A crop planting management system that combines above-ground and underground data

Abstract

This invention discloses a crop planting management system that combines above-ground and underground data, including a planting area acquisition module, a crop acquisition module, an above-ground data acquisition module, an underground data acquisition module, a data analysis module, a data processing module, and an information sending module. The planting area acquisition module is configured to collect the area information and geographic boundary coordinates of the planting area. The crop acquisition module is configured to collect the crop type information and growth stage identifiers. The data analysis module generates above-ground data acquisition instructions and underground data acquisition instructions based on the area information, geographic boundary coordinates, and crop type information. The above-ground data acquisition module collects above-ground data from the planting area in response to the above-ground data acquisition instructions; the underground data acquisition module collects underground data from the planting area in response to the underground data acquisition instructions. This invention achieves intelligent crop planting management by combining above-ground and underground data.

Description

Technical Field

[0001] This invention relates to the field of crop management, specifically to a crop planting management system that combines above-ground and underground data. Background Technology

[0002] Crop cultivation requires comprehensive control of multiple dimensions such as soil environment and crop growth stage. Meeting the water and fertilizer needs of crops through reasonable irrigation and precise fertilization, and ensuring yield and quality, is one of the core links in agricultural production.

[0003] To improve planting efficiency and management precision, crop planting management systems have emerged. These systems collect information such as planting area, geographical boundaries, crop type, and growth stage, combining above-ground and below-ground data to assist in planting decisions. However, traditional systems suffer from weak data correlation and limited evaluation dimensions, making it difficult to accurately match the dynamic needs of crops. Therefore, this paper proposes a crop planting management system that combines above-ground and below-ground data. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a crop planting management system that combines above-ground and underground data, including a planting area data acquisition module, a crop data acquisition module, an above-ground data acquisition module, an underground data acquisition module, a data analysis module, a data processing module, and an information transmission module.

[0005] The planting area data collection module is configured to collect the area information and geographic boundary coordinates of the planting area;

[0006] The crop collection module is configured to collect crop type information and growth stage identifiers;

[0007] The data analysis module generates above-ground data acquisition instructions and underground data acquisition instructions based on the area information, geographic boundary coordinates, and type information.

[0008] The above-ground data acquisition module collects above-ground data of the planting area in response to the above-ground data acquisition command;

[0009] The underground data acquisition module collects underground data from the planting area in response to underground data acquisition commands;

[0010] The data processing module processes above-ground data to generate above-ground assessment information, processes underground data to generate underground assessment information, analyzes underground data to obtain water and fertilizer demand index, and integrates above-ground and underground assessment information to generate comprehensive assessment information.

[0011] The calculation of the above-ground assessment information, below-ground assessment information, and water and fertilizer demand index is all based on the growth stage identifier, and is achieved by matching the physiological parameter thresholds of the corresponding growth stage in the crop demand database.

[0012] The information sending module will send comprehensive evaluation information to the user terminal.

[0013] Furthermore, the data processing module includes:

[0014] Real-time verification unit: Receives user feedback data and calculates the dynamic verification factor δ based on the comprehensive evaluation information, specifically:

[0015]

[0016] Model iteration unit, used in ||E user -E integrated When ||<∈ and δ<θ, crop demand library parameter calibration is triggered for model iteration;

[0017] E user The actual growth status is manually scored by the user, on a scale of 0-100; E integrated The system generates a comprehensive evaluation value, on a scale of 0-100; max(·) is the maximum value function; ∈ is the data credibility threshold, which is the maximum allowable absolute deviation between the user rating and the system evaluation value, and is the boundary value for judging the reliability of the data source; θ is the credibility threshold, which is used as the lower limit of the consistency between the comprehensive evaluation information and the user feedback.

[0018] The real-time verification unit is connected to the UAV re-inspection system:

[0019] When ||E user -E integrated When ||≥∈ and δ<θ, the ground data acquisition module is controlled to re-acquire data in the abnormal area.

[0020] Furthermore, the data processing module also integrates a multimodal disaster prediction submodule, which is used to construct a heat wave / freezing damage disaster risk index R based on historical meteorological data and real-time ground data. d The specific process is as follows:

[0021]

[0022] T canopy The infrared thermometer value of the canopy; T opt The optimal temperature for the current growth stage in the crop demand pool; W leaf This refers to the water content of the leaves. σ is the rate of change in leaf water content; σ is the temperature deviation weighting coefficient, with heat wave disaster >0 and frost damage disaster <0; λ is the water change sensitivity coefficient.

[0023] Furthermore, the crop planting management system also includes a disaster response unit, a protection intensity calculation module, and an equipment driving module;

[0024] The disaster response unit is used to connect the shade net and the anti-frost fan equipment;

[0025] The protection strength calculation module is used to calculate the heat wave / freezing damage risk index R. d Calculate the protection strength P:

[0026]

[0027] In the formula, R safe R represents the crop's tolerance safety threshold. critical Critical threshold for irreversible crop damage; μ is the maximum power coefficient of the equipment;

[0028] The device driver module is used to convert the protection strength P into a device control signal.

[0029] Furthermore, the crop planting management system also includes a water and fertilizer early warning module, which is used to control irrigation / fertilization equipment according to the water and fertilizer demand index. When the water and fertilizer demand index exceeds the dynamic threshold, the irrigation / fertilization equipment control signal is triggered.

[0030] The control signal strength I of irrigation / fertilizer equipment control signals is calculated based on the relative deviation between the actual value and the target value:

[0031]

[0032] E actual E represents the real-time water and fertilizer demand index. target γ is the target water and fertilizer index; γ is the soil type correction factor.

[0033] Furthermore, the disaster response unit executes a tiered response strategy, the specific content of which is as follows:

[0034] R d ∈[R safe R critical When [the condition is met], the basic protection mode is activated;

[0035] R d ≥R critical At that time, activate emergency response mode.

[0036] Furthermore, the data analysis module also includes a spatial heterogeneity analysis unit, a dynamic partitioning engine, and an adaptive sampling unit:

[0037] Spatial heterogeneity analysis unit, used to calculate the coefficient of variation C of raster regions based on historical subsurface data, such as soil electrical conductivity and organic matter content. v The specific process is as follows:

[0038]

[0039] In the formula, σ is the standard deviation of soil parameters within the grid area, reflecting spatial dispersion; μ is the average value of soil parameters within the grid area.

[0040] Historical underground data is stored and categorized by growth stage identifiers, and during analysis, a subset of historical data is matched to the current crop growth stage.

[0041] Dynamic partitioning engine, used in C v When the value is greater than τ, the raster is split into sub-partitions, and each sub-partition is assigned an independent identifier code.

[0042] Adaptive sampling unit, used to sample according to the coefficient of variation C v Adjust the sampling density of the underground data acquisition module:

[0043]

[0044] In the formula, ρ0 is the basic sampling density; τ is the coefficient of variation threshold; and k is the density adjustment factor, ranging from 1.5 to 2.0.

[0045] Furthermore, the data analysis module also includes:

[0046] Spatiotemporal correlation predictor: Establishing the above-ground and below-ground data correlation matrix Mt for a raster region:

[0047]

[0048] NDVI is the Normalized Difference Vegetation Index, ranging from [-1, 1]; EC is the Soil Electrical Conductivity; T canopy Canopy temperature; N root α represents root density; 11 α 1n α m1 With α mn The weight coefficients are obtained by training with historical data and need to be optimized by grouping according to the growth stage identifier to adapt to the physiological characteristics of crops at different stages.

[0049] The strategy generator is used to predict the optimal acquisition strategy based on the correlation matrix Mt and historical above-ground and below-ground data using an LSTM time series model.

[0050] S t+1 =LSTM(M t M t-1 M t-Δt );

[0051] M t-1 M is the correlation matrix at the previous time step; t-Δt The historical time-time correlation matrix is ​​given, where Δt is the time window; LSTM(·) long short-term memory neural network operation; S t+1The optimal acquisition strategy is predicted.

[0052] The resource optimizer is used to calculate energy consumption cost E based on the forecasting strategy. cost The specific process is as follows:

[0053] E cost =β1·D flight +β2·N sensor ;

[0054] D flight N represents the flight distance of the drone. sensor β1 represents the number of activated underground sensors; β2 represents the flight distance weighting coefficient; and β3 represents the sensor activation weighting coefficient.

[0055] Furthermore, the information sending module includes an AR visualization interface:

[0056] A three-dimensional growth pattern is generated by overlaying underground root distribution data with above-ground canopy data.

[0057] Spatial heat map marking areas with abnormal water and fertilizer demand indices.

[0058] The beneficial effects of this invention are reflected in:

[0059] It has achieved full-dimensional monitoring capabilities, enabling three-dimensional monitoring of crop growth status through the collaborative collection and analysis of above-ground and underground data, thus overcoming the limitations of traditional single data sources.

[0060] It can accurately warn and respond to disasters, integrate multimodal data such as temperature deviation and moisture change rate, construct a dynamic heat wave / freezing damage disaster risk index, improve the accuracy of extreme weather warnings, automatically trigger graded protection strategies based on disaster risk level, and dynamically adjust equipment power to achieve efficient resource utilization.

[0061] It achieves intelligent water and fertilizer management, dynamically calculates the intensity of water and fertilizer control signals by combining soil type correction factors, realizes precise irrigation / fertilization on demand, avoids resource waste, and visualizes abnormal areas through spatial heat maps to help quickly locate problems.

[0062] The system also implemented model self-optimization and introduced a user feedback verification mechanism. When there is a significant difference between the system assessment and human judgment, it automatically triggers model calibration or drone re-inspection to continuously improve the accuracy of the assessment. It also uses historical data to train the association matrix weights to enhance the reliability of predictions.

[0063] Furthermore, based on the spatial variation coefficient of soil parameters, the underground sampling density is dynamically adjusted to improve accuracy in sparse data areas and reduce energy consumption in homogeneous areas. The optimal acquisition strategy is predicted through the LSTM model, which optimizes the UAV flight path and the number of activated sensors, significantly reducing monitoring costs. Attached Figure Description

[0064] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0065] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0066] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0067] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0068] like Figure 1 As shown, a crop planting management system that combines above-ground and underground data includes a planting area data acquisition module, a crop data acquisition module, an above-ground data acquisition module, an underground data acquisition module, a data analysis module, a data processing module, and an information sending module.

[0069] The planting area data collection module is configured to collect the area information and geographic boundary coordinates of the planting area;

[0070] The crop collection module is configured to collect crop type information and growth stage identifiers;

[0071] The data analysis module generates above-ground data acquisition instructions and underground data acquisition instructions based on the area information, geographic boundary coordinates, and type information.

[0072] The above-ground data acquisition module collects above-ground data of the planting area in response to the above-ground data acquisition command;

[0073] The underground data acquisition module collects underground data from the planting area in response to underground data acquisition commands;

[0074] The data processing module processes above-ground data to generate above-ground assessment information, processes underground data to generate underground assessment information, analyzes underground data to obtain water and fertilizer demand index, and integrates above-ground and underground assessment information to generate comprehensive assessment information.

[0075] The calculation of the above-ground assessment information, below-ground assessment information, and water and fertilizer demand index is all based on the growth stage identifier, and is achieved by matching the physiological parameter thresholds of the corresponding growth stage in the crop demand database.

[0076] The information sending module will send comprehensive evaluation information to the user terminal.

[0077] The data processing module includes:

[0078] Real-time verification unit: Receives user feedback data and calculates the dynamic verification factor δ based on the comprehensive evaluation information, specifically:

[0079]

[0080] Model iteration unit, used in ||E user -E integrated When ||≤∈ and δ<θ, the crop demand library parameter calibration is triggered to perform model iteration;

[0081] E user The actual growth status is manually scored by the user, on a scale of 0-100; E integrated The system generates a comprehensive evaluation value on a scale of 0-100; max(·) is the function to take the maximum value.

[0082] The real-time verification unit is connected to the UAV re-inspection system:

[0083] When ||E user -E integrated When ||≥∈ and δ<θ, control the ground data acquisition module to re-acquire data in the abnormal area;

[0084] By calculating the validation factor δ in real time, the deviation between system evaluation and human experience is identified, avoiding model distortion. Based on the type of deviation, an optimization strategy is intelligently selected, namely model iteration or data resampling, balancing efficiency and accuracy.

[0085] For example, the overall system evaluation value: E integrated =85

[0086] User rating: E user =70;

[0087] Right now

[0088] Preset thresholds: ∈ = 10, θ = 0.85;

[0089] |70-85|=15>∈, and δ=0.823<θ, at this time the drone re-inspection is automatically started;

[0090] When there is a small deviation, E user -E integrated=8 < ∈, δ = 0.8 < θ, at this point, model parameter calibration is performed;

[0091] When significant local anomalies exist, E user -E integrated =15≥∈, δ=0.82<θ, at this time, the drone is needed to accurately re-inspect the abnormal area.

[0092] The data processing module also integrates a multimodal disaster prediction submodule, which is used to construct a heat wave / freezing damage disaster risk index R based on historical meteorological data and real-time ground data. d The specific process is as follows:

[0093]

[0094] T canopy The infrared thermometer value of the canopy; T opt The optimal temperature for the current growth stage in the crop demand pool;

[0095] W leaf This refers to the water content of the leaves. σ is the rate of change in leaf water content; σ is the temperature deviation weighting coefficient (heat wave disaster > 0, frost damage disaster < 0); λ is the water change sensitivity coefficient;

[0096] Historical meteorological data needs to be filtered according to the growth stage identifier to select the corresponding historical threshold.

[0097] The sign and magnitude of δ are used to automatically distinguish disaster types and quantify the degree of risk:

[0098] When σ > 0, the focus is on heat wave warnings, such as when σ = 0.8;

[0099] When σ < 0, the focus is on frost damage warning, such as σ - 0.6;

[0100] Measured data: T canopy =42℃, T opt =30℃;

[0101] The leaves lose water rapidly;

[0102] Parameters: σ=0.8, λ=1.2, R d =0.8×|42-30|+1.2×(-0.3)=9.24, indicating high risk;

[0103] When R d >R safe , such as R safe =5, the system immediately activates the shade net and adjusts irrigation;

[0104] This sensor detects abnormal changes in moisture content and provides early warnings even before absolute moisture content.

[0105] On the eve of the cold wave: T canopy =0.2℃, T opt =8℃;

[0106] This indicates that condensation of dew suggests a risk of freezing.

[0107] R d = -0.6 × |2 - 8| + 0.9 × 0.4 = -3.24, a negative value triggers a frost damage response;

[0108] σ<0 and R d <0, the anti-frost fan starts in advance to prevent the blades from icing;

[0109] λ is adjusted according to crop type. For example, rice prefers moist conditions: λ = 0.5, which weakens the impact of water fluctuations.

[0110] Cacti are drought resistant, such as λ = 1.8, which means they are more sensitive to changes in water availability.

[0111] The crop planting management system also includes a disaster response unit, a protection strength calculation module, and an equipment driver module;

[0112] The disaster response unit is used to connect the shade net and the anti-frost fan equipment;

[0113] The protection strength calculation module is used to calculate the heat wave / freezing damage risk index R. d Calculate the protection strength P:

[0114]

[0115] In the formula, R safe R represents the crop's tolerance safety threshold. critical Critical threshold for irreversible crop damage; μ is the maximum power coefficient of the equipment;

[0116] The device driver module is used to convert the protection strength P into a device control signal;

[0117] The heat wave / freezing damage risk index R d The linear mapping is to the equipment power P, enabling more refined control with stronger protection as the risk increases;

[0118] Heatwave Disaster Index:

[0119] R d =7.5, which is the heatwave risk index;

[0120] R safe =5;

[0121] R critical=10;

[0122] μ = 1000, maximum power of the shade net;

[0123]

[0124] At this time, the system automatically sets the power of the shade net to 50%, 500W / 1000W;

[0125] This approach avoids excessive energy consumption during low-risk periods while ensuring effective protection against medium-risk situations.

[0126] It also enables emergency response to critical disasters, when R d ≥R critical When the time comes, it automatically switches to the maximum power mode P=μ to prevent irreversible damage;

[0127] Extreme freezing damage index:

[0128] R d = -8.2, σ<0 indicates a frost damage pattern;

[0129] R safe =-5;

[0130] R critical =-8;

[0131] μ = 3000, maximum fan power: 3000W;

[0132] At this point, it is determined that: R d =-8.2<R critical =-8;

[0133] Exceeding μ;

[0134] The system automatically activates at this time, Emergency Response Mode R. d <R critical ;

[0135] At this point, the formula limit is ignored, and the system is forced to run at full power, while the backup heater is activated.

[0136] This enables dynamic adaptation, automatically adjusting equipment power based on real-time risks to avoid energy waste during periods of low risk.

[0137] Catastrophe fallback, when R d When the critical threshold is exceeded, maximum power is forced and backup equipment is activated.

[0138] Equipment lifespan optimization reduces the operating time of equipment under low-risk conditions, extending the lifespan of shade nets / fans.

[0139] The crop planting management system also includes a water and fertilizer early warning module, which is used to control irrigation / fertilization equipment based on the water and fertilizer demand index. When the water and fertilizer demand index exceeds the dynamic threshold, the control signal of the irrigation / fertilization equipment is triggered. The calculation model of the water and fertilizer demand index needs to input the growth stage identifier in order to call the water and fertilizer demand benchmark value of the corresponding stage.

[0140] The process of obtaining the water and fertilizer demand index is as follows:

[0141]

[0142] In the formula, n represents the number of parameter combinations or influencing factors involved in the calculation, which is a positive integer. The specific value depends on the classification of different influencing factors considered when actually constructing the model.

[0143] w i It is the weight coefficient corresponding to the i-th influencing factor, used to reflect the importance of different influencing factors in calculating the water and fertilizer demand index. The weight can be determined by experimental data, historical experience or model training.

[0144] f i It is a function of soil moisture, nutrient content, and growth stage, used to characterize how these factors work together to affect water and fertilizer requirements in the i-th case. For example, it can be a functional relationship obtained by fitting through regression analysis, neural networks, etc., and the input soil environment and crop growth stage information is transformed into contribution values ​​to water and fertilizer requirements.

[0145] Soil moisture is an indicator that reflects the water content in the soil. Common ways of expressing it include volumetric water content and gravimetric water content. It is one of the basic environmental factors that affect the water and fertilizer requirements of crops.

[0146] Nutrients refer to the nutrient content in the soil, such as the content of major nutrients like nitrogen, phosphorus, and potassium. Nutrient status directly affects the crop's fertilizer requirements and the relationship between fertilizer and water requirements. For example, when nutrients are deficient, water may need to be added to promote fertilizer absorption.

[0147] Growth represents the different growth stages of crops, such as seedling stage, flowering stage, and maturity stage. The water and fertilizer requirements of crops vary greatly at different growth stages, which is a key biological factor that must be considered when calculating the water and fertilizer requirement index.

[0148] The control signal strength I of irrigation / fertilizer equipment control signals is calculated based on the relative deviation between the actual value and the target value:

[0149]

[0150] E actualE represents the real-time water and fertilizer demand index. target The target water and fertilizer index is γ; γ is the soil type correction factor.

[0151] By dynamically calculating the control intensity I based on the relative deviation, precise resource allocation is achieved by supplementing only what is lacking.

[0152] For example, in the monitoring and management process of cornfields:

[0153] The soil type is sandy soil, γ = 1.5;

[0154] Target water and fertilizer index, E target =80;

[0155] Real-time monitoring value, E actual =60;

[0156] That is, a medium-intensity signal;

[0157] At this time, the system automatically turns on the drip irrigation equipment and replenishes water at a maximum flow rate of 37.5%.

[0158] To avoid the waste of water resources caused by traditional fixed-duration irrigation, sandy soil is prone to seepage.

[0159] The disaster response unit implements a tiered response strategy, the specific content of which is as follows:

[0160] R d ∈[R safe R critical When [the condition is met], the basic protection mode is activated;

[0161] R d ≥R critical When this occurs, activate emergency response mode;

[0162] Risk classification and precise resource allocation have been achieved.

[0163] The basic protection mode enables low-power operation of core equipment, saving energy, while the emergency response mode activates all equipment and backup systems, providing a safety net in case of disaster.

[0164] The data analysis module also includes a spatial heterogeneity analysis unit, a dynamic partitioning engine, and an adaptive sampling unit.

[0165] Spatial heterogeneity analysis unit, used to calculate the coefficient of variation C of raster regions based on historical subsurface data, such as soil electrical conductivity or organic matter content. v The specific process is as follows:

[0166]

[0167] In the formula, σ is the standard deviation of soil parameters within the grid area, reflecting the spatial dispersion;

[0168] μ represents the average value of soil parameters within the grid area;

[0169] Dynamic partitioning engine, used in C v When the value is greater than τ, the raster is split into sub-partitions, and each sub-partition is assigned an independent identifier code.

[0170] Adaptive sampling unit, used to sample according to the coefficient of variation C v Adjust the sampling density of the underground data acquisition module:

[0171]

[0172] In the formula, ρ0 is the basic sampling density; τ is the coefficient of variation threshold; and k is the density adjustment factor, ranging from 1.5 to 2.0.

[0173] Through the coefficient of variation C v Automatically locate soil parameters, such as electrical conductivity or areas with drastic fluctuations in organic matter.

[0174] System scan revealed:

[0175] Soil electrical conductivity in area A is C v =8%, which is homogeneous;

[0176] Area B, Section C v =32%, severely uneven;

[0177] Automatic determination: C v >τ, preset threshold τ = 15%;

[0178] At this point, area B is divided into three partitions: B1, B2, and B3.

[0179] Salt-alkali plaques were found in the B2 subregion, indicating abnormal conductivity, and targeted improvements were made.

[0180] The data analysis module also includes:

[0181] Spatiotemporal correlation predictor: Establishing the above-ground and below-ground data correlation matrix Mt for a raster region:

[0182]

[0183] NDVI is the Normalized Difference Vegetation Index, ranging from [-1, 1]; EC is the Soil Electrical Conductivity; T canopy Canopy temperature; N root α represents root density; 11 α 1n α m1 With α mn These are the weighting coefficients, obtained through training with historical data;

[0184] The strategy generator is used to predict the optimal acquisition strategy based on the correlation matrix Mt and historical above-ground and below-ground data using an LSTM time series model.

[0185] S t+1 =LSTM(M t M t-1 ,...,M t-Δt );

[0186] M t-1 M is the correlation matrix at the previous time step; t-Δt The historical time-time correlation matrix is ​​given, where Δt is the time window; LSTM(·) long short-term memory neural network operation; S t+1 The optimal acquisition strategy is predicted.

[0187] The resource optimizer is used to calculate energy consumption cost E based on the forecasting strategy. cost The specific process is as follows:

[0188] E cost =β1·D flight +β2·N sensor ;

[0189] D flight N represents the flight distance of the drone. sensor β1 represents the number of activated underground sensors; β2 represents the flight distance weighting coefficient; β3 represents the sensor activation weighting coefficient.

[0190] For example, leaf curling occurs during the corn tasseling stage in July;

[0191] Traditional methods require manual inspection of the entire farmland, blindly digging up roots to check the soil, and uniform irrigation, which is time-consuming, damages the root system, and wastes water resources.

[0192] Spatiotemporal correlation matrix analysis:

[0193] Detected: Above ground, canopy temperature 42℃, normal value ≤35℃, NDVI index 0.48, normal >0.7; Below ground, root density at a depth of 20cm decreased by 50%, soil moisture content 8%, critical value 12%; Correlation matrix reveals key patterns:

[0194]

[0195] This means that the diagnosis is deep soil drought, not disease, or shallow water shortage.

[0196] LSTM predicts the optimal strategy:

[0197] Input historical data: correlation matrix of the past 3 days + weather forecast

[0198] Output:

[0199] St+1 = Immediately scan the B7 subregion 30-50cm underground;

[0200] At this point, the deep soil layer where water loss is fastest is identified;

[0201] Energy consumption optimization execution:

[0202] The drone activated eight deep-soil sensors after flying only 3km.

[0203] Energy consumption calculation:

[0204] E cost =0.5×0.3+0.1×0.8=2.3kWh.

[0205] The information sending module includes an AR visualization interface:

[0206] A three-dimensional growth pattern is generated by overlaying underground root distribution data with above-ground canopy data.

[0207] Spatial heat map marking areas with abnormal water and fertilizer demand indices.

[0208] The specific process for obtaining above-ground assessment information is as follows:

[0209] Data Acquisition and Preprocessing:

[0210] Ground data is acquired through the ground data acquisition module and includes:

[0211] Canopy morphology data, such as plant height and leaf area index, are collected by image recognition or sensors;

[0212] Canopy physiological data, such as leaf temperature and chlorophyll fluorescence, are collected by infrared / spectroscopy equipment; meteorological data, such as light intensity and air temperature and humidity, are collected by environmental sensors.

[0213] Combining the spatial heterogeneity analysis unit and the adaptive sampling unit, the collected data is processed as follows: spatial calibration, matching planting area raster partitions with geographic boundary coordinates, such as sub-partition identifier association; temporal alignment, based on a spatiotemporal correlation predictor, unifying data collection timestamps;

[0214] Anomaly filtering utilizes a real-time verification unit to remove noisy data that deviates from the normal range, such as sudden changes in leaf temperature.

[0215] In the construction of the above-ground assessment model: LSTM time series model and multimodal disaster prediction submodule, a two-dimensional assessment model is built.

[0216] Growth status assessment involves feature extraction from canopy morphology and physiological data, such as leaf area index fitting and chlorophyll content inversion. The crop demand library model, trained through the data processing module, outputs a growth status score E.canopy For example, a score of 0-100 can reflect health and growth.

[0217] Disaster risk assessment combines meteorological data with heat wave / freezing risk indices to calculate the current disaster threat value faced by the canopy and generate risk level labels (such as safe, warning, emergency).

[0218] By integrating growth status scores and disaster risk levels, structured above-ground assessment information is generated.

[0219] Above-ground assessment information {E canopy R d , Coordinates of the abnormal region, matching degree of the growth stage;

[0220] The matching degree of growth stage is calculated by comparing the data collected by the crop acquisition module with the optimal parameters of the crop demand database, such as the deviation rate between the current leaf area index and the standard value of the seedling stage.

[0221] The specific process of processing underground data to generate underground assessment information is as follows:

[0222] Subsurface data is acquired through a subsurface data acquisition module and includes:

[0223] Soil physical data, such as moisture and bulk density, are collected by soil sensors;

[0224] Soil chemical data, such as nutrient content and electrical conductivity, are collected by rapid soil testing equipment;

[0225] Root distribution data, such as root length density and depth, are obtained by ground-penetrating radar or image reconstruction technology.

[0226] Based on the spatial heterogeneity analysis unit, the underground data is processed as follows:

[0227] Rasterized mapping, aligned with planting area and geographic boundary coordinates, generates a heat map of the spatial distribution of soil parameters;

[0228] Deep stratification processing is used to split the data according to the main distribution layer of the root system, such as 0-20cm, 20-40cm;

[0229] Dynamic sampling optimization utilizes adaptive sampling units to supplement sampling in areas with high coefficients of variation, such as nutrient gradient zones.

[0230] Construction of underground assessment model:

[0231] By combining the data processing module with the spatiotemporal correlation predictor, a soil-root synergistic assessment model is constructed:

[0232] Soil fertility assessment involves normalizing soil nutrient and moisture data, matching optimal parameters for the current growth stage (such as the nitrogen requirement threshold during seedling stage) using a crop requirement database, and outputting the soil fertility index F. soil For example, a score of 0-100 reflects the fertilizer supply capacity.

[0233] Root health assessment involves morphological analysis of root distribution data, such as root surface area and branch number, combined with multimodal data, such as the correlation between canopy water content and root water uptake, to calculate the root vigor index V. root This reflects the root system's absorption efficiency.

[0234] Subsurface assessment information is output, integrating soil fertility index and root vitality index to generate structured subsurface assessment information: Subsurface assessment information = {F soil V root Spatial distribution of soil parameters, root stratification health;

[0235] The spatial distribution of soil parameters is output as rasterized heat map data, and the root health of the layers is marked according to the depth layer, such as the vitality index of the 0-20cm layer and the vitality index of the 20-40cm layer.

[0236] The specific process of fusing above-ground and underground assessment information to generate comprehensive assessment information is as follows:

[0237] Extracting common dimensions from above-ground and below-ground data: Growth stage correlation: By identifying crop growth stages, a unified benchmark for above-ground and below-ground growth stages is established;

[0238] Based on the geographical boundary coordinates and grid partitions of the planting area, align the above-ground anomaly areas with the below-ground soil parameter anomaly areas.

[0239] Water and fertilizer demand correlation: through the defined water and fertilizer demand index E actual This relates to the aboveground canopy's transpiration demand and the underground soil's ability to supply fertilizer and water.

[0240] Construct a comprehensive evaluation value E integrated The calculation formula is:

[0241] (E integrated =α·E canopy +b·F soil +c·E actual -δ·R d Where: a, b, and c are the weighting coefficients for aboveground growth, soil fertility, and water and fertilizer requirements, calibrated through model iteration units;

[0242] δ is the negative correction coefficient for disaster risk; the higher the risk, the more the comprehensive assessment value is reduced.

[0243] The fusion process generates comprehensive evaluation information that includes numerical scores, spatial distribution, and decision recommendations.

[0244]

[0245] The information is sent to the user terminal in the form of a 3D growth trend map, abnormal area annotation, and quantitative scoring table through the information sending module.

[0246] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A crop planting management system that combines above-ground and underground data, characterized in that, It includes a planting area data collection module, a crop data collection module, a surface data collection module, an underground data collection module, a data analysis module, a data processing module, and an information transmission module; The planting area data collection module is configured to collect the area information and geographic boundary coordinates of the planting area; The crop collection module is configured to collect crop type information and growth stage identifiers; The data analysis module generates above-ground data acquisition instructions and underground data acquisition instructions based on area information, geographic boundary coordinates, and type information. The above-ground data acquisition module collects above-ground data of the planting area in response to the above-ground data acquisition command; The underground data acquisition module collects underground data from the planting area in response to underground data acquisition commands; The data processing module processes above-ground data to generate above-ground assessment information, processes underground data to generate underground assessment information, analyzes underground data to obtain water and fertilizer demand index, and integrates above-ground and underground assessment information to generate comprehensive assessment information. The specific process for generating above-ground assessment information is as follows: Ground-level data is acquired through the ground-level data acquisition module, including: Canopy morphology data, including plant height and leaf area index, are collected by image recognition or sensors; Canopy physiological data, including leaf temperature and chlorophyll fluorescence, were acquired using infrared / spectroscopy equipment; Meteorological data, including light intensity, air temperature and humidity, are collected by environmental sensors; Combining the pre-set spatial heterogeneity analysis unit and adaptive sampling unit, the collected ground data are processed as follows: Spatial calibration involves matching planting area raster partitions with geographic boundary coordinates. Temporal alignment, based on a spatiotemporal correlation predictor, unifies data collection timestamps; Anomaly filtering utilizes a real-time verification unit to remove noisy data that deviates from the normal range; In the construction of the above-ground assessment model: LSTM time series model and multimodal disaster prediction submodule, a two-dimensional assessment model is built. Growth status assessment involves feature extraction from canopy morphology and physiological data, including leaf area index fitting and chlorophyll content inversion. A crop demand library model trained through the data processing module then outputs a growth status score. ; Disaster risk assessment combines meteorological data with heat wave / freezing risk indices to calculate the current disaster threat value faced by the canopy and generate risk level labels, including safe, warning, or emergency. By integrating growth status scores and disaster risk levels, structured above-ground assessment information is generated. ； The specific process of processing underground data to generate underground assessment information is as follows: Subsurface data is acquired through a subsurface data acquisition module and includes: Soil physical data, including moisture and bulk density, are collected by soil sensors; Soil chemical data, including nutrient content and electrical conductivity, are collected by soil rapid testing equipment; Root distribution data, including root length density and depth, is obtained by ground-penetrating radar or image reconstruction technology; Based on the spatial heterogeneity analysis unit, the underground data is processed as follows: Rasterized mapping, aligned with planting area and geographic boundary coordinates, generates a heat map of the spatial distribution of soil parameters; Deep hierarchical processing is used to split the data according to the main distribution layer of the root system; Dynamic sampling optimization utilizes adaptive sampling units to supplement sampling in areas with high coefficients of variation, including nutrient gradient zones; Construction of underground assessment model: By combining the data processing module with the spatiotemporal correlation predictor, a soil-root synergistic assessment model is constructed: Soil fertility assessment involves normalizing soil nutrient and moisture data, matching optimal parameters for the current growth stage (including seedling nitrogen requirement thresholds) using a crop requirement database, and outputting a soil fertility index. ; Root health assessment involves morphological analysis of root distribution data, including root surface area and branch number, combined with multimodal data, including the correlation between canopy water content and root water uptake, to calculate the root vigor index. This reflects the root system's absorption efficiency; The underground assessment information output integrates soil fertility index and root vitality index to generate structured underground assessment information: ； The calculation of the above-ground assessment information, below-ground assessment information, and water and fertilizer demand index is all based on the growth stage identifier, and is achieved by matching the physiological parameter thresholds of the corresponding growth stage in the crop demand database. The specific process of fusing above-ground and underground assessment information to generate comprehensive assessment information is as follows: Extracting common dimensions from above-ground and below-ground data: Growth stage correlation: By identifying crop growth stages, a unified benchmark for above-ground and below-ground growth stages is established; Based on the geographical boundary coordinates and grid partitions of the planting area, align the above-ground anomaly areas with the below-ground soil parameter anomaly areas. Water and fertilizer demand correlation: through the defined water and fertilizer demand index This relates to the aboveground canopy's transpiration demand and the underground soil's nutrient / water supply capacity; Construct a comprehensive evaluation value The calculation formula is: Where: a, b and c are the weight coefficients of aboveground growth, soil fertility and water and fertilizer requirements, which are calibrated through model iteration units; This is a negative correction coefficient for disaster risk; the higher the risk, the greater the reduction in the overall assessment value. The fusion process generates comprehensive evaluation information that includes numerical scores, spatial distribution, and decision recommendations. ； The information sending module will send comprehensive evaluation information to the user terminal; The crop planting management system also includes a water and fertilizer early warning module, which is used to control irrigation / fertilization equipment based on the water and fertilizer demand index. When the water and fertilizer demand index exceeds the dynamic threshold, the control signal of the irrigation / fertilization equipment is triggered. The calculation model of the water and fertilizer demand index needs to input the growth stage identifier in order to call the water and fertilizer demand benchmark value of the corresponding stage. The process of obtaining the water and fertilizer demand index is as follows: ； In the formula, n represents the number of parameter combinations or influencing factors involved in the calculation, which is a positive integer. The specific value depends on the classification of different influencing factors considered when actually building the model. It is the weight coefficient corresponding to the i-th influence factor; It is a function of soil moisture, nutrient content, and growth stage; soil moisture is an indicator that reflects the water content in the soil. Common ways of expressing it include volumetric water content and gravimetric water content. It is one of the basic environmental factors that affect the water and fertilizer requirements of crops. Nutrient refers to the nutrient content in the soil, including the content of major nutrients such as nitrogen, phosphorus, and potassium. Growth represents the growth stages of crops, including the seedling stage, flowering stage, and maturity stage. The water and fertilizer requirements of crops vary greatly at different growth stages.

2. The crop planting management system combining above-ground and underground data according to claim 1, characterized in that: The data processing module includes: The real-time verification unit is used to receive user feedback data and calculate dynamic verification factors for comprehensive evaluation information. Specifically: ； in, The actual growth status is scored manually entered by the user, on a scale of 0-100. The system generates a comprehensive evaluation score, on a scale of 0-100. The maximum value function is the maximum permissible absolute deviation between user ratings and system evaluation values, which is the boundary value for judging the reliability of the data source. θ is the credibility threshold, which is used to comprehensively evaluate the lower limit of the consistency between information and user feedback. Model iteration unit, used in ,and The crop demand database parameter calibration is triggered in a timely manner to iterate the model. This is the data credibility threshold; The real-time verification unit is connected to the UAV re-inspection system: when ,and At that time, the ground-based data acquisition module will re-acquire data in the abnormal area.

3. A crop planting management system combining above-ground and underground data according to claim 2, characterized in that: The data processing module also integrates a multimodal disaster prediction submodule, which is used to construct a heat wave / freezing damage risk index based on historical meteorological data and real-time ground data. The specific process is as follows: ； The infrared thermometer value is for the canopy. The optimal temperature for the current growth stage in the crop demand pool; This refers to the water content of the leaves. The rate of change in leaf water content; The temperature deviation weighting coefficient is used, with heat wave disaster > 0 and frost damage disaster < 0. This represents the sensitivity coefficient to changes in moisture content.

4. A crop planting management system combining above-ground and underground data according to claim 3, characterized in that: The crop planting management system also includes a disaster response unit, a protection strength calculation module, and an equipment driver module; The disaster response unit is used to connect the shade net and the anti-frost fan equipment; The protection strength calculation module is used to calculate the risk index of heat waves / freezing damage. Calculate the protection strength P: ； In the formula, This represents the crop's tolerance safety threshold. This represents the critical threshold for irreversible crop damage. This refers to the maximum power factor of the equipment. The device driver module is used to convert the protection strength P into a device control signal.

5. A crop planting management system combining above-ground and underground data according to claim 1, characterized in that: The crop planting management system also includes a water and fertilizer early warning module, which is used to control irrigation / fertilization equipment based on the water and fertilizer demand index. When the water and fertilizer demand index exceeds the dynamic threshold, the irrigation / fertilization equipment control signal is triggered.

6. A crop planting management system combining above-ground and underground data according to claim 4, characterized in that: The disaster response unit implements a tiered response strategy, the specific content of which is as follows: Basic protection mode: When a heat wave occurs Frost damage may occur. When activated, the core of the system is a low-power operation of the core protection equipment, which only maintains basic protection strength. The core protection equipment includes a shade net and a frost-proof fan. Emergency Response Mode: When a heat wave occurs Frost damage may occur. The system is activated at any time, with the core being full-load operation of all equipment plus the activation of the backup protection system, which provides disaster backup protection. The backup protection system includes a backup heater.

7. A crop planting management system combining above-ground and underground data according to claim 2, characterized in that: The data analysis module also includes a spatial heterogeneity analysis unit, a dynamic partitioning engine, and an adaptive sampling unit. Spatial heterogeneity analysis unit, used to calculate the coefficient of variation of raster regions based on historical underground data; The dynamic partitioning engine is used to split the raster into sub-partitions when the coefficient of variation meets preset conditions, and assign an independent identifier code to each sub-partition. An adaptive sampling unit is used to adjust the sampling density of the underground data acquisition module based on the coefficient of variation.

8. A crop planting management system combining above-ground and underground data according to claim 1, characterized in that: The data analysis module also includes: A spatiotemporal correlation predictor is used to establish a correlation matrix between above-ground and underground data in a raster area; A strategy generator is used to predict the optimal acquisition strategy based on the correlation matrix and historical above-ground and below-ground data using an LSTM time series model.

9. A crop planting management system combining above-ground and underground data according to claim 1, characterized in that: The information sending module includes an AR visualization interface, which is used for: A three-dimensional growth pattern is generated by overlaying underground root distribution data with above-ground canopy data. Spatial heat map marking areas with abnormal water and fertilizer demand indices.

Images