A precise fertilization decision-making and plot-level display method and system

Abstract

The present application relates to a precise fertilization decision-making and plot-level display method and system, including the steps of reading in map information; marking the plots of planted crops on the map according to the field ridges; calculating the input-output ratio values for each plot; and marking the input-output ratio values for each plot or marking each plot with a specific color according to the input-output ratio values. By directly displaying the plot shapes and input-output ratios on the map, relevant personnel can intuitively see the balanced fertilization status of the plots within a certain area.

Description

Technical Field

[0001] This application belongs to the technical field of green agricultural big data, and in particular relates to a precise fertilization decision-making and plot-level display method and system. Background Art

[0002] Variable fertilization technology is one of the core contents of precision agriculture. It uses information technology and, according to the spatial and temporal variations of crops and their growth environments, implements fertilization prescription farming with positioning, timing, and quantification. Its purpose is to reduce fertilizer input to achieve the same yield, improve the farmland ecological environment, increase fertilizer utilization rate, and obtain the best environmental and economic benefits.

[0003] In order to better display the economy of fertilization in a certain area, there is a need for a method and system for precise fertilization decision-making and plot-level display, which can use the system to display the economic situation of fertilization amounts in different plots on a map. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: to provide a method and system for plot-level display based on precise fertilization decision-making to solve the deficiencies in the prior art.

[0005] The technical solution adopted by the present invention to solve its technical problems is:

[0006] A precise fertilization decision-making and plot-level display method includes the following steps:

[0007] S1: Read in map information;

[0008] S2: Mark the plots where crops are planted on the map according to the field ridges;

[0009] S3: Calculate the production-input ratio numerical curves of nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer on each plot, and solve the minimum value of the production-input ratio numerical curve as the recommended fertilization amount;

[0010] S4: Take the recommended fertilization amounts of nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer minus the original fertilizer contents of nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer in the plot as the fertilization recommendation plan, and mark this fertilization recommendation plan on the plots of the map information;

[0011] S5: Click on the corresponding plot on the map information to view the fertilization recommendation plans of nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer for this plot;

[0012] The calculation method of the production-input ratio value of each fertilizer in step S3 is:

[0013] S31: Obtain the initial soil fertilizer content Z0 of the area where the crop to be planted is located before cultivation, obtain the current fertilizer price P, obtain the crop yield Yk of the previous year and the crop yield Yp of the current year, obtain the remaining soil fertilizer content Z after the crop is harvested in the current year, obtain the fertilizer absorption content Sk per unit crop of the previous year and the fertilizer absorption content Sp per unit crop of the fertilizer applied in the current year, and the amount of fertilizer used for this cultivation is H;

[0014] S32: Calculate the total crop absorption Xp of the fertilizer applied in the current year. The total crop absorption Xp of the fertilizer applied in the current year = the crop yield Yp of the current year × the fertilizer absorption content Sp per unit crop of the fertilizer applied in the current year;

[0015] Calculate the total crop absorption Xk of the previous year. The total crop absorption Xk of the previous year = the crop yield Yk of the previous year × the fertilizer absorption content Sk per unit crop of the previous year;

[0016] Calculate the fertilizer loss. The fertilizer loss S in the current year = (1 - I%) × the total crop absorption Xp of the fertilizer applied in the current year / the area of the region; The fertilizer utilization rate I% in the current year = (the total crop absorption Xp of the fertilizer applied in the current year - the remaining soil fertilizer content Z after the crop is harvested in the current year) / the amount of fertilizer used for this cultivation H × 100%;

[0017] S33: Calculate the input-output ratio curve D, ; where P represents the price of the fertilizer, Py represents the price of the crop product, and solve for the minimum value of the input-output ratio numerical curve as the recommended fertilizer application rate.

[0018] Preferably, for the precise fertilization decision-making and plot-level display method of the present application, in step S4, the plots are divided into high, medium, and low fertilization degrees according to the ratio of the fertilizer application rate to the original fertilizer content of the plot, and classified according to the three fertilization degrees of nitrogen fertilizer, phosphate fertilizer, and potassium fertilizer, and the classification is marked on the map. In step S5, click on the corresponding plot information on the map to view the classification mark of the plot.

[0019] Preferably, for the precise fertilization decision-making and plot-level display method of the present application, different classification marks are represented by different colors on the map.

[0020] Preferably, for the precise fertilization decision-making and plot-level display method of the present application, the method for obtaining the fertilizer absorption content Sp per unit crop of the fertilizer applied in the current year and the fertilizer absorption content Sm per unit crop without fertilization is as follows:

[0021] Air-dry the plants of the harvested crops, use statistical means, randomly select some crops as samples, measure the content of fertilizer elements in the crops, and then estimate the fertilizer absorption content Sp per unit crop of the fertilizer applied in the current year and the fertilizer absorption content Sm per unit crop without fertilization according to the sample ratio;

[0022] The air-dried plants include crop seeds, stems and leaves, and roots. After crushing the air-dried plants, the determination of fertilizer element content is carried out.

[0023] Use a convolutional neural network model to measure the fertilizer element content of the air-dried stems and leaves and the fertilizer element content of the grains.

[0024] The method includes the following steps:

[0025] A1: Training process of the convolutional neural network model device: Use the photos of the air-dried stems and leaves and grains with specific moisture content and the corresponding fertilizer element content of the stems and leaves and the fertilizer element content of the grains as the training set to train the convolutional neural network model device to obtain the convolutional neural network model device; when constructing the training set, divide several intervals according to the normal distribution based on the mass of the stems and leaves or grains, select the corresponding number of samples to construct the training set, and the stems and leaves or grains as the training set are complete photos, and the parts in the photos that are not stems and leaves or grains are adjusted to be transparent through processing.

[0026] A2: Element content recognition process. Use the photos of the air-dried stems and leaves and grains with new unknown fertilizer element content of the stems and leaves and the fertilizer element content of the grains as the input, and use the output of the convolutional neural network model as the fertilizer element content of the stems and leaves and the fertilizer element content of the grains.

[0027] During the recognition process, also select the stems and leaves and grains whose mass of the stems and leaves or grains conforms to the normal distribution as samples. After summing the fertilizer element content in each recognized stem and leaf or grain and then dividing by the corresponding weight percentage, the fertilizer element content of the experimental plot is obtained.

[0028] The present invention also provides a precise fertilization decision-making and plot-level display system, including:

[0029] Map module. In the map module, there are plots with the planting crop ranges marked according to the field ridges.

[0030] Recommended fertilization amount calculation module, which is used to calculate the production-input ratio numerical curves of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer on each plot, and solve the minimum value of the production-input ratio numerical curve as the recommended fertilization amount.

[0031] Map marking module, which is used to take the recommended fertilization amounts of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer minus the original fertilizer contents of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer in the plot as the fertilization recommendation plan, and mark this fertilization recommendation plan on the plots of the map information.

[0032] Query module: It is used to display the fertilization recommendation plans of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer for the corresponding plot when clicking on the corresponding plot in the map information.

[0033] Preferably, in the precise fertilization decision-making and plot-level display system of the present invention, in the map marking module, plots are divided into high, medium, and low fertilization degrees according to the ratio of the fertilization amount to the original fertilizer content of the plot, and classified according to the three fertilization degrees of nitrogen fertilizer, phosphate fertilizer, and potassium fertilizer, and the classification is marked on the map. In the query module, click on the corresponding plot in the map information to view the classification mark of the plot.

[0034] Preferably, the precise fertilization decision-making and plot-level display system of the present invention

[0035] The recommended fertilization amount calculation module includes:

[0036] The data acquisition sub-module: used to obtain the initial soil fertilizer content Z0 of the area where the crop to be planted is located before cultivation, obtain the current fertilizer price P, obtain the crop yield Yk of the previous year and the crop yield Yp of the current year, obtain the remaining soil fertilizer content Z after the crop is harvested in the current year, obtain the unit crop fertilizer absorption content Sk of the previous year and the unit crop fertilizer absorption content Sp of the current year for fertilization, and the amount of fertilizer used for this cultivation is H;

[0037] The data processing sub-module: used to calculate the total crop absorption Xp of fertilization in the current year, the total crop absorption Xp of fertilization in the current year = the crop yield Yp of the current year × the unit crop fertilizer absorption content Sp of fertilization in the current year; calculate the total crop absorption Xk of the previous year, the total crop absorption Xk of the previous year = the crop yield Yk of the previous year × the unit crop fertilizer absorption content Sk of the previous year; calculate the fertilizer loss amount, the fertilizer loss amount S in the current year = (1 - I%) × the total crop absorption Xp of fertilization in the current year / the area of the region; the fertilizer utilization rate I% in the current year = (the total crop absorption Xp of fertilization in the current year - the remaining soil fertilizer content Z after the crop is harvested in the current year) / the amount of fertilizer used for this cultivation H × 100%;

[0038] The result calculation sub-module: used to calculate the input-output ratio curve D, ; where P represents the price of the fertilizer, Py represents the crop product price, and solve for the minimum value of the input-output ratio numerical curve as the recommended fertilization.

[0039] Preferably, in the precise fertilization decision-making and plot-level display system of the present invention, the method for obtaining the unit crop fertilizer absorption content Sp of fertilization in the current year and the unit crop fertilizer absorption content Sm of non-fertilization is as follows:

[0040] Air-dry the plants of the harvested crops, use statistical means, randomly select some crops as samples, measure the content of fertilizer elements in the crops, and then estimate the unit crop fertilizer absorption content Sp of fertilization in the current year and the unit crop fertilizer absorption content Sm of non-fertilization according to the sample ratio;

[0041] The air-dried plants include crop seeds, stems and leaves, and roots. After crushing the air-dried plants, the determination of the fertilizer element content is carried out.

[0042] Use a convolutional neural network model to measure the fertilizer element content of the air-dried stems and leaves and the fertilizer element content of the grains.

[0043] The method includes the following steps:

[0044] A1: Training process of the convolutional neural network model device: Use the photos of the air-dried stems and leaves and grains with specific moisture content and the corresponding fertilizer element content of the stems and leaves and the fertilizer element content of the grains as the training set to train the convolutional neural network model device to obtain the convolutional neural network model device; when constructing the training set, divide several intervals according to the normal distribution based on the mass of the stems and leaves or grains, and select the corresponding number of samples to construct the training set. The stems and leaves or grains used as the training set are complete photos, and the parts in the photos that are not stems and leaves or grains are adjusted to be transparent through processing.

[0045] A2: Element content recognition process. Use the photos of the air-dried stems and leaves and grains with new unknown fertilizer element content of the stems and leaves and the fertilizer element content of the grains as the input, and use the output of the convolutional neural network model as the fertilizer element content of the stems and leaves and the fertilizer element content of the grains.

[0046] During the recognition process, also select the stems and leaves and grains whose mass of the stems and leaves or grains conforms to the normal distribution as samples. After summing the fertilizer element content in each recognized stem and leaf or grain and then dividing by the corresponding weight percentage, the fertilizer element content of the test plot is obtained.

[0047] The present invention also provides a computer storage medium, which stores one or more instructions, and the instructions are suitable for being loaded and executed by a processor to perform the above method.

[0048] The beneficial effects of the present invention are:

[0049] The present invention provides a precise fertilization decision-making and plot-level display method and system, including the step of reading map information; the step of marking the plots where crops are planted on the map according to the field ridges; the step of calculating the input-output ratio value for each plot; the step of marking the input-output ratio value for each plot or marking a specific color for each plot according to the input-output ratio value. By directly displaying the plot shape and input-output ratio on the map, relevant personnel can intuitively see the balanced fertilization status of the plots within a certain area. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The technical solutions of the present application will be further described below with reference to the drawings and embodiments.

[0051] Figure 1It is a schematic diagram after marking a plot on the map in the precise fertilization decision-making and plot-level display method of the embodiment of the present application;

[0052] Figure 2 It is a schematic diagram showing the input-output ratio value on the map in the precise fertilization decision-making and plot-level display method of the embodiment of the present application;

[0053] Figure 3 It is a flowchart of the precise fertilization decision-making and plot-level display method of the embodiment of the present application;

[0054] Figure 4 It is a schematic column diagram of "control", "stable", and "increase" for the fertilization amounts of N nitrogen, P phosphorus, and K potassium in the embodiment of the present application;

[0055] Figure 5 It is a structural diagram of the convolutional neural network model device of the embodiment of the present application. Detailed implementation manners

[0056] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0057] The technical solutions of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0058] Embodiment 1

[0059] This embodiment provides a precise fertilization decision-making and plot-level display method, and the flowchart is as Figure 3 shown, including the following steps:

[0060] S1: Read in map information;

[0061] S2: Mark the plots where crops are planted on the map according to the field ridges, as Figure 1 shown;

[0062] S3: Calculate the input-output ratio value curves of nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer on each plot, and solve the minimum value of the input-output ratio value curve as the recommended fertilization amount;

[0063] S4: Use the recommended fertilization amount minus the original fertilizer content of nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer in the plot as the fertilization recommendation plan, and mark the fertilization recommendation plan on the plots of the map information;

[0064] S5: Click on the corresponding plot on the map information to view the fertilization recommendation plans for nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer of the plot.

[0065] In step S4, the plots are divided into high, medium, and low fertilization degrees according to the ratio of the fertilization amount to the original fertilizer content of the plot. For example (only for illustration), if the fertilization amount is less than 0 times the original fertilizer content, it is "control" and is called low fertilization degree; if the fertilization amount is 0 - 20% of the original fertilizer content, it is "stable" and is called medium fertilization degree; if the fertilization amount is more than 20% of the original fertilizer content, it is "increase" and is called high fertilization degree. The fertilization recommendation scheme can be divided into 27 categories (permutations and combinations of three elements and three fertilization degrees):

[0066] Control nitrogen - control phosphorus - control potassium, control nitrogen - control phosphorus - stable potassium, control nitrogen - control phosphorus - increase potassium, control nitrogen - stable phosphorus - control potassium, control nitrogen - increase phosphorus - control potassium, control nitrogen - stable phosphorus - stable potassium, control nitrogen - stable phosphorus - increase potassium, control nitrogen - increase phosphorus - stable potassium, control nitrogen - increase phosphorus - increase potassium;

[0067] Stable nitrogen - control phosphorus - control potassium, stable nitrogen - control phosphorus - stable potassium, stable nitrogen - control phosphorus - increase potassium, stable nitrogen - stable phosphorus - control potassium, stable nitrogen - increase phosphorus - control potassium, stable nitrogen - stable phosphorus - stable potassium, stable nitrogen - stable phosphorus - increase potassium, stable nitrogen - increase phosphorus - stable potassium, stable nitrogen - increase phosphorus - increase potassium;

[0068] Increase nitrogen - control phosphorus - control potassium, increase nitrogen - control phosphorus - stable potassium, increase nitrogen - control phosphorus - increase potassium, increase nitrogen - stable phosphorus - control potassium, increase nitrogen - increase phosphorus - control potassium, increase nitrogen - stable phosphorus - stable potassium, increase nitrogen - stable phosphorus - increase potassium, increase nitrogen - increase phosphorus - stable potassium, increase nitrogen - increase phosphorus - increase potassium;

[0069] The above 27 categories can be classified and marked and finally displayed on the map for the operator to view. Since usually, there will not be 27 classification marks at the same time, usually 3 - 10 classification marks, different colors can be used to represent different classification marks to make it more intuitive to display on the map.

[0070] As Figure 4 shown, the figure shows a columnar schematic diagram of "control", "stable", and "increase" for the fertilization amounts of N nitrogen, P phosphorus, and K potassium. Currently, it is shown as "control nitrogen - stable phosphorus - stable potassium" in the figure;

[0071] The calculation method of the input - output ratio value in step S3 is as follows:

[0072] S31: Obtain the initial soil fertilizer content Z0 of the area where the crop to be planted is located before cultivation, obtain the current fertilizer price P, obtain the crop yield Yk of the previous year and the crop yield Yp of the current year, obtain the remaining soil fertilizer content Z after the current year's crop is harvested, obtain the unit crop fertilizer absorption content Sk of the previous year and the unit crop fertilizer absorption content Sp of the current year's fertilization, and the fertilizer amount used for this cultivation is H;

[0073] S32: Calculate the total crop absorption Xp of the current year's fertilization. The total crop absorption Xp of the current year's fertilization = the current year's crop yield Yp × the unit crop fertilizer absorption content Sp of the current year's fertilization;

[0074] Calculate the total crop uptake Xk in the previous year. The total crop uptake Xk in the previous year = the crop yield Yk in the previous year × the unit crop fertilizer uptake content Sk in the previous year;

[0075] Calculate the fertilizer loss. The fertilizer loss S in the current year = (1 - I%) × the total crop uptake Xp of the fertilizer applied in the current year / the area of the region; The fertilizer utilization rate I% in the current year = (the total crop uptake Xp of the fertilizer applied in the current year - the residual fertilizer content Z in the soil after crop harvest in the current year) / the amount of fertilizer used H in this cultivation × 100%;

[0076] S33: Calculate the input-output ratio curve D, ; where P represents the price of the fertilizer, and Py represents the price of the crop product. Solve for the minimum value of the input-output ratio numerical curve as the recommended fertilizer application rate.

[0077] The precise fertilization decision-making and plot-level display method of this embodiment calculates the input-output ratio D, which is the ratio of the additional net income from fertilization to the fertilization cost. In this embodiment, the profitability in the case of fertilization and non-fertilization is compared, and the economic value of the residual fertilizer in the region and the environmental impact value caused by fertilizer loss are comprehensively considered. The smaller the input-output ratio, the higher the economic efficiency. By calculating the input-output ratio of different regions (plots), it is convenient to select the optimal fertilization plan.

[0078] The fertilizer can be the most basic nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer;

[0079] During the calculation process, obtain the amounts of the three fertilizers H, the residual fertilizer content Z in the soil after crop harvest, the unit crop fertilizer uptake content Sk without fertilization, and the unit crop fertilizer uptake content Sp with fertilization in the current year respectively; Calculate the fertilizer losses of the three fertilizers respectively;

[0080] When calculating the input-output ratio D, the three fertilizers need to be accumulated into the formula described in S33.

[0081] ; i = 1 can represent nitrogen fertilizer, 2 can represent phosphorus fertilizer, and 3 can represent potassium fertilizer. Of course, other fertilizer elements can also be evaluated.

[0082] In step S32: The method for obtaining the unit crop fertilizer uptake content Sp with fertilization in the current year and the unit crop fertilizer uptake content Sm in the previous year is as follows:

[0083] Air-dry the plants of the harvested crops, use statistical means, randomly select some crops as samples, measure the content of fertilizer elements in the crops, and then estimate the unit crop fertilizer uptake content Sp with fertilization in the current year and the unit crop fertilizer uptake content Sm without fertilization according to the sample ratio.

[0084] The air-dried plants include crop seeds, stems and leaves, and roots. After crushing the air-dried plants, the determination of fertilizer element content is carried out.

[0085] Taking rice as an example, the methods for measuring yield and fertilizer element content are as follows:

[0086] Sampling: Randomly take 3 sample segments each 1 m long. After air-drying the whole crop, weigh the air-dried weight of the stems and leaves and the air-dried weight of the panicles. After threshing the panicles, weigh the air-dried grains and the air-dried grain weight (g). Since the rice roots are still in the ground during harvesting, the fertilizer element content of the roots is not calculated here.

[0087] Measure the fertilizer element content of the air-dried stems and leaves and the fertilizer element content of the grains (fertilizer element content of the husk + fertilizer element content of the rice grains). The measurement of the fertilizer element content of the stems and leaves and the fertilizer element content of the grains (fertilizer element content of the husk + fertilizer element content of the rice grains) is obtained by using national standards.

[0088] It is also possible to use a convolutional neural network model to measure the fertilizer element content of the stems and leaves and the fertilizer element content of the grains;

[0089] The specific method is as follows:

[0090] A1: Training process of the convolutional neural network model device: Use the photos of the air-dried stems and leaves and grains and their corresponding fertilizer element content of the stems and leaves and grains after air-drying to a specific moisture content as the training set to train the convolutional neural network model device to obtain the convolutional neural network model device. When constructing the training set, divide several intervals according to the normal distribution based on the mass of the stems and leaves or grains, and select the corresponding number of samples to construct the training set (that is, the number of samples close to the average value in the whole experimental plot is the largest, and the number of samples farther away from the average value is smaller;). The stems and leaves or grains used as the training set are complete photos, and the parts in the photos that are not stems and leaves or grains are adjusted to be transparent through processing;

[0091] A2: Element content recognition process. Use the photos of the air-dried stems and leaves and grains with unknown fertilizer element content of the stems and leaves and grains after air-drying to a specific moisture content as the input, and take the output of the convolutional neural network model as the fertilizer element content of the stems and leaves and the fertilizer element content of the grains;

[0092] During the recognition process, the stems and leaves and grains with the mass of the stems and leaves or grains conforming to the normal distribution are also selected as samples. After summing the fertilizer element content in each recognized stem and leaf or grain and then dividing by the corresponding weight percentage, the fertilizer element content of the experimental plot is obtained. Since the moisture content during training and recognition is quite the same, the influence of water on the recognition result can be ignored.

[0093] It should be noted that the convolutional neural network models for recognizing the fertilizer element content of the stems and leaves and the fertilizer element content of the grains should be constructed separately.

[0094] The photos in the training set have the same size as those of the air-dried stems, leaves and grains in the recognition procedure, and the pixel ratio of the photos to the actual objects is the same (for example, 10 pixels correspond to 1 cm of the object).

[0095] According to the statistical principle, the grain yield (i.e., crop yield) of the entire experimental plot is calculated according to the sampling ratio, and the fertilizer element content of the stems and leaves and the fertilizer element content of the grains in the entire experimental plot are calculated. According to the proportion of the elements in the fertilizer, the amount of fertilizer used by the crop is calculated. At the same time, constructing the training set and recognition samples based on the normal distribution of mass can improve the recognition accuracy.

[0096] The larger the amount of data during training, the better, and it should be at least no less than 10,000 pieces.

[0097] The basic principle of the convolutional neural network model device is: image input → feature calculation → output category.

[0098] Specifically, it includes:

[0099] Input layer;

[0100] Feature extraction layer, each layer includes a convolutional layer and a pooling layer, such as layer1-layer7, a total of 7 layers; (the size of the convolutional kernel in the convolutional layer gradually decreases, and the stride and kernel size of the pooling layer are the same)

[0101] First fully connected layer;

[0102] Second fully connected layer;

[0103] Output layer.

[0104] Common algorithms can be used for the training algorithm, such as the stochastic gradient descent algorithm, Adam algorithm, RMSProp algorithm, Adagrad algorithm, etc.

[0105] The convolutional neural network model device is composed of an input layer, several feature extraction layers, fully connected layers, and an output layer. The parameters of each layer, such as the pooling window size, stride size, convolutional kernel size, node number, etc., are all related to the photo size of the training set, which is not the inventive point of this embodiment and will not be elaborated here.

[0106] The fertilizer element content of the stems and leaves + the fertilizer element content of the grains is used as the total absorption amount of the crops in the entire experimental plot (divided into the total absorption amount of fertilized crops and the total absorption amount of unfertilized crops according to different source regions).

[0107] Example 2

[0108] This embodiment provides a precise fertilization decision-making and plot-level display system, which is characterized in that it includes:

[0109] A map module, in which there are plots with the range of planted crops marked according to the ridges.

[0110] A recommended fertilization amount calculation module, which is used to calculate the value curves of the input-output ratios of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer on each plot, and solve the minimum value of the input-output ratio value curve as the recommended fertilization amount.

[0111] A map marking module, which is used to take the recommended fertilization amounts of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer minus the original fertilizer content in the plot as the fertilization recommendation plan, and mark this fertilization recommendation plan on the plot of the map information.

[0112] A query module: which is used to display the fertilization recommendation plans of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer for the corresponding plot when clicking on the corresponding plot in the map information.

[0113] In the map marking module, the plots are divided into high, medium and low fertilization degrees according to the ratio of the fertilization amount to the original fertilizer content in the plot, and classified according to the three fertilization degrees of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer, and the classification is marked on the map. When clicking on the corresponding plot in the map information in the query module, view the classification mark of this plot.

[0114] The recommended fertilization amount calculation module specifically includes:

[0115] A data acquisition sub-module: which is used to obtain the initial soil fertilizer content Z0 of the area where the crop to be planted is located before cultivation, obtain the current fertilizer price P, obtain the crop yield Yk of the previous year and the crop yield Yp of the current year, obtain the remaining soil fertilizer content Z after the current year's crop is harvested, obtain the unit crop fertilizer absorption content Sk of the previous year and the unit crop fertilizer absorption content Sp of the current year's fertilization, and the amount of fertilizer used for this cultivation is H.

[0116] A data processing sub-module: which is used to calculate the total amount of fertilizer absorbed by the crops of the current year Xp, the total amount of fertilizer absorbed by the crops of the current year Xp = the crop yield Yp of the current year × the unit crop fertilizer absorption content Sp of the current year's fertilization; calculate the total amount of fertilizer absorbed by the crops of the previous year Xk, the total amount of fertilizer absorbed by the crops of the previous year Xk = the crop yield Yk of the previous year × the unit crop fertilizer absorption content Sk of the previous year; calculate the fertilizer loss amount, the fertilizer loss amount S of the current year = (1 - I%) × the total amount of fertilizer absorbed by the crops of the current year Xp / the area of the region; the current year's fertilizer utilization rate I% = (the total amount of fertilizer absorbed by the crops of the current year Xp - the remaining soil fertilizer content Z after the current year's crop is harvested) / the amount of fertilizer used for this cultivation H × 100%.

[0117] A result calculation sub-module: which is used to calculate the input-output ratio curve D, ; where P represents the price of the fertilizer, Py represents the price of the crop product, and solve the minimum value of the input-output ratio value curve as the recommended fertilization.

[0118] The results of the three fertilizers are obtained by separately calculating nitrogen fertilizer, phosphate fertilizer, and potassium fertilizer.

[0119] During the calculation processes of the data processing sub-module and the result calculation sub-module, the fertilizer amounts H of the three fertilizers, the remaining fertilizer content Z in the soil after the crops are harvested in the current year, the fertilizer absorption content Sk per unit crop in the previous year, and the fertilizer absorption content Sp per unit crop for the fertilizer applied in the current year are respectively obtained; the fertilizer losses of the three fertilizers are calculated separately.

[0120] When calculating the input-output ratio D, the three fertilizers need to be added to the formula in the result calculation sub-module.

[0121] Furthermore, the method for obtaining the fertilizer absorption content Sp per unit crop for the fertilizer applied in the current year and the fertilizer absorption content Sm per unit crop without fertilization is as follows:

[0122] The plants of the harvested crops are air-dried, and by using statistical means, some crops are randomly selected as samples to measure the content of fertilizer elements in the crops, and then the fertilizer absorption content Sp per unit crop for the fertilizer applied in the current year and the fertilizer absorption content Sm per unit crop without fertilization are estimated according to the sample ratio.

[0123] Preferably, for the intelligent decision-making device for agricultural balanced fertilization considering environmental impacts of the present invention, the air-dried plants include crop seeds, stems and leaves, and roots, and after the air-dried plants are crushed, the determination of the fertilizer element content is carried out.

[0124] Example 3

[0125] This embodiment provides a computer storage medium, which stores one or more instructions, and the instructions are suitable for being loaded and executed by a processor to perform the intelligent decision-making method for agricultural balanced fertilization considering environmental impacts as in Example 1.

[0126] Inspired by the above ideal embodiments according to the present application, through the above description content, relevant staff can completely make various changes and modifications without departing from the technical idea of this application. The technical scope of this application is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

[0127] Inspired by the above ideal embodiments according to the present application, through the above description content, relevant staff can completely make various changes and modifications without departing from the technical idea of this application. The technical scope of this application is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

[0128] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0129] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0130] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0131] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

Claims

1. A precise fertilization decision-making and plot-level display method, characterized in that, It includes the following steps: S1: Read in the map information; S2: Mark the plots for growing crops on the map according to the ridge; S3: Calculate the value curves of the input-output ratios of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer for each plot, and solve the minimum value of the input-output ratio value curve as the recommended fertilization amount; S4: Take the recommended fertilization amounts of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer minus the original fertilizer contents of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer in the plot as the fertilization recommendation plan, and mark this fertilization recommendation plan on the plots of the map information; S5: Click on the corresponding plot on the map information to view the fertilization recommendation plan of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer for this plot; The calculation method of the input-output ratio value of each fertilizer in step S3 is as follows: S31: Obtain the initial soil fertilizer content Z0 of the area where the crop to be planted is located before cultivation, obtain the current fertilizer price P, obtain the crop yield Yk of the previous year and the crop yield Yp of the current year, obtain the remaining soil fertilizer content Z after the crop is harvested in the current year, obtain the unit crop fertilizer absorption content Sk of the previous year and the unit crop fertilizer absorption content Sp of the current year's fertilization, and the amount of fertilizer used for this cultivation is H; S32: Calculate the total crop absorption amount Xp of the current year's fertilization. The total crop absorption amount Xp of the current year's fertilization = the crop yield Yp of the current year × the unit crop fertilizer absorption content Sp of the current year's fertilization; Calculate the total crop absorption amount Xk of the previous year. The total crop absorption amount Xk of the previous year = the crop yield Yk of the previous year × the unit crop fertilizer absorption content Sk of the previous year; Calculate the fertilizer loss amount. The current-year fertilizer loss amount S = (1 - I%) × the total crop absorption amount Xp of the current year's fertilization / the area of the region; the current-year fertilizer utilization rate I% = (the total crop absorption amount Xp of the current year's fertilization - the remaining soil fertilizer content Z after the crop is harvested in the current year) / the amount of fertilizer used for this cultivation H × 100%; S33: Calculate the input-output ratio curve D, where P represents the price of fertilizer and Py represents the price of crop products, and solve for the minimum value of the input-output ratio numerical curve as the recommended fertilization amount.

2. The precise fertilization decision-making and plot-level display method according to claim 1, characterized in that, In step S4, the plots are divided into high, medium and low fertilization degrees according to the ratio of the fertilization amount to the original fertilizer content of the plot, and classified according to the three fertilization degrees of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer, and the classification is marked on the map. In step S5, click on the corresponding plot on the map information to view the classification mark of this plot.

3. The precise fertilization decision-making and plot-level display method according to claim 2, characterized in that, Different classification marks are represented by different colors on the map.

4. The precise fertilization decision-making and plot-level display method according to claim 1, characterized in that, The obtaining methods of the unit crop fertilizer absorption content Sp of the current year's fertilization and the unit crop fertilizer absorption content Sm of non-fertilization are as follows: Air-dry the plants of the harvested crops, and use statistical means to randomly select some crops as samples to measure the fertilizer element contents in the crops, and then estimate the unit crop fertilizer absorption content Sp of the current year's fertilization and the unit crop fertilizer absorption content Sm of non-fertilization according to the sample ratio; The air-dried plants include crop seeds, stems and leaves, and roots. After the air-dried plants are crushed, the fertilizer element contents are measured; Use a convolutional neural network model to measure the fertilizer element contents of the air-dried stems and leaves and the fertilizer element contents of the grains; The method includes the following steps: A1: Training process of the convolutional neural network model device: Photos of stems, leaves and grains dried to a specific moisture content and their corresponding fertilizer element contents in stems, leaves and grains are used as the training set to train the convolutional neural network model device, and the convolutional neural network model device is obtained; When constructing the training set, several intervals are divided according to the normal distribution based on the mass of stems, leaves or grains, and corresponding numbers of samples are selected to construct the training set. The stems, leaves or grains used as the training set are complete photos, and the parts in the photos that are not stems, leaves or grains are adjusted to be transparent through processing; A2: Element content recognition process. Photos of stems, leaves and grains dried to a specific moisture content with new unknown fertilizer element contents in stems, leaves and grains are used as the input, and the output of the convolutional neural network model is used as the fertilizer element contents in stems, leaves and grains; In the recognition process, stems and leaves with the mass of stems or grains conforming to the normal distribution are also selected as samples. After summing up the fertilizer element contents in each recognized stem or grain and then dividing by the corresponding weight percentage, the fertilizer element content of the experimental plot is obtained.

5. A precise fertilization decision-making and plot-level display system, characterized in that, It includes: Map module. In the map module, there are plots with the range of planted crops marked according to the ridges; Recommended fertilizer application rate calculation module, which is used to calculate the production-to-input ratio numerical curves of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer on each plot, and solve the minimum value of the production-to-input ratio numerical curve as the recommended fertilizer application rate; Map marking module, which is used to take the recommended fertilizer application rates of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer minus the original fertilizer contents of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer in the plot as the fertilizer application recommendation plan, and mark this fertilizer application recommendation plan on the plot of the map information; Query module: It is used to display the fertilizer application recommendation plans of nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer for the corresponding plot when clicking on the map information; The recommended fertilizer application rate calculation module includes: Data acquisition sub-module: It is used to obtain the initial soil fertilizer content Z0 in the area where the crop to be planted is located before cultivation, obtain the current fertilizer price P, obtain the crop yield Yk in the previous year and the crop yield Yp in the current year, obtain the remaining soil fertilizer content Z after the crop is harvested in the current year, obtain the unit crop fertilizer absorption content Sk in the previous year and the unit crop fertilizer absorption content Sp of the fertilizer applied in the current year. The amount of fertilizer used in this cultivation is H; Data processing sub-module: It is used to calculate the total crop absorption Xp of the fertilizer applied in the current year. The total crop absorption Xp of the fertilizer applied in the current year = the crop yield Yp in the current year × the unit crop fertilizer absorption content Sp of the fertilizer applied in the current year; Calculate the total crop absorption Xk in the previous year. The total crop absorption Xk in the previous year = the crop yield Yk in the previous year × the unit crop fertilizer absorption content Sk in the previous year; Calculate the fertilizer loss. The fertilizer loss S in the current year = (1 - I%) × the total crop absorption Xp of the fertilizer applied in the current year / the area of the region; The fertilizer utilization rate I% in the current year = (the total crop absorption Xp of the fertilizer applied in the current year - the remaining soil fertilizer content Z after the crop is harvested in the current year) / the amount of fertilizer used H in this cultivation × 100%; Result calculation sub-module: used to calculate the input-output ratio curve D, where P represents the price of fertilizer, Py represents the price of crop products, and the minimum value of the input-output ratio numerical curve is solved as the recommended fertilization.

6. The precise fertilization decision-making and plot-level display system according to claim 5, characterized in that, In the map identification module, plots are divided into high, medium, and low fertilization degrees according to the ratio of the fertilization amount to the original fertilizer content of the plot, and classified according to the three fertilization degrees of nitrogen, phosphorus, and potassium fertilizers. The classification is marked on the map. In the query module, click on the corresponding plot in the map information to view the classification mark of the plot.

7. The precise fertilization decision-making and plot-level display system according to claim 5, characterized in that The method for obtaining the fertilizer absorption content Sp per unit crop with fertilization in the current year and the fertilizer absorption content Sm per unit crop without fertilization is as follows: Air-dry the plants of the harvested crops, and use statistical means to randomly select some crops as samples to measure the content of fertilizer elements in the crops. Then, estimate the fertilizer absorption content Sp per unit crop with fertilization in the current year and the fertilizer absorption content Sm per unit crop without fertilization according to the sample ratio. The air-dried plants include crop seeds, stems and leaves, and roots. After pulverizing the air-dried plants, the determination of the fertilizer element content is carried out. Use a convolutional neural network model to measure the fertilizer element content of the air-dried stems and leaves and the fertilizer element content of the grains. The method includes the following steps: A1: The training process of the convolutional neural network model device: Use the photos of the air-dried stems and leaves and grains with a specific moisture content and the corresponding fertilizer element content of the stems and leaves and the fertilizer element content of the grains as the training set to train the convolutional neural network model device to obtain the convolutional neural network model device. When constructing the training set, divide several intervals according to the normal distribution based on the mass of the stems and leaves or grains, and select the corresponding number of samples to construct the training set. The stems and leaves or grains used as the training set are complete photos, and the parts of the photos that are not stems and leaves or grains are adjusted to be transparent through processing. A2: The element content recognition process: Use the photos of the air-dried stems and leaves and grains with a new unknown fertilizer element content of the stems and leaves and the fertilizer element content of the grains as the input, and use the output of the convolutional neural network model as the fertilizer element content of the stems and leaves and the fertilizer element content of the grains. During the recognition process, select the stems and leaves and grains whose mass of the stems and leaves or grains conforms to the normal distribution as samples. After summing the fertilizer element content in each recognized stem and leaf or grain and then dividing by the corresponding weight percentage, the fertilizer element content of the experimental plot is obtained.

8. A computer storage medium, characterized in that The computer storage medium stores one or more instructions, and the instructions are suitable for being loaded and executed by a processor to perform the method according to any one of claims 1-4.

Images