A farm-scale wheat and corn crop water and fertilizer integrated operation decision method and device

Abstract

The present application relates to a kind of farm scale wheat and corn crop water and fertilizer integrated operation decision method and device, belong to intelligent agricultural crop production management field.It includes: the automatic acquisition of planting area thermal infrared image, multispectral image, obtains ground crop vegetation nitrogen content, plant number, vegetation dry weight sample data, calculates the above-ground sample biomass;Nitrogen content, biomass dynamic sample data of different growth stages of crops are combined, vegetation nitrogen-biomass characteristic space is constructed, and nitrogen dilution curve is obtained by fitting;Establish vegetation nitrogen, biomass intelligent remote sensing inversion model, and invert the whole area vegetation nitrogen content and biomass spatial distribution atlas;Temperature drought vegetation index is calculated, combined with nitrogen-biomass atlas, nitrogen nutrition index is analyzed pixel by pixel;Temperature drought index, nitrogen nutrition index are established respectively, and temperature drought index-nitrogen nutrition index characteristic space is formed, and water and fertilizer deficiency diagnosis is carried out pixel by pixel by gridding characteristic space.

Description

Technical Field

[0001] This invention belongs to the field of intelligent agricultural crop production management, and in particular relates to a farm-scale decision-making method and device for integrated water and fertilizer operation of wheat and corn crops. Background Technology

[0002] Intelligent diagnosis and precise operational decision-making regarding water and fertilizer deficits are core components of integrated water and fertilizer management for wheat and corn. Their accuracy directly determines the efficiency of water and fertilizer resource utilization, crop yield and quality, and the synergy of field production. Currently, traditional water and fertilizer management relies on experience-based judgment, making it difficult to accurately identify the type and degree of water and fertilizer deficits in crops, leading to imbalances in water and fertilizer application and serious resource waste. To address this challenge, intelligent technical solutions based on remote sensing, the Internet of Things (IoT), and machine learning have gradually emerged in the industry. However, in actual crop growth, water and fertilizer operations are usually carried out simultaneously, and existing technologies operate independently in terms of water and fertilizer stress diagnosis and operational decision-making, failing to comprehensively consider the crop's needs for water and nutrients. For example, invention patent CN120167207A discloses an intelligent water and fertilizer application system for wheat. This system collects data such as soil moisture, plant drought status, and NDVI through IoT and remote sensing equipment, and constructs a predictive model based on deep learning algorithms to predict the timing and amount of water and fertilizer application. While this solution achieves data-driven intelligent decision-making, it only indirectly infers water and fertilizer status through a single indicator and lacks a water and fertilizer interaction analysis mechanism, making it unable to accurately distinguish the synergistic effects of "water stress" and "nitrogen deficit." Furthermore, its block-based management model fails to consider the heterogeneity of water and fertilizer within the same plot, limiting decision-making granularity to the block level and failing to support pixel-level precision operations, thus falling short of the needs for refined management at the farm scale. For example, invention patent CN118067709A proposes a method for simultaneous monitoring of crop water and fertilizer balance. This method uses multispectral images from UAVs to invert crop nitrogen and water content, and calculates a comprehensive water and fertilizer balance index based on the entropy weight method, achieving plot-level monitoring of water and fertilizer status. This scheme overcomes the limitations of single-factor monitoring, but its core flaw lies in the ambiguity of the diagnostic dimensions: the comprehensive index can only reflect the overall profit and loss level, and cannot quantify the respective deficit levels and interactive effects of water and nitrogen. Moreover, the weighting coefficients are difficult to obtain accurately, making it difficult to accurately output differentiated decisions such as "water supplementation as the main focus and nitrogen supplementation as a supplement" or "water and fertilizer synergistic supplementation". Furthermore, the monitoring results are only used as a reference and have not been transformed into operational instructions that can directly drive irrigation and fertilization equipment. The diagnosis and execution links are disconnected, limiting its practicality.

[0003] In summary, existing water and fertilizer deficit diagnosis and decision-making technologies generally suffer from bottlenecks: First, the diagnostic logic is one-sided, either relying on monitoring only a single factor or making vague judgments based solely on comprehensive indices, lacking analysis of water and fertilizer interactions, and failing to accurately distinguish the dominant type and level of deficit; second, the decision-making granularity is coarse, mostly remaining at the block or plot level, ignoring the heterogeneity of water and fertilizer in the field, and failing to achieve pixel-level precise control; third, decision-making is disconnected from operations, with monitoring results unable to be directly converted into executable operational instructions for equipment, requiring secondary manual processing, which reduces decision-making efficiency and implementation. Summary of the Invention

[0004] This paper proposes a technical solution for integrated water and fertilizer management in the agricultural production process of wheat and corn crops on a farm, aiming to improve the planting and management efficiency of wheat and corn crops. The specific technical solution is as follows:

[0005] A farm-scale decision-making method for integrated water and fertilizer management in wheat and corn crops, comprising:

[0006] The data obtained includes UAV thermal infrared image data and multispectral image data in the image coordinate system of the crop planting area of ​​the farm. The two types of image data are pixel registered, and ground sampling survey data obtained from ground sampling and testing are also included, including vegetation nitrogen content data, vegetation dry weight data, and plant number data. Biomass sample data are calculated based on vegetation dry weight and plant number.

[0007] Based on vegetation nitrogen content data from ground sampling surveys at different times and calculated biomass data, a vegetation nitrogen-biomass feature space was constructed. The functional equation between the two, i.e., the nitrogen dilution curve, was obtained by fitting the feature space.

[0008] A farm-scale intelligent remote sensing inversion model for vegetation nitrogen and biomass was established using ground survey sampling data and UAV multispectral image data to invert the spatial distribution map of vegetation nitrogen content and biomass in the planting area.

[0009] Based on the collected UAV thermal infrared and multispectral image data, the surface temperature and NDVI index are calculated, and the temperature-vegetation drought index is further generated. At the same time, the nitrogen nutrition index is analyzed pixel by pixel by combining the vegetation nitrogen content and biomass map.

[0010] Based on the temperature and vegetation drought index and the nitrogen nutrition index, and after standardization, the index is classified into the [0,1] interval. A feature space is established with the temperature and vegetation drought index as the abscissa and the nitrogen nutrition index as the ordinate, and the feature space is divided into grids based on the degree of water and fertilizer stress.

[0011] Based on the integrated pixel-by-pixel determination of crop water and fertilizer stress by grid partitioning, water and fertilizer operation instructions are dynamically output according to the water and fertilizer stress level to guide irrigation and fertilization equipment to perform precise operations.

[0012] The present invention has the following beneficial effects:

[0013] This invention comprehensively quantifies the degree of crop water and nutrient deficiency and their interaction, accurately outputs differentiated operational decisions such as "water supplementation as the main method and nitrogen supplementation as the auxiliary method" or "water and fertilizer synergistic supplementation", and transforms the results into operational instructions that can directly drive irrigation and fertilization equipment, thus connecting the operational decision-making and execution links and improving the practicality of water and fertilizer operation decisions for wheat and corn crops on farms. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating a farm-scale decision-making method for integrated water and fertilizer management of wheat and corn crops according to an embodiment of the present invention.

[0015] Figure 2 This is a schematic diagram of the temperature drought vegetation index-nitrogen nutrition index feature space gridding of a farm-scale fertigation operation decision-making method for wheat and corn crops according to an embodiment of the present invention.

[0016] Figure 3 A schematic diagram of a farm-scale fertigation decision-making device for wheat and corn crops according to an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above objectives, this invention adopts the following technical solution.

[0018] To achieve the above objectives, the present invention adopts the following technical solution:

[0019] Figure 1 This is a schematic diagram of a farm-scale fertigation decision-making method for wheat and corn crops, as described in an embodiment of the present invention. Figure 1 The farm-scale fertigation operation decision-making method for wheat and corn crops shown in this embodiment may include:

[0020] S101: Acquire drone imagery data of the crop planting area of ​​the farm, as well as ground sampling survey data obtained from ground sampling and testing, and calculate biomass sample data.

[0021] The drone imagery data includes multispectral imagery data and thermal infrared imagery data.

[0022] UAV imagery data can be obtained by using a multi-rotor UAV equipped with multispectral and thermal infrared cameras (such as the M350RTK UAV + AQ300 Pro camera).

[0023] Ground sampling survey data includes vegetation nitrogen content, number of plants, and vegetation dry weight. Vegetation nitrogen content can be obtained by collecting plant samples from the farm's planting area, grinding them, and then analyzing them in a laboratory. Vegetation dry weight is obtained by blanching the aboveground plant samples at 105℃ for 30 minutes and then drying them at 65-70℃ to constant weight.

[0024] Biomass is calculated using the following formula:

[0025] ;

[0026] Where AGB represents the biomass of the sampled cell. This refers to the total dry weight of the aboveground parts of a single crop plant. This represents the total number of crop plants within the sampling plot.

[0027] S102, combining dynamic sample data of nitrogen content and biomass at different growth stages, constructs a vegetation nitrogen-biomass feature space with biomass as the abscissa and vegetation nitrogen content as the ordinate, and obtains the nitrogen dilution curve equation by fitting.

[0028] In a specific embodiment, the nitrogen dilution curve is determined by fitting using machine learning methods, and it takes the form of a power function:

[0029] ;

[0030] in, Let W be the vegetation nitrogen content, W be the vegetation biomass, and a and b be the nitrogen dilution curve parameters (a > 0, 0 < b < 1). These parameters were obtained through nonlinear least squares fitting, with the fitting objective being to minimize... ². This represents the fitting error term.

[0031] In other embodiments, nitrogen dilution curve equations can also be obtained according to other preset rules.

[0032] S103. Based on UAV multispectral image data and ground sampling survey data, a remote sensing inversion model for vegetation nitrogen and biomass was established, and a spatial distribution map of vegetation nitrogen and biomass in the planting area was obtained based on the inversion model.

[0033] In a specific embodiment, the vegetation nitrogen content inversion model and the biomass inversion model are determined based on machine learning methods:

[0034] ;

[0035] in, Let represent the vegetation nitrogen content and biomass at the i-th sampling point. For the model intercept term, The fitting coefficients for the spectral indices at each sampling point are... Let q be the spectral index of the i-th sampling point.

[0036] In other embodiments, vegetation nitrogen content and biomass models can also be obtained according to other preset rules.

[0037] S104 calculates the temperature drought vegetation index based on UAV imagery data, and combines vegetation nitrogen content map and biomass map to analyze the nitrogen nutrition index pixel by pixel based on nitrogen dilution curve.

[0038] In a specific embodiment, the nitrogen nutritional index is calculated using the following formula.

[0039] ;

[0040] in, It is the calculated nitrogen nutrient index. It refers to the pixel nitrogen concentration in the crop vegetation nitrogen map. The critical nitrogen concentration for crops is calculated based on biomass maps and nitrogen dilution curves. When = 1, it indicates that the crop nitrogen content is appropriate at this stage and no additional fertilizer is needed. A value less than 1 indicates a nitrogen deficiency in the crop, requiring fertilizer supplementation. A value >1 indicates that the current crop has excessive nitrogen and no additional fertilizer is needed.

[0041] The temperature drought vegetation index is calculated using the following formula:

[0042] ;

[0043] ;

[0044] ;

[0045] ;

[0046] in, It is the surface temperature value. It is the normalized vegetation index. Based on The calculated vegetation fractal cover was used as a step size of 0.01, and a temperature-fractal cover feature space was established with vegetation fractal cover as the x-axis and land surface temperature as the y-axis. and In the temperature-fractal coverage feature space, the maximum and minimum temperatures corresponding to different fractal coverage values ​​are used to establish the dry edge equation and the wet edge equation. and These are the fitting parameters for the dry-edge equation. and These are the fitting parameters for the wet edge equation. It is the calculated temperature drought vegetation index.

[0047] S105, based on the temperature-drought vegetation index and the nitrogen nutrition index, with the temperature-drought vegetation index as the x-axis and the nitrogen nutrition index as the y-axis, when... When >1, let = 1, establish a characteristic space of temperature drought vegetation index-nitrogen nutrition index, with 0.2 The feature space is divided into grids with a step size of 0.2.

[0048] S106 determines the crop's water and fertilizer deficit status pixel by pixel and outputs water and fertilizer operation instructions according to the stress level to guide irrigation and fertilization equipment to operate precisely.

[0049] Figure 2 This is a schematic diagram of the temperature drought vegetation index-nitrogen nutrition index feature space of a farm-scale fertigation decision-making method for wheat and maize crops according to an embodiment of the present invention.

[0050] Figure 3 This is a schematic diagram of a device structure for farm-scale integrated water and fertilizer management decision-making for wheat and corn crops, according to an embodiment of the present invention. This embodiment is related to... Figure 1 Corresponding to a farm-scale decision-making method for integrated water and fertilizer management in wheat and corn crops, such as Figure 3 As shown, the apparatus for a farm-scale fertigation operation decision-making method for wheat and corn crops in this embodiment may include:

[0051] The data acquisition and storage module 201 is used to store the acquired UAV data, as well as vegetation nitrogen content data and biomass data obtained from ground surveys and ground sampling tests.

[0052] The nitrogen dilution curve construction module 202 is used to combine dynamic sample data of nitrogen content and biomass at different growth stages, construct a vegetation nitrogen-biomass feature space with biomass as the abscissa and vegetation nitrogen content as the ordinate, and fit and obtain the nitrogen dilution curve.

[0053] The agricultural condition inversion model and map inversion module 203 is used to establish a remote sensing inversion model of vegetation nitrogen and biomass based on UAV multispectral image data and ground sampling survey data, and to obtain a spatial distribution map of vegetation nitrogen and biomass in the planting area based on the inversion model.

[0054] Index calculation module 204 is used to calculate the temperature drought vegetation index and the nitrogen nutrition index.

[0055] The water and fertilizer stress characteristic spatial construction module 205 is used to combine the temperature drought vegetation index and the nitrogen nutrition index, with a value of 0.2. A grid for water and fertilizer stress levels is established with a step size of 0.2.

[0056] The water and fertilizer integration decision and operation instruction output module 206 is used to output water and fertilizer operation instructions according to the degree of water and fertilizer stress, so as to guide the precise operation of irrigation and fertilization equipment.

[0057] It should be noted here that, Figure 3 The data acquisition and storage module 201, nitrogen dilution curve construction module 202, crop condition inversion model and map inversion module 203, index calculation module 204, water and fertilizer stress feature space construction module 205, and water and fertilizer integrated decision and operation instruction output module 206 of a farm-scale wheat and corn crop fertigation operation decision-making device are included. Figure 1 The specific implementation process in steps S101 to S106 of the farm-scale decision-making method for integrated water and fertilizer operation of wheat and corn crops is the same, and will not be repeated here.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A farm-scale decision-making method for integrated water and fertilizer management in wheat and corn crops, characterized in that, include: UAV thermal infrared image data and multispectral image data in the image coordinate system of the crop planting area of ​​the farm, as well as ground sampling survey data obtained from ground sampling and testing, including vegetation nitrogen content data, vegetation dry weight data, and plant number data, and biomass sample data are calculated based on vegetation dry weight and plant number in the sample area. Based on vegetation nitrogen content data from ground sampling surveys at different times and calculated biomass data, a vegetation nitrogen-biomass feature space was constructed. The functional equation between the two, i.e., the nitrogen dilution curve, was obtained by fitting the feature space. A farm-scale intelligent remote sensing inversion model for vegetation nitrogen and biomass was established using ground survey sampling data and UAV multispectral image data to invert the spatial distribution map of vegetation nitrogen content and biomass in the planting area. Based on the collected UAV thermal infrared and multispectral image data, the surface temperature and NDVI index are calculated, and the temperature-vegetation drought index is further generated. At the same time, the nitrogen nutrition index is analyzed pixel by pixel by combining the vegetation nitrogen content and biomass map. Based on the temperature and vegetation drought index and the nitrogen nutrition index, and after standardization, they were classified into the [0,1] interval. A feature space was established with the temperature and vegetation drought index as the abscissa and the nitrogen nutrition index as the ordinate. The feature space was then divided into gridded partitions based on the degree of water and fertilizer stress. Based on the integrated pixel-by-pixel determination of crop water and fertilizer stress by grid partitioning, water and fertilizer operation instructions are dynamically output according to the water and fertilizer stress level to guide irrigation and fertilization equipment to perform precise operations.

2. The farm-scale fertigation operation decision-making method for wheat and corn crops as described in claim 1, characterized in that, Before calculating the temperature drought vegetation index, the following also applies: The fractal vegetation cover is calculated using the following formula: ； ； ； ； in, It is the surface temperature value. It is the normalized vegetation index. Based on The calculated vegetation fractal cover was used as a step size of 0.01, and a temperature-fractal cover feature space was established with vegetation fractal cover as the x-axis and land surface temperature as the y-axis. and In the temperature-fractal coverage feature space, the maximum and minimum temperatures corresponding to different fractal coverage values ​​are used to establish the dry edge equation and the wet edge equation. and These are the fitting parameters for the dry-edge equation. and These are the fitting parameters for the wet edge equation. It is the calculated temperature drought vegetation index.

3. The farm-scale fertigation operation decision-making method for wheat and corn crops as described in claim 1, characterized in that, When making decisions about integrated water and fertilizer management, the following should also be included: A characteristic space of temperature drought vegetation index-nitrogen nutrient index was established, with 0.2... The feature space is divided into gridded partitions based on the degree of water and fertilizer stress, with a step size of 0.

2. The degree of water and fertilizer stress is divided into five categories: suitable, mild stress, moderate stress, severe stress, and extreme stress. Corresponding operational decision management is carried out according to the state of water and fertilizer stress, so as to realize differentiated and integrated decision-making of "water supplementation as the main method and nitrogen supplementation as the auxiliary method" or "water and fertilizer synergistic supply".

4. The farm-scale fertigation operation decision-making method for wheat and corn crops as described in claim 1, characterized in that, Biomass is calculated using the following formula: ； Where AGB represents the biomass of the sampled cell. This refers to the total dry weight of the aboveground parts of a single crop plant. This represents the total number of crop plants within the sampling plot.

5. The farm-scale fertigation operation decision-making method for wheat and corn crops as described in claim 1, characterized in that, The nitrogen dilution curve was determined by fitting using machine learning methods, and it is in the form of a power function. ； in, Let W be the vegetation nitrogen content, W be the vegetation biomass, and a and b be the nitrogen dilution curve parameters. These parameters were obtained through nonlinear least squares fitting, with the fitting objective being to minimize [the nitrogen content]. ², This represents the fitting error term.

6. The farm-scale fertigation operation decision-making method for wheat and corn crops as described in claim 1, characterized in that, The vegetation nitrogen content inversion model and biomass inversion model were determined based on machine learning methods: in, Let represent the vegetation nitrogen content and biomass at the i-th sampling point. For the model intercept term, The fitting coefficients for the spectral indices at each sampling point are... Let q be the spectral index of the i-th sampling point.

7. The farm-scale fertigation operation decision-making method for wheat and corn crops as described in claim 1, characterized in that, The nitrogen nutritional index is calculated using the following formula: ； in, It is the calculated nitrogen nutrient index. It refers to the pixel nitrogen concentration in the crop vegetation nitrogen map. It is the critical nitrogen concentration for crops calculated based on biomass maps and nitrogen dilution curves; when When = 1, it indicates that the crop nitrogen content is appropriate at this stage and no additional fertilizer is needed. When < 1, it indicates that the crop is currently deficient in nitrogen and needs additional fertilizer; when A value >1 indicates that the current crop has excessive nitrogen and no additional fertilizer is needed.

8. A farm-scale decision-making device for integrated water and fertilizer management of wheat and corn crops, characterized in that, include: The data acquisition and storage module is used to store the acquired UAV data, as well as vegetation nitrogen content data and biomass data obtained from ground surveys and ground sampling tests. The nitrogen dilution curve construction module is used to combine dynamic sample data of nitrogen content and biomass at different growth stages, construct a vegetation nitrogen-biomass feature space with biomass as the abscissa and vegetation nitrogen content as the ordinate, and fit and obtain the nitrogen dilution curve. The agricultural condition inversion model and map inversion module are used to establish a remote sensing inversion model of vegetation nitrogen and biomass based on UAV multispectral image data and ground sampling survey data, and obtain a spatial distribution map of vegetation nitrogen and biomass in the planting area based on the inversion model. The index calculation module is used to calculate the temperature drought vegetation index and the nitrogen nutrition index. A spatial construction module for water and fertilizer stress characteristics is used to combine temperature drought vegetation index and nitrogen nutrition index to establish a grid for water and fertilizer stress intensity. The integrated water and fertilizer decision-making and operation instruction output module is used to output water and fertilizer operation instructions according to the degree of water and fertilizer stress, so as to guide the precise operation of irrigation and fertilization equipment.

9. An electronic device, characterized in that, include: One or more processors; A memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores executable instructions that, when executed by a processor, cause the processor to perform the method described in any one of claims 1 to 7.

Images