is most active, affects the nutrient absorption, but the effect is indirectly reflected through the physiological response, and is therefore given the same weight of 0.25 as the environmental coefficient, so as to form a weight distribution in which the self growth is dominant and the internal and external influences are cooperative;

[0095] The current physiological characteristic prediction value is compared with the dynamic physiological benchmark vector to obtain a leading prediction deviation, and the method specifically comprises: using a vector difference algorithm to calculate the difference between the current physiological characteristic prediction value and the dynamic physiological benchmark vector according to the corresponding characteristic dimensions, specifically: subtracting the leaf number in the prediction value from the leaf number in the benchmark vector to obtain the difference value, and according to the same calculation method, the area, plant height and contour characteristic difference values can be obtained. The four characteristic difference values are arranged in the original sequence to form a vector, which is the leading prediction deviation;

[0096] The leading prediction deviation is calculated to generate an aboveground physiological leading value, and the method specifically comprises: for each sampling point , the time series image change rate is divided by the historical same-period image characteristic vector mean to obtain a relative change rate term , the time derivative of the relative change rate term is calculated to obtain a time change rate term , the time benchmark quantity is divided by the change rate benchmark quantity to obtain a product term , the product term is multiplied by the time change rate term to obtain a normalized benchmark quantity term ; the comprehensive environmental influence index is divided by the environmental response sensitivity adjustment coefficient to obtain an environmental index ratio term , the environmental index ratio term is applied to the first type of modified Bessel function to obtain an environmental impact correction term , the normalized reference quantity term is multiplied by the environmental impact correction term to obtain a multiplication result , the multiplication results of the sampling points are summed and averaged to obtain a time series feature term ; along the soil depth variable , from 0 to the shallow soil depth threshold , the integral is obtained to obtain the soil feature change rate term , the soil feature reference quantity is multiplied by the shallow soil depth threshold , to obtain a product result, the time reference quantity is divided by the product result to obtain , the is multiplied by the soil feature change rate term to obtain the normalized reference quantity , the normalized reference quantity is multiplied by the soil above-ground correlation gain coefficient to obtain a gain coefficient term ; the time series feature term and the gain coefficient term are added to obtain the above-ground physiological lead value , in specific applications, the following calculation formula can be used, for example:

[0097] ;

[0098] In the formula, , the above-ground physiological lead value is represented, , the total number of time series data sampling points is represented, , the index of is represented, , the time reference quantity is represented, , the change rate reference quantity is represented, , the time partial derivative operator is represented, , the time series image change rate is represented, , the historical contemporaneous image feature vector mean is represented, , the first type of modified Bessel function is represented, , the comprehensive environmental impact index is represented, , the adjustment coefficient of environmental response sensitivity is represented, 1 5, , the soil above-ground correlation gain coefficient is represented, 2 7, , the soil feature reference quantity is represented, , the shallow soil depth threshold is represented, , the soil depth is represented, , the time change rate of the soil feature tensor is represented, represents the depth differential;

[0099] The following are Explanation of the value range:

[0100] When the root redundancy radius When the diameter is 1.5 times that of the aboveground root, it indicates that the potential of the root system to absorb water and fertilizer is not fully utilized, and the driving force of the root activity signal on aboveground physiology is weakened. The value is low, The value is 1 3;

[0101] When the root redundancy radius When the diameter of the root system is 1.5 times that of the crown, it indicates that the root system expansion range is smaller than the crown width (underground space is limited), the competition for water and nutrients between roots is intensified, and the risk of root hypoxia increases. The value is too high. The value is 4 5;

[0102] The following are Explanation of the value range:

[0103] When crops are planted densely, that is, the side length of the dynamic grid of a single crop At 50cm, because the number of plants per unit area exceeds the reasonable range, the competition for light energy, water and nutrients is fierce, and the root system is intertwined, resulting in a decrease in oxygen content in the root zone. The value is low, The value is 2 4.7;

[0104] When crops are planted sparsely, that is, the side length of the dynamic grid of a single crop When the height is 50cm, the number of plants per unit area is lower than the appropriate range, the individual growth space of crops is excessive, the water absorption resistance of the root system is reduced, but the light energy utilization rate of the group is reduced. The value is too high. The value is 4.8 7.

[0105] By integrating the rate of change of time-series images, environmental impacts and shallow soil weights, with self-growth as the main factor and internal and external factors coordinated in weight distribution, the current physiological characteristic prediction value is accurately calculated. After comparing with the dynamic physiological reference vector, it can capture subtle deviations in characteristics such as leaf number and plant height through advanced prediction deviation. Combined with the parameter adjustment of root redundant radius and planting density, it can reflect the root activity and competition status in a targeted manner, and warn of root hypoxia and other problems 3-5 days before leaf wilt, greatly shortening the time difference between physiological changes and manifestations, and solving the monitoring lag.

[0106] Through the integration of the time sequence feature term and the soil gain coefficient term, multi-dimensional data such as time change rate, environmental correction and soil integration are fused to form a quantifiable index that can be directly applied. The parameter values under different root system redundancy radius and planting density allow the index to adapt to different growth scenarios, avoid misjudgment of a single standard, and make quantitative analysis from the physiological mechanism. The monitoring is deepened from surface morphology observation to physiological state prediction, and the growth status of crops in smart agriculture can be understood in a timely manner based on big data analysis.

[0107] In one case of the embodiment, the root depth activity coefficient is obtained by analyzing the big data fusion matrix, including:

[0108] The soil feature tensor is analyzed to extract the nutrient availability and water conductivity of each depth soil to obtain the depth soil activity weight, specifically including: extracting three principal component eigenvalues of each depth layer from the soil feature tensor, using a multivariate linear regression algorithm to map these eigenvalues to the nutrient availability and water conductivity of the corresponding depth, setting the weight of the nutrient availability as 0.6 and the weight of the water conductivity as 0.4, multiplying the nutrient availability of each depth layer by 0.6 and the water conductivity by 0.4, and adding the two to obtain the activity score of the depth layer. The activity scores of each depth layer are normalized to make the results between 0 and 1. The obtained value is the depth soil activity weight. The depth soil activity weight is three, corresponding to the depth soil of three layers respectively.

[0109] Based on the dynamic change values of temperature and light duration in the comprehensive environmental influence index, the depth soil activity weight is corrected to generate a depth correction coefficient, specifically including: using a sliding window method (window 24 hours, step 6 hours) to process the temperature and light duration data in the comprehensive environmental influence index, specifically: calculating the average value of temperature and the average value of light duration in each window, and then calculating the mean difference of adjacent windows. Divide the mean difference by 24 hours to obtain the temperature dynamic change rate and the light duration dynamic change rate. Set the temperature change rate weight as 0.6 and the light change rate weight as 0.4. Add the temperature dynamic change rate and the light duration dynamic change rate multiplied by the corresponding weight to obtain the environmental correction factor. Multiply the environmental correction factor by the depth proportion (1 / 3) to obtain the basic correction value of the corresponding depth. Multiply the basic correction value of each depth with the depth soil activity weight of the layer respectively to obtain three depth correction coefficients, corresponding to the depth soil of three layers respectively.

[0110] The image feature vector is extracted to obtain the aboveground growth rate feature, and the aboveground growth rate feature is fused with the depth correction coefficient to obtain a preliminary root activity coefficient, specifically including: extracting the time sequence data corresponding to the leaf area and plant height from the image feature vector, sorting according to the collection time, calculating the average value of the leaf area and the average value of the plant height in each window by using the sliding window method (window 24 hours, step 6 hours), and then calculating the mean difference of the adjacent two windows, that is, the mean value of the latter window minus the mean value of the former window, and then dividing the mean difference by 24 hours to obtain the unit time change of the leaf area and the plant height, and then the two change amounts are respectively fitted by using the linear regression algorithm, and the slope of the fitted straight line is the aboveground growth rate feature, the growth rate feature is set to be 0.3, and the three depth correction coefficients are each one third of 0.7, the growth rate feature is multiplied by 0.3, the three depth correction coefficients are respectively multiplied by 0.7 and divided by 3, and then all the products are added, and the addition result is the preliminary root activity coefficient;

[0111] The historical data is used to determine the historical root activity reference value of the same period;

[0112] The deviation rate of the preliminary root activity coefficient and the root activity reference value is calculated, specifically including: subtracting the root activity reference value from the preliminary root activity coefficient to obtain a difference value, and then dividing the difference value by the root activity reference value, and the obtained result is the deviation rate;

[0113] The preliminary root activity coefficient is calibrated based on the deviation rate to obtain a root depth activity coefficient, specifically including: calculating the exponential decay weight of each soil layer , multiplying the exponential decay weight of each soil layer , the soil activity weight and the environmental correction coefficient to obtain a product term , summing the product terms of all soil layers, and then dividing by the sum of the exponential decay weights to obtain a normalized depth weighted average term ; multiplying the three-day growth acceleration rate by the growth sensitivity and taking the negative exponent to obtain an exponential response term ; dividing 2 by to obtain a response ratio term ; multiplying the depth weight of each layer by the environmental correction coefficient to obtain a product result , summing the product results of all soil layers to obtain a sum result , subtracting the historical reference value from the sum result and dividing by the reference standard deviation to obtain a deviation calibration term​ , apply the S-type function to the bias calibration term and map it to The normalized deviation calibration term is obtained ; The aboveground growth rate , the normalized depth weighted average term, the response ratio term, and the normalized deviation calibration term are multiplied to obtain the root depth activity coefficient , when applied specifically, it can be achieved through the following calculation formula, for example:

[0114] ;

[0115] Where, represents the root depth activity coefficient, represents the aboveground growth rate, represents the number of soil layers, express The index of represents the base of natural logarithms, represents the depth attenuation factor (0.5), Indicates the Layer soil activity weight, , Indicates the Layer environment correction factor, , represents growth sensitivity (0.8), Indicates the three-day growth acceleration rate, represents the sigmoid function, represents the depth weight, , Indicates the historical benchmark value, represents the benchmark standard deviation;

[0116] The following are Explanation of the value range:

[0117] When dense planting occurs, competition for resources such as light, water and nutrients becomes intense. The soil structure deteriorates, resulting in the inhibition of root activity and poor soil activity. The value is low, The value is ;

[0118] When sparsely planted, resources such as light, water and nutrients can be fully utilized by crops, and the soil activity is better, so The value is too high. The value is ;

[0119] The following are Explanation of the value range:

[0120] When the soil water content 40% of the water holding capacity of the entire planting area, the root system of the crop has difficulty in absorbing water, so the value is on the low side, the value is ;

[0121] When the soil water content 40% of the water holding capacity of the entire planting area, the root system of the crop has difficulty in absorbing water, so the value is on the high side, the value is ;

[0122] The following is an explanation of the value range of When the root depth of the crop is mainly concentrated in the shallow layer of the soil, that is, 0-20 cm, at this time the soil depth is mainly reflected in the surface layer, so

[0123] the value is on the low side, the value is ;

[0124] When the root depth of the crop is mainly concentrated in the deep layer of the soil, that is, 20 40 cm, 40 60 cm, at this time the soil depth is mainly reflected in the deep layer, so the value is on the high side, the value is .

[0125] By integrating multi-dimensional data such as soil activity weight, environmental correction coefficient and above-ground growth rate, the combination of big data and smart agriculture is realized, and through exponential decay weight and bias calibration, the root activity of different soil depths is analyzed in layers. Through parameter adaptation of planting density, soil water content and root distribution depth, the influence of resource competition, water absorption efficiency and other factors on the root system can be reflected, which breaks through the limitation of only monitoring the above-ground part, and enables the grower to intuitively understand the potential problems such as root anoxia.

[0126] The dynamic calculation of the root depth activity coefficient effectively solves the problem of monitoring lag, wherein the dynamic changes of the environment and growth are captured through a sliding window, and the bias is calibrated using historical benchmark values, so that the coefficient can respond to physiological changes in the root system in real time. Compared with the lagging symptoms such as leaf wilting, this coefficient can detect abnormalities in soil activity and growth rate to provide early warning of crop root problems, which facilitates timely intervention in greenhouse planting.

[0127] In one case of the embodiment, the historical root activity benchmark value of the same period is determined based on historical data, including:

[0128] ​Obtaining the current date and the planting date of the crop, and determining a growth stage matching factor based on the two dates, specifically including: calculating the difference between the current date and the planting date of the crop, which is the number of days that have grown, based on the number of days in each growth stage in the historical data, such as the seedling stage (1-30 days), the jointing stage (31-60 days), etc., to determine the current stage, and subtract the number of days that have grown in this stage from the starting day of the current stage, and then divide by the total number of days in the stage (such as 30 days in the seedling stage), and the result obtained is the growth stage matching factor;

[0129] Screening the growth stage matching factor from the historical data 5% of all healthy crop samples, constructing a spatiotemporal similar sample set, specifically including: extracting the growth stage matching factor of all healthy crop samples from the historical data, calculating the difference between each sample factor and the current growth stage matching factor, and screening out samples with an absolute difference value not exceeding 5% of the current factor, grouping these samples by planting year and growth stage to form a spatiotemporal similar sample set;

[0130] Performing clustering analysis on the spatiotemporal similar sample set to obtain a basic activity distribution interval, specifically including: using the K-means clustering algorithm to analyze the spatiotemporal similar sample set, specifically: taking the root activity value of each healthy crop in the spatiotemporal similar sample set as the clustering feature, setting the number of clusters to 3, calculating the Euclidean distance between the root activity value of each sample and the 3 initial cluster centers, dividing the sample into the nearest cluster group, iteratively updating the cluster center until the center no longer changes, at this time, the root activity value of each class of samples is counted, and the minimum and maximum values in each class are taken to form the distribution interval of the class, and the three intervals are combined to obtain the basic activity distribution interval;

[0131] Extracting environmental data to construct an environmental correction tensor, specifically including: extracting temperature, humidity, light intensity and carbon dioxide concentration from the environmental data, taking the data of the last 7 days and calculating the daily average, using the analytic hierarchy process to determine the weight of each parameter, multiplying the daily average by the corresponding weight to obtain the daily correction value, arranging these correction values by day as rows and parameter as columns to form a 4-column 7-row matrix, which is the environmental correction tensor;

[0132] Adjusting the basic activity distribution interval according to the environmental correction tensor to generate an environmental adaptability benchmark interval, specifically including: calculating the arithmetic mean of the daily correction values in the environmental correction tensor to obtain a correction coefficient, multiplying the minimum and maximum values of the basic activity distribution interval by the correction coefficient to obtain the adjusted minimum and maximum values, and the interval formed by the two is the environmental adaptability benchmark interval;

[0133] The environmental adaptability benchmark interval was disassembled to separate the trend part and the cycle part, and the root activity benchmark value for the same period in history was calculated, specifically including: using the 3-day moving average method to process the values ​​of the environmental adaptability benchmark interval, specifically: taking the arithmetic average of the values ​​of 3 consecutive days in the environmental adaptability benchmark interval to obtain the trend part, subtracting the corresponding trend part value from the original interval value, and the remaining part is the cycle part. The arithmetic average of the trend part and the cycle part were calculated respectively, and the result of adding the two is the root activity benchmark value for the same period in history.

[0134] By accurately determining the baseline value of root activity during the same period in history, the reliability of root monitoring is improved. By combining growth stage matching with K-means clustering, the basic activity interval is extracted from healthy samples, and then dynamically adjusted through the environmental correction tensor to make the baseline value adapt to the current environment. This not only reflects the growth patterns of crops, but also incorporates environmental influences, avoiding judgment bias caused by environmental differences. It can more accurately identify root abnormalities, make up for the shortcomings of monitoring based solely on ground data, and enhance the ability of smart agriculture to perceive the true growth status of crops.

[0135] In one case of this embodiment, correlation analysis is performed on the aboveground physiological advance value and the root depth activity coefficient to obtain a crop health correlation index, including:

[0136] The aboveground physiological advance value and the root depth activity coefficient are decomposed to generate a multi-dimensional growth coordination tensor, which includes: extracting the four basic sub-items of leaf number, area, plant height, and contour characteristics from the aboveground physiological advance value; for the root depth activity coefficient, 20cm, 20 40cm, 40 Three 60cm soil layers were used to extract the active sub-items corresponding to each layer. These were then classified according to three dimensions: time (daily), growth indicator type (aboveground / belowground), and soil depth. A normalization algorithm was used to convert all sub-item values ​​to the 0-1 interval to obtain a baseline value. Using the analytic hierarchy process, the importance of each sub-item in the corresponding dimension was compared pairwise to determine its respective weight. Under the cross-classification of the three dimensions, the baseline value of each sub-item was multiplied by its weight. The results for the same dimensional combinations were then summed to form a three-dimensional array, namely, a multi-dimensional growth coordination tensor.

[0137] The matching degree of the aboveground and underground characteristic components in the growth coordination tensor is calculated to obtain the dynamic correlation degree of the growth period and the spatial distribution, specifically including: from the growth coordination tensor, aligning the aboveground characteristic components (the number of leaves, area, plant height, and contour characteristics) with the underground characteristic components (three layers of soil activity) according to the daily time dimension, the cosine similarity algorithm is used to calculate the similarity degree of the aboveground and underground components at the same time point (by comparing the coincidence degree of the numerical distribution of the two), to obtain the daily matching degree, the average value of the daily matching degree in each period (such as every 7 days as a period) is calculated as the period matching degree, and then the period matching degree is combined with the soil depth weight of the underground component (the weight of the shallow layer 0-20 cm is 0.5, the weight of the middle layer 20 40 cm is 0.3, and the weight of the deep layer 40 60 cm is 0.2), the period matching degree is multiplied by the corresponding depth weight and then summed, and the sum is the dynamic correlation degree of the growth period and the spatial distribution.

[0138] Based on the historical data, the growth correlation of the healthy crops in the same period is determined, and a healthy fluctuation feature library is established.

[0139] The difference value between the current dynamic correlation degree and the healthy fluctuation feature library is calculated to generate a health feature deviation degree, specifically including: extracting the historical dynamic correlation degree data matched with the current growth stage from the healthy fluctuation feature library, and calculating the arithmetic mean value thereof, then subtracting the arithmetic mean value from the current dynamic correlation degree to obtain the difference value, and then dividing the difference value by the standard deviation of the historical dynamic correlation degree data, and the result obtained is the health feature deviation degree.

[0140] The dynamic correlation degree and the health feature deviation degree are analyzed to obtain a crop health correlation index, specifically including: setting the dynamic correlation degree weight as 0.6 and the health feature deviation degree weight as 0.4, taking the absolute value of the health feature deviation degree, then subtracting the absolute value from 1 to obtain a health feature adaptation degree, multiplying the dynamic correlation degree by 0.6, multiplying the health feature adaptation degree by 0.4, and then adding the two product results to obtain the crop health correlation index.

[0141] By generating the crop health correlation index, the linkage analysis of the aboveground and underground growth states is realized, which decomposes the aboveground physiological lead value and the root depth activity coefficient into multiple dimensions, constructs a growth coordination tensor, calculates the dynamic correlation degree by combining the cosine similarity, compares it with the health fluctuation feature library to generate the deviation degree, and fuses the aboveground and underground data, which can accurately reflect the growth coordination of the two, and can warn the root problem through the correlation degree change before the leaves appear abnormally, solve the hysteresis of monitoring only the aboveground part, provide comprehensive and timely health evaluation for greenhouse planting, and improve the accuracy of intelligent agricultural monitoring.

[0142] In one case of the embodiment, the growth correlation of the historical healthy crops in the same period is determined based on historical data, and a healthy fluctuation feature library is established, including:

[0143] Based on historical data, the healthy crop data of different growth stages in the same period is determined, specifically including: collecting growth data and environmental data of historical crops in the same period, screening out the identified healthy crop data, dividing the stages according to the growth days, such as the seedling stage (1-30 days), the jointing stage (31-60 days) and other growth stages, for the healthy data of each stage, classifying according to the indexes (such as leaf number, area, plant height, contour features, etc.), using the median algorithm to statistically process the same index values of all healthy samples in the stage, and the obtained result is the index data of the healthy crops in the same period and the growth stage, and thus the healthy crop data of different growth stages in the same period can be determined;

[0144] The growth stage matching degree and the environmental factor similarity of the healthy crop data and the current crop are calculated, and the healthy sample screening index is generated, specifically including: for the growth stage matching degree, the growth stage matching degree is calculated, specifically: calculating the difference between the current crop growth days and the historical healthy sample growth days, such as the 15th day of the current crop in a certain growth stage, and the 18th day of a certain historical sample in the same stage, the difference is 3 days, subtract 1 by (the absolute value of the difference divided by the typical duration of the growth stage), and then take the maximum value with 0, which can obtain the growth stage matching degree, and can ensure that the matching degree is in the range of 0-1 and reflect the continuity; for the environmental factor similarity, specifically: taking the temperature, humidity and other environmental factors, calculating the factor mean value of the current and historical sample in the same period, for each factor, subtract 1 by (the absolute value of the mean difference divided by the historical standard deviation of the factor), and limit the result to 0-1, calculate the arithmetic mean of all factor similarities, and obtain the environmental factor similarity; set the growth stage matching degree weight to 0.6 and the environmental factor similarity weight to 0.4, multiply the growth stage matching degree and the environmental factor similarity by the corresponding weight and sum them up, and the result is the healthy sample screening index;

[0145] Based on the healthy sample screening index, the target historical healthy crop sample is screened out, and the growth correlation feature matrix is obtained, specifically including: setting a screening threshold (0.8), retaining the historical samples with a healthy sample screening index not lower than the screening threshold, taking the historical samples as target samples, extracting the above-ground indexes (leaf number, plant height, etc.) and underground indexes (root activity of each layer) of these target samples, sorting the samples according to the growth time, taking the above-ground indexes of each sample as the rows of the matrix in turn, and taking the underground indexes as the columns of the matrix in turn, the index data of different time points correspond to the numbers in different positions of the matrix, and the matrix formed by the arrangement and combination is the growth correlation feature matrix;

[0146] The growth correlation feature matrix is analyzed and a health fluctuation feature library is established, specifically including: splitting the growth correlation feature matrix according to the growth stage to obtain a sub-matrix of each stage, calculating the mean and standard deviation of each index of the aboveground and underground indexes of each stage, taking the mean plus or minus 3 times the standard deviation as the normal fluctuation range of the index, classifying each sub-matrix according to the growth stage and index type (aboveground underground), and recording the mean, standard deviation and fluctuation range of each index, that is, establishing a health fluctuation feature library.

[0147] By combining the growth stage matching degree and the environmental factor similarity to screen the target sample, it is ensured that the historical healthy sample screened is highly similar to the current crop, and the growth correlation feature matrix can accurately reflect the correlation rule of the aboveground and underground indexes. Meanwhile, the health fluctuation feature library effectively improves the forward-looking of the monitoring. The library covers the aboveground and underground index correlation data of different growth stages, which can be quickly compared with the real-time data of the current crop. The root problem is detected in advance through index fluctuation anomaly before the appearance of symptoms such as wilting of leaves, which makes up for the lag of monitoring only the aboveground part. The greenhouse grower can timely master the real growth status of the crop, and the monitoring effect of intelligent agriculture is achieved.

[0148] In one case of the embodiment, the crop health condition is determined according to the crop health correlation index, including:

[0149] The health threshold interval is determined based on the aboveground physiological advance value, the root depth activity coefficient and the historical data, specifically including: selecting healthy crop samples of the last 5 growth cycles from the historical data, extracting the aboveground physiological advance value and the root depth activity coefficient thereof, calculating the mean and standard deviation of the aboveground physiological advance value and the root depth activity coefficient, taking the mean plus or minus 3 times the standard deviation as the normal fluctuation range of the aboveground physiological advance value and the root depth activity coefficient respectively, and taking the intersection of the two as the health threshold interval.

[0150] The health threshold interval is compared with the crop health correlation index to obtain the crop health condition, specifically including: comparing the crop health correlation index with the health threshold interval, if the crop health correlation index is within the health threshold interval, it is determined to be healthy, and if the crop health correlation index is lower than the lower limit of the health threshold interval or higher than the upper limit of the health threshold interval, it is determined to be abnormal growth.

[0151] The health threshold interval is obtained by calculating the mean value and standard deviation through the aboveground physiological advanced value of the historical health sample and the root depth activity coefficient, and the normal fluctuation range of the aboveground and underground growth indexes is considered, the crop health can be directly determined by comparing the health correlation index with the interval, the subjectivity of the determination of the crop health is avoided, the monitoring lag problem is effectively solved, the timeliness of the health determination is improved, the health threshold interval fuses the aboveground physiological advanced value and the root depth activity coefficient, the two indexes can reflect the crop physiological change in advance, when the health correlation index exceeds the interval, the growth abnormality can be determined before the appearance of the leaf wilting and other symptoms, the grower can take intervention in time, the monitoring and determination method based on the big data can capture the real growth state of the crop in time, and the monitoring effect of the smart agriculture is ensured.

[0152] Although the embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made thereto without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. The smart agricultural monitoring system based on big data is characterized by: include: The division unit is used to obtain the maximum diameter of the ground and the maximum extension radius of the root system of each crop in the greenhouse, determine the dynamic grid size of the single crop, and divide the planting area into multiple dynamic grids based on the dynamic grid size, each grid covering the single crop and the extension range of its root system; The integration unit is used to collect image data, environmental data, and soil data at different depths of individual crops in each grid in real time, and integrate the three pre-processed data to obtain a big data fusion matrix for individual crops; The analysis unit is used to analyze the big data fusion matrix to obtain the aboveground physiological advance value and the root depth activity coefficient, including: Perform image extraction on the big data fusion matrix to obtain the time series image change rate; Determine the environmental physiological response coefficient based on the time series image change rate and the comprehensive environmental impact index; Screen the soil characteristic tensor and generate shallow soil impact weights; Obtain historical data of crops, analyze the historical data, and obtain the historical mean of image feature vectors at the same growth stage; The historical mean is corrected based on the comprehensive environmental impact index of the same period to obtain the dynamic physiological baseline vector; The time series image change rate, environmental physiological response coefficient, and shallow soil impact weight are integrated to obtain the current physiological characteristic prediction value; Compare the current physiological characteristic prediction value with the dynamic physiological reference vector to obtain the advance prediction deviation; Calculate the lead prediction deviation to generate the ground physiological lead value; The soil characteristic tensor is parsed to extract the nutrient availability and water conductivity of the soil at each depth, and the depth soil activity weight is obtained; Based on the dynamic changes of temperature and light duration in the comprehensive environmental impact index, the depth soil activity weight is corrected to generate a depth correction coefficient; Extract the image feature vector to obtain the aboveground growth rate characteristics, and fuse the aboveground growth rate characteristics with the depth correction coefficient to obtain the preliminary root activity coefficient; Determine the root activity baseline value for the same period in history based on historical data, and calculate the deviation rate between the preliminary root activity coefficient and the root activity baseline value; The preliminary root activity coefficient was calibrated based on the deviation rate to obtain the root depth activity coefficient; The correlation unit is used to perform correlation analysis on the aboveground physiological advance value and the root depth activity coefficient to obtain the crop health correlation index, including: Decompose the aboveground physiological advance value and root depth activity coefficient to generate a multi-dimensional growth coordination tensor; The matching degree of above-ground and underground characteristic components in the growth coordination tensor is calculated to obtain the dynamic correlation between the growth cycle and spatial distribution. Determine the growth correlation of healthy crops in the same period based on historical data and establish a health fluctuation feature library; Calculate the difference between the current dynamic correlation and the health fluctuation feature library to generate the health feature deviation; The dynamic correlation degree and health characteristic deviation degree were analyzed to obtain the crop health correlation index; The evaluation unit is used to determine the health status of crops based on crop health correlation indexes.

2. The smart agriculture monitoring system based on big data according to claim 1 is characterized in that: Determine dynamic grid size for individual crops, including: The maximum above-ground diameter and the maximum root extension radius of each crop were calculated to obtain the proportional coefficient; Determine the redundant root coverage radius based on the proportional coefficient and the maximum above-ground diameter of each crop plant; Taking a single crop plant as the center, the planting area is divided into square grids with a side length of twice the root redundancy coverage radius to generate a dynamic grid size for a single crop plant.

3. The smart agriculture monitoring system based on big data according to claim 1 is characterized in that: The three pre-processed data are integrated to obtain the big data fusion matrix of individual crops, including: Extract image data to obtain image feature vector; Analyze the correlation between each data in the environmental data, and integrate the correlation to obtain a comprehensive environmental impact index; Analyze soil data at different depths to obtain soil characteristic tensors; The image feature vector, comprehensive environmental impact index and soil feature tensor are fused to generate a big data fusion matrix of a single crop.

4. The smart agriculture monitoring system based on big data according to claim 1 is characterized in that: Determine the baseline root activity value for the same period based on historical data, including: Get the current date and the crop planting date, and determine the growth stage matching factor based on the two dates; All healthy crop samples within the growth stage matching factor of ±5% were screened from historical data to construct a spatiotemporal similarity sample set; Cluster analysis is performed on the spatiotemporally similar sample sets to obtain the basic activity distribution interval; Extract environmental data and construct environmental correction tensor; Adjust the basic activity distribution interval according to the environmental correction tensor to generate the environmental adaptability benchmark interval; The environmental adaptability benchmark interval was disassembled to separate the trend part and the cycle part, and the root activity benchmark value for the same historical period was calculated.

5. The smart agriculture monitoring system based on big data according to claim 1 is characterized in that: Based on historical data, we determine the growth correlation of healthy crops in the same period in history and establish a health fluctuation feature library, including: Based on historical data, determine the healthy crop data at different growth stages during the same period in history; Calculate the matching degree between healthy crop data and the growth stage of current crops and the similarity of environmental factors to generate a healthy sample screening index; Based on the healthy sample screening index, target historical healthy crop samples are screened to obtain the growth correlation feature matrix; The growth-related characteristic matrix was analyzed and a health fluctuation characteristic library was established.

6. The smart agriculture monitoring system based on big data according to claim 1 is characterized in that: Determine crop health status based on crop health related indices, including: Determine the health threshold interval based on aboveground physiological advance values, root depth activity coefficients, and historical data; The health threshold interval is compared with the crop health association index to obtain the crop health status.

Images