A rice nitrogen fertilization method, system, device and storage medium

Abstract

The application discloses a rice nitrogen fertilizer application method, system, device and storage medium, comprising obtaining a to-be-detected rice photo of a to-be-fertilized area, inputting the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo, when the predicted growth grade is less than a preset grade, calculating a first nitrogen application interval according to the predicted growth period, calculating a second nitrogen application interval according to the predicted growth period and a preset corresponding proportion of each period, and calculating a rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval, thereby improving the fertilizer utilization rate, reducing the fertilizer cost, and improving the quality of the planned crops.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent fertilization, and in particular to a rice nitrogen fertilization method, system, device and storage medium. BACKGROUND

[0002] As one of the main grain crops in China, the yield of rice not only affects the development of agricultural economy, but also affects people's life. The management of the whole growth cycle of rice plays a decisive role in its final yield and quality, and the fertilization directly affects the yield and quality of rice in the whole growth cycle of rice.

[0003] But the current fertilization method can be summarized as soil testing fertilization method, fertilizer effect function method and nutrient diagnosis method, among which, the soil testing fertilization method is a method of formulating accurate fertilization scheme according to the nutrient status of soil. But the soil testing fertilization method needs soil testing, which needs certain technology and equipment support, and the cost is high. Secondly, different crops have different nutrient requirements, and different fertilization schemes need to be formulated according to different crops, which is more cumbersome to operate. The test cycle of the fertilizer effect function method is relatively long, which needs to consume a lot of manpower and material resources, and needs professional technical personnel to model and analyze, and the repeatability between years is poor, which is easy to appear "saddle type" curve, and the error of predicted fertilization amount is large. The nutrient diagnosis method is a method for evaluating the nutrient status of plants, which analyzes the nutrient content in plant tissues to judge whether the plant lacks a certain nutrient or absorbs too much, so as to guide the decision of fertilization management, but in order to carry out nutrient diagnosis, laboratory test is needed, including sampling, sample preparation and analysis, which involves certain cost and time cost, and the sampling process may have certain influence on the result, and the analysis and interpretation of nutrient diagnosis result also needs certain professional knowledge. SUMMARY

[0004] The present application aims at at least solving the technical problems in the prior art. To this end, the present application provides a rice nitrogen fertilization method, system, device and storage medium, which can improve the fertilizer utilization rate, reduce the fertilization cost and improve the quality of planned crops.

[0005] In a first aspect of the present application, a rice nitrogen fertilization method is provided, comprising the following steps:

[0006] Obtaining a to-be-detected rice photo of a to-be-fertilized area;

[0007] Inputting the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo;

[0008] When the predicted growth grade is less than a preset grade, calculating a first nitrogen application interval according to the predicted growth period;

[0009] calculating a second nitrogen application interval according to the predicted growth period and a preset proportion corresponding to each period;

[0010] calculating the nitrogen fertilizer application amount of the rice corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval.

[0011] According to the embodiments of the present application, at least the following technical effects are achieved:

[0012] The method comprises the following steps: obtaining a to-be-detected rice photo of a to-be-fertilized area, inputting the to-be-detected rice photo into a trained classification neural network model, obtaining a predicted growth period and a predicted growth grade of the to-be-detected rice photo, when the predicted growth grade is less than a preset grade, calculating a first nitrogen application interval according to the predicted growth period, calculating a second nitrogen application interval according to the predicted growth period and a preset proportion corresponding to each period, and calculating the nitrogen fertilizer application amount of the rice corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval. The method improves the fertilizer utilization rate, reduces the fertilization cost, and improves the quality of the planned crops.

[0013] According to some embodiments of the present application, the training process of the classification neural network model comprises the following steps:

[0014] obtaining a preprocessed historical rice photo;

[0015] labeling the preprocessed historical rice photo according to a preset growth period and a preset growth grade to obtain a labeled training image;

[0016] constructing an initial classification neural network model, inputting the labeled training image into the initial classification neural network model for training, and obtaining the trained classification neural network model.

[0017] According to some embodiments of the present application, when the predicted growth grade is less than the preset grade, the first nitrogen application interval is calculated according to the predicted growth period, which comprises the following steps:

[0018] obtaining a historical SPAD value and a historical nitrogen fertilizer application amount of a historical rice photo;

[0019] constructing a fitting equation corresponding to each preset growth period according to the historical SPAD value and the historical nitrogen fertilizer application amount;

[0020] When the predicted growth grade is less than the preset grade, the first nitrogen application interval is calculated according to the predicted growth period and the corresponding fitting equation.

[0021] According to some embodiments of the present application, the first nitrogen application interval is calculated according to the predicted growth period and the corresponding fitting equation, which comprises the following steps:

[0022] According to the predicted growth period and the preset growth period corresponding SPAD table data matching, the first SPAD value interval corresponding to the predicted growth period is obtained;

[0023] According to the first SPAD value interval and the preset growth period corresponding preset SPAD minimum value SPAD value calculation, the second SPAD value interval is obtained.

[0024] According to the second SPAD value interval and the corresponding fitting equation interval calculation, the first nitrogen application interval is obtained.

[0025] According to some embodiments of the present application, the second nitrogen application interval is calculated according to the predicted growth period and the preset each period corresponding weight, comprising:

[0026] The yield range of rice under the optimal nitrogen application level and the nitrogen application amount of the previous period of the predicted growth period are obtained.

[0027] The total nitrogen application amount is calculated according to the yield range of rice under the optimal nitrogen application level.

[0028] The predicted nitrogen application interval of the predicted growth period is calculated according to the total nitrogen application amount and the preset each period corresponding weight.

[0029] According to the predicted nitrogen application interval and the nitrogen application amount of the previous period, the nitrogen application amount is calculated, and the second nitrogen application interval is obtained.

[0030] According to some embodiments of the present application, the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo is calculated according to the first nitrogen application interval and the second nitrogen application interval, comprising:

[0031] The minimum value of the first nitrogen application interval and the minimum value of the second nitrogen application interval are obtained, and the maximum value of the two minimum values is taken as the minimum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo.

[0032] The maximum value of the first nitrogen application interval and the maximum value of the second nitrogen application interval are obtained, and the minimum value of the two maximum values is taken as the maximum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo.

[0033] The nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo is obtained according to the minimum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo and the maximum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo.

[0034] According to some embodiments of the present application, the rice nitrogen fertilizer application method further comprises:

[0035] The nitrogen content of the to-be-purchased fertilizer product is obtained.

[0036] According to the nitrogen content and the rice nitrogen fertilizer application amount, the number of purchase bags of the to-be-purchased chemical fertilizer product is calculated.

[0037] In a second aspect, the present application provides a rice nitrogen fertilizer application system, comprising:

[0038] a data acquisition module configured to acquire a to-be-detected rice photo of a to-be-fertilized area;

[0039] a prediction module configured to input the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo;

[0040] a first nitrogen application interval calculation module configured to, when the predicted growth grade is less than a preset grade, calculate a first nitrogen application interval according to the predicted growth period;

[0041] a second nitrogen application interval calculation module configured to calculate a second nitrogen application interval according to the predicted growth period and a preset corresponding proportion of each period;

[0042] a rice nitrogen fertilizer application amount calculation module configured to calculate a rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval.

[0043] The system acquires a to-be-detected rice photo of a to-be-fertilized area, inputs the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo, calculates a first nitrogen application interval according to the predicted growth period when the predicted growth grade is less than a preset grade, calculates a second nitrogen application interval according to the predicted growth period and a preset corresponding proportion of each period, and calculates a rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval, thereby improving fertilizer utilization rate, reducing fertilization cost, and improving the quality of planned crops.

[0044] In a third aspect, the present application provides a rice nitrogen fertilizer application electronic device, comprising at least one control processor and a memory in communication connection with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the above-mentioned rice nitrogen fertilizer application method.

[0045] In a fourth aspect, the present application provides a computer readable storage medium storing computer executable instructions for causing a computer to execute the above-mentioned rice nitrogen fertilizer application method.

[0046] It should be noted that the beneficial effects between the second to fourth aspects of the present application and the prior art are the same as the beneficial effects between the above-mentioned rice nitrogen fertilizer application system and the prior art, which will not be described here.

[0047] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0048] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the appended drawings.

[0049] Figure 1 is a flow chart of a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0050] Figure 2 is a schematic diagram of a rice tillering stage according to a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0051] Figure 3 is a schematic diagram of a rice heading stage according to a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0052] Figure 4 is a schematic diagram of a rice yield promotion stage according to a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0053] Figure 5 is a schematic diagram of a relationship between a field nitrogen application amount and a SPAD value at a rice tillering stage according to a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0054] Figure 6 is a schematic diagram of a relationship between a field nitrogen application amount and a SPAD value at a rice heading stage according to a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0055] Figure 7 is a schematic diagram of a relationship between a field nitrogen application amount and a SPAD value at a rice yield promotion stage according to a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0056] Figure 8 is a schematic diagram of a rice yield at different nitrogen application levels at a rice yield promotion stage according to a rice nitrogen fertilizer application method according to an embodiment of the present application;

[0057] Figure 9 is a schematic diagram of a structure of a rice nitrogen fertilizer application system according to an embodiment of the present application. DETAILED DESCRIPTION

[0058] Embodiments of the present application are described in detail below with reference to several drawings. The embodiments described below are illustrative only and are not intended to limit the present application, as defined by the appended claims, in any way.

[0059] In the description of the present application, if there is a description to first, second, etc. is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or the order of the indicated technical features.

[0060] In the description of the present application, it is to be understood that the orientation description, such as up, down, etc. indicates the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present application.

[0061] In the description of the present application, it is to be understood that, unless otherwise explicitly limited, the words such as setting, mounting, connecting, etc. should be broadly understood, and the person skilled in the art can reasonably determine the specific meaning of the above words in the present application in combination with the specific content of the technical solution.

[0062] Before introducing the present application, the fertilization amount of each growth period of rice is briefly introduced.

[0063] Fertilization amount: The fertilization amount of rice can be calculated according to the expected yield, the nutrient requirement of rice, the supply amount of soil nutrients, and the nutrient content and utilization rate of the applied fertilizer.

[0064] Fertilization period: It can be divided into base fertilizer, tillering fertilizer, ear fertilizer, and grain fertilizer (to be selected according to the growth potential of rice) four periods. The fertilization time and distribution ratio of each period are as follows.

[0065] Base fertilizer: It is applied into the soil before rice transplanting, combined with the last time of field plowing.

[0066] Tillering fertilizer: The tillering period is an important period to increase the number of plants, and is applied half a month after transplanting or seedling.

[0067] Ear fertilizer: It is divided into flower promoting fertilizer and flower preserving fertilizer. Flower promoting fertilizer is applied during the differentiation period of ear axis to the differentiation period of glume flower, and the application of nitrogen at this period can increase the number of glume flowers per ear. Flower preserving fertilizer is applied slightly before the pollen cell meiosis period, and has the effects of preventing glume flower degradation and increasing stem sheath storage accumulation.

[0068] Granular fertilizer: Granular fertilizer can prolong leaf function, increase photosynthetic intensity, increase grain weight, and reduce empty grains. Especially for rice fields with small plant populations and varieties with large panicles and long grain-filling periods, it is recommended to apply a small amount of urea, but it is important to avoid excessive nitrogen to prevent excessive vegetative growth and delayed maturity.

[0069] Main fertilization methods for rice:

[0070] The "light-heavy-supplemented" method involves applying sufficient and appropriate amounts of basal fertilizer and tillering fertilizer, along with reasonable application of panicle fertilizer and appropriate application of grain fertilizer. This aims to achieve early and stable growth, prevent excessive growth in the early stage, promote flowering in the middle stage, and prevent premature senescence in the later stage. This method ensures a sufficient number of panicles while also promoting large panicles and heavy grains. This fertilization method is commonly used for single-season late-season rice and late-maturing mid-season rice in southern China.

[0071] Early-Stability to Mid-Attack Method: This method saves fertilizer, ensures stable and high yields. It mainly aims to increase the effective tillering rate, promote larger ears to improve the grain filling rate, and increase grain weight to achieve higher yields. Characteristics: Strong plants with large tillers in a small population; early tillering control; robust stems and strong roots; mid-stage promotion of large ears; and mid-to-late-stage promotion of grain filling rate and ear weight.

[0072] The "Early Promotion-Mid-Control-Later Supplementation" method is similar to the "V"-shaped fertilization method. It involves heavy application of base fertilizer and tillering fertilizer, with appropriate application of granular fertilizer, aiming to achieve "vigorous growth in the early stage, stable development in the mid-stage, and robust growth in the late stage." This fertilization method is widely used in Northeast China's rice-growing areas, most early rice fields in southern China, and wheat-stubble rice fields in North China. Its drawback is that excessive early growth can easily lead to dense canopy cover in the field, resulting in more severe pests and diseases.

[0073] As one of my country's main food crops, rice yield not only affects agricultural economic development but also people's lives. Management throughout the entire rice growth cycle plays a decisive role in its final yield and quality, and fertilization directly impacts both yield and quality throughout the rice's growth cycle.

[0074] But the current fertilization method can be summarized as soil testing fertilization method, fertilizer effect function method and nutrient diagnosis method, wherein the soil testing fertilization method is a method for formulating an accurate fertilization scheme according to the nutrient status of the soil. But the soil testing fertilization method needs soil testing, which needs certain technical and equipment support, and the cost is high. Secondly, different crops have different nutrient requirements, and different fertilization schemes need to be formulated according to different crops, which is more complicated. The test cycle of the fertilizer effect function method is relatively long, and a large amount of manpower and material resources are needed, and professional technical personnel are needed for modeling and analysis, and the repeatability between years is poor, and the "saddle type" curve is easy to appear, and the error of predicted fertilization amount is large. The nutrient diagnosis method is a method for evaluating the nutrient status of plants, which judges whether the plants lack a certain nutrient or absorb too much by analyzing the nutrient content in the plant tissues, so as to guide the decision-making of fertilization management, but in order to carry out nutrient diagnosis, laboratory testing is needed, including sampling, sample preparation and analysis, which involves certain cost and time cost, and the sampling process may have certain influence on the result, and the analysis and interpretation of nutrient diagnosis result also need certain professional knowledge.

[0075] In order to solve the above technical defects, with reference to Figure 1 The application also provides a rice nitrogen fertilization method, comprising:

[0076] Step S101, obtaining a to-be-detected rice photo of a to-be-fertilized area;

[0077] Step S102, inputting the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo;

[0078] Step S103, when the predicted growth grade is less than a preset grade, calculating a first nitrogen application interval according to the predicted growth period;

[0079] Step S104, calculating a second nitrogen application interval according to the predicted growth period and a preset corresponding proportion of each period;

[0080] Step S105, calculating a rice nitrogen fertilization amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval.

[0081] The method obtains a to-be-detected rice photo of a to-be-fertilized area, inputs the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo, calculates a first nitrogen application interval according to the predicted growth period when the predicted growth grade is less than a preset grade, calculates a second nitrogen application interval according to the predicted growth period and a preset corresponding proportion of each period, and calculates a rice nitrogen fertilization amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval, thereby improving the fertilizer utilization rate, reducing the fertilization cost, and improving the quality of the planned crops.

[0082] In some embodiments, the training process of the classification neural network model comprises:

[0083] Step S201, obtaining the pre-processed historical rice photos;

[0084] Step S202, marking the pre-processed historical rice photos according to the preset growth period and the preset growth vigor grade to obtain the marked training images;

[0085] Step S203, constructing an initial classification neural network model, inputting the marked training images into the initial classification neural network model for training to obtain a trained classification neural network model.

[0086] Specifically, the preset growth period includes the tillering period, the heading period, the yield promotion period and others, and the preset growth vigor grade includes zero level, first level, second level and third level.

[0087] Specifically, the rice growth cycle has four main fertilization stages, namely, base fertilizer, tillering fertilizer, ear fertilizer and grain fertilizer, wherein the base fertilizer is applied before rice seeding and transplanting, and the tillering fertilizer, ear fertilizer and grain fertilizer correspond to the division period, the heading period and the yield promotion period, respectively.

[0088] Specifically, the present application divides the rice growth period into four categories: tillering period, heading period, yield promotion period and others, but only in the tillering period, heading period and yield promotion period, the present application will give the fertilization strategy.

[0089] For the growth vigor of rice, the present application ensures that all other factors are the same except for the different fertilization amounts in the rice fertilization experiment. Four groups of control experiments are constructed, with four nitrogen application levels, namely N0 (0), N1 (75 kg·hm -2 ), N2 (150 kg·hm -2 ) and N3 (225 kg·hm -2 ), corresponding to poor, relatively poor, normal and over-fertilization, respectively. Each nitrogen application treatment adopts the method of split fertilization, and the nitrogen application amounts of base fertilizer, tillering fertilizer, ear fertilizer and grain fertilizer account for 35%, 20%, 30% and 15% of the total nitrogen application amount, respectively. The application amounts of phosphorus fertilizer and potassium fertilizer are consistent in each treatment, the phosphorus fertilizer is applied as base fertilizer once, the application amount is 96 kg P2O5·hm -2 , and the potassium fertilizer is applied half in base fertilizer and half in ear fertilizer, the total application amount is 135 kg K2O·hm -2 .

[0090] Therefore, the corresponding labels of each photo are tillering period N0: 0, tillering period N1: 1, tillering period N2: 2, tillering period N3: 3, …, yield promotion period N0: 12, yield promotion period N1: 13, yield promotion period N2: 14, yield promotion period N3: 15, a total of 4x4 = 16 classification labels.

[0091] The same data set is used to identify the growth period and growth of rice, and the same deep learning algorithm, ResNet50, is used. In order to better extract the features of the data, the neural network needs to design a deeper network level and increase the number of network nodes, but with the deepening of the level of the deep neural network and the increase of the number of nodes, the traditional neural network will have the problem of gradient disappearance or gradient explosion, and ResNet50 can solve this problem well. ResNet50 is a deep residual network (ResNet), mainly used to solve the problem of gradient disappearance in neural network training. The "residual module" is introduced, which can design a very deep network structure without increasing the additional calculation amount, so that the neural network can learn more complex features. The convolution module in the ResNet50 network structure does not change the size of the residual block, but only changes the dimension of the residual block. In the ResNet50 network structure, the residual block has three layers of convolution, so the network has a total of 1+3x(3+4+6+3)=49 convolution layers, plus the last fully connected layer, a total of 50 layers. The core idea of the residual module is to introduce an identity mapping, so that the input of the network can be directly transmitted to the output, and a residual function is introduced to describe the difference between the input and the output. This difference is added to the output to obtain the final output result. In this way, the residual module can help the network better learn and optimize features, improve classification accuracy and robustness.

[0092] The original ResNet50 is a 1000-class network, and the data set used is ImageNet-1000, but the network output of this application is only 16 classifications, so the fully connected output of ResNet50 needs to be changed from 1000 to 16.

[0093] This application uses a pre-training mechanism, which only changes the last fully connected model of the original model. Not only can it speed up the training of the network, but it also retains the ability of the previously trained network to extract features, and can reduce the dependence on data volume, without training a neural network from scratch.

[0094] In some embodiments, when the predicted growth level is less than the preset level, a first nitrogen application interval is calculated according to the predicted growth period, comprising:

[0095] The historical SPAD value and the historical nitrogen fertilizer application amount of the historical rice photo are obtained;

[0096] According to the historical SPAD value and the historical nitrogen fertilizer application amount, a fitting equation corresponding to each preset growth period is constructed;

[0097] When the predicted growth level is less than the preset level, the first nitrogen application interval is calculated according to the predicted growth period and the corresponding fitting equation.

[0098] Specifically, referring to Figures 2 to 7 In the process of taking photos of the growth of rice, the SPAD-502 chlorophyll meter is used to determine the chlorophyll value of different rice fields, and 3 main stem top third and fourth leaves, i.e. top 3 leaves (L3) and top 4 leaves (L4), are randomly determined at five points in the southeast, northwest and center of the rice field (taking photos). When measuring, select 1 / 2 of the fully expanded or semi-expanded leaves and the positions 3 cm above and below the 1 / 4 or 3 / 4 of the leaf width. The average of the SPAD values of each plant is taken as the SPAD value of the plant, and the average of the SPAD values of 3 plants in each corner is taken as the SPAD value of the corner, and the average of the SPAD values of all corners is taken as the SPAD value of the field (taking photos). The camera is used to take photos at about 7 days before and after the tillering stage, heading stage and yield promotion stage (long seed stage) of the rice field, and 100 photos are taken for each field in each period, and a total of 500 photos are taken. The schematic diagram of the rice photos in different periods is shown in Figure 3 .

[0099] Specifically, the SPAD values and the data of the nitrogen application amount of the rice field sampled in different periods are plotted as shown in Figures 5 to 7 .

[0100] Specifically, referring to Table 1, Table 1 is an equation obtained by modeling the data SPAD and nitrogen application amount measured and recorded in three periods.

[0101] Table 1

[0102]

[0103] Wherein, x is the nitrogen application amount, and y is the measured SPAD value.

[0104] In some embodiments, according to the prediction growth period and the corresponding fitting equation, a first nitrogen application interval is obtained by interval calculation, including:

[0105] According to the prediction growth period and the corresponding SPAD table of the preset growth period, data matching is performed to obtain a first SPAD value interval corresponding to the prediction growth period;

[0106] According to the first SPAD value interval and the preset minimum SPAD value corresponding to the preset growth period, a second SPAD value interval is obtained by SPAD value calculation;

[0107] According to the second SPAD value interval and the corresponding fitting equation, a first nitrogen application interval is obtained by interval calculation.

[0108] Specifically, referring to Table 2, Table 2 is the minimum and maximum values of SPAD under different fertilization levels in different periods of rice.

[0109] Table 2

[0110]

[0111] Specifically, according to the rice growth level, the SPAD range of the normal rice in the growth period is found, the minimum SPAD and the maximum SPAD corresponding to the table 2 are found, and the nitrogen fertilizer increment N is calculated according to the SPAD and the fertilization relationship according to table 1 and table 2. SMin and N SMax The minimum value in table 1 is subtracted from the minimum SPAD and the maximum SPAD of the normal rice in table 2 as the y value, and the formula is substituted to obtain the x value.

[0112] In some embodiments, the second nitrogen application interval is calculated according to the predicted growth period and the preset proportion corresponding to each period, comprising:

[0113] The yield range of the rice under the optimal nitrogen application level and the nitrogen application amount of the previous period of the predicted growth period are obtained;

[0114] The total nitrogen application amount is calculated according to the yield range of the rice under the optimal nitrogen application level;

[0115] The predicted nitrogen application interval of the predicted growth period is calculated according to the total nitrogen application amount and the preset proportion corresponding to each period;

[0116] The nitrogen application amount is calculated according to the predicted nitrogen application interval and the nitrogen application amount of the previous period, and the second nitrogen application interval is obtained.

[0117] Specifically, refer to table 3, table 3 is some numerical value of rice growth (harvest period) under different nitrogen application levels.

[0118] Table 3

[0119]

[0120] Referring to table 3 and Figure 8 , Figure 8 The x-axis is the nitrogen application amount, and the Y-axis is the rice yield; a quadratic curve is used for fitting, and the fitting equation is: y=a+b·x+c·x 2 , the parameters are: a=300.25, b=2.463, c=-0.0071, and the correlation coefficient R 2 is 0.987. The fitting curve is as Figure 8 . It can be seen from table 3 and Figure 8 that with the increase of the nitrogen application amount, the yield per mu of rice also increases, but after the optimal nitrogen application amount, the increase of the yield of rice becomes slow. It can be calculated that the increment of the yield of rice per kg / mu of nitrogen fertilizer under four different nitrogen application levels is 2.16, 0.53 and 0.04 respectively. It can be seen that subsequent increase of the fertilizer amount cannot bring more benefits, therefore, in order to avoid waste of nitrogen fertilizer and reduce environmental pollution, the application of nitrogen fertilizer should be controlled.

[0121] Specifically, the yield of rice under the optimal nitrogen application level ranges from 450 to 500 kg per mu, and according to the yield range, the total nitrogen application amount N Figure 8 Correspondingly, the total nitrogen application amount N TMin , N TMax According to the foregoing, the split fertilization method is used for each nitrogen application amount treatment, and the nitrogen application amounts of base fertilizer, tillering fertilizer, ear fertilizer and grain fertilizer account for 35%, 20%, 30% and 15% of the total nitrogen application amount respectively, so as to obtain the nitrogen application proportion of each period, and according to the proportion, it is calculated that how much nitrogen fertilizer TN TMin , TN TMax , the nitrogen application amount N P of the previous period is obtained, TN TMin , TN TMax and N P are calculated, and the nitrogen application increment range N NMin , N NMax .

[0122] In some embodiments, the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo is calculated according to the first nitrogen application interval and the second nitrogen application interval, comprising:

[0123] Obtaining the minimum value of the first nitrogen application interval and the minimum value of the second nitrogen application interval; and taking the maximum value of the two minimum values as the minimum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo;

[0124] Obtaining the maximum value of the first nitrogen application interval and the maximum value of the second nitrogen application interval; and taking the minimum value of the two maximum values as the maximum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo;

[0125] According to the minimum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo and the maximum value of the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo, the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo is obtained.

[0126] Specifically, the nitrogen fertilizer application amount of rice corresponding to the to-be-detected rice photo is max(N SMin , N NMin )~min(N SMax , N NMax ).

[0127] In some embodiments, the rice nitrogen fertilizer application method further comprises:

[0128] Obtaining the nitrogen content of the to-be-purchased fertilizer product;

[0129] According to the nitrogen content and the nitrogen fertilizer application amount of rice, the purchase bag number of the to-be-purchased fertilizer product is calculated.

[0130] Specifically, according to the nitrogen content of different chemical fertilizer products, the number of bags to be purchased is calculated, and one decimal place is reserved. Since there may be remaining chemical fertilizers (previously remaining or used), the user decides whether to purchase an integer bag, so one decimal place is reserved.

[0131] In order to facilitate the understanding of the person skilled in the art, a set of experimental data is provided below:

[0132] In addition, with reference to Figure 9 An embodiment of the present application provides a rice nitrogen fertilizer application system, which comprises a data acquisition module 1100, a prediction module 1200, a first nitrogen application interval calculation module 1300, a second nitrogen application interval calculation module 1400 and a rice nitrogen fertilizer application amount calculation module 1500, wherein:

[0133] The data acquisition module 1100 is used for acquiring a to-be-detected rice photo of a to-be-fertilized area.

[0134] The prediction module 1200 is used for inputting the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo.

[0135] The first nitrogen application interval calculation module 1300 is used for calculating a first nitrogen application interval according to the predicted growth period when the predicted growth grade is less than a preset grade.

[0136] The second nitrogen application interval calculation module 1400 is used for calculating a second nitrogen application interval according to the predicted growth period and a preset corresponding proportion of each period.

[0137] The rice nitrogen fertilizer application amount calculation module 1500 is used for calculating a rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval.

[0138] The system acquires a to-be-detected rice photo of a to-be-fertilized area, inputs the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo, calculates a first nitrogen application interval according to the predicted growth period when the predicted growth grade is less than a preset grade, calculates a second nitrogen application interval according to the predicted growth period and a preset corresponding proportion of each period, and calculates a rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval, thereby improving fertilizer utilization rate, reducing fertilization cost and improving the quality of planned crops.

[0139] It should be noted that the system embodiment and the above-mentioned system embodiment are based on the same inventive concept, and therefore the related content of the above-mentioned method embodiments is also applicable to the system embodiment, which will not be described here again.

[0140] The application further provides a rice nitrogen fertilizer application electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the rice nitrogen fertilizer application method when executing the computer program.

[0141] The processor and the memory can be connected through a bus or other means.

[0142] The memory, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0143] The non-transitory software programs and instructions required for the rice nitrogen fertilizer application method of the above-mentioned embodiments are stored in the memory, and when executed by the processor, the rice nitrogen fertilizer application method in the above-mentioned embodiments is executed, for example, the method steps S101 to S105 in the above-mentioned Figure 1 are executed.

[0144] The application further provides a computer readable storage medium, which stores computer executable instructions for executing the rice nitrogen fertilizer application method.

[0145] The computer readable storage medium stores computer executable instructions, which are executed by a processor or a controller, for example, by a processor in the above-mentioned electronic device embodiment, so that the above-mentioned processor executes the rice nitrogen fertilizer application method in the above-mentioned embodiments, for example, executes the method steps S101 to S105 in the above-mentioned Figure 1

[0146] ​As will be appreciated by one of ordinary skill in the art, all or some of the steps, systems, and techniques disclosed herein can be embodied in software, firmware, hardware, or any suitable combination thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as an application- specific integrated circuit. Such software can be distributed on computer readable media, which can comprise computer storage media (or non-transitory media), and communication media (or transitory media). As will be appreciated by one of ordinary skill in the art, the term computer storage media includes all physical and tangible computer storage media, such as a volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Further, as will be appreciated by one skilled in the art, communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. In the foregoing specification, the concepts have been described with reference to specific embodiments. However, one of ordinary skill in the art appreciates that various modifications and changes can be made without departing from the scope of the application as set forth in the claims below. Accordingly, the specification is to be regarded in an illustrative rather than a restrictive sense.

[0147] The embodiments of the present application disclosed above are only used to explain the principle of the present application, and the present application is not limited to the above embodiments. Various changes can be made without departing from the spirit of the present application.

Claims

1. A method for nitrogen fertilizer application for rice, characterized by, The rice nitrogen fertilizer application method comprises: obtaining a to-be-detected rice photo of a to-be-fertilized area; inputting the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo; when the predicted growth grade is less than a preset grade, calculating a first nitrogen application interval according to the predicted growth period, specifically as follows: obtaining historical SPAD values and historical nitrogen fertilizer application amounts of historical rice photos; constructing a fitting equation corresponding to each preset growth period according to the historical SPAD values and the historical nitrogen fertilizer application amounts; when the predicted growth grade is less than the preset grade, performing interval calculation according to the predicted growth period and the corresponding fitting equation to obtain the first nitrogen application interval, specifically as follows: performing data matching according to the predicted growth period and a SPAD table corresponding to a preset growth period to obtain a first SPAD value interval corresponding to the predicted growth period; performing SPAD value calculation according to the first SPAD value interval and a preset minimum SPAD value corresponding to the preset growth period to obtain a second SPAD value interval; performing interval calculation according to the second SPAD value interval and the corresponding fitting equation to obtain the first nitrogen application interval; calculating a second nitrogen application interval according to the predicted growth period and corresponding proportions of preset periods, specifically as follows: obtaining a yield range of rice under an optimal nitrogen application level and a nitrogen application amount of a previous period of the predicted growth period; calculating a total nitrogen application amount according to the yield range of rice under the optimal nitrogen application level; calculating a predicted nitrogen application interval of the predicted growth period according to the total nitrogen application amount and the corresponding proportions of the preset periods; performing nitrogen application amount calculation according to the predicted nitrogen application interval and the nitrogen application amount of the previous period to obtain the second nitrogen application interval; calculating a rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval, specifically as follows: obtaining a minimum value of the first nitrogen application interval and a minimum value of the second nitrogen application interval; and taking a maximum value of the two minimum values as a minimum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo; obtaining a maximum value of the first nitrogen application interval and a maximum value of the second nitrogen application interval; and taking a minimum value of the two maximum values as a maximum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo; obtaining the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the minimum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo and the maximum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo.

2. The method for nitrogenous fertilizer application to rice according to claim 1, characterized by, The training process of the classification neural network model comprises: obtaining preprocessed historical rice photos; labeling the preprocessed historical rice photos according to preset growth periods and preset growth grades to obtain labeled training images; constructing an initial classification neural network model, inputting the labeled training images into the initial classification neural network model for training, and obtaining the trained classification neural network model.

3. The method for nitrogen fertilizer application to rice according to claim 1, wherein The rice nitrogen fertilizer application method further comprises: obtaining a nitrogen content of a to-be-purchased fertilizer product; According to the nitrogen content and the rice nitrogen fertilizer application amount, the purchase bag number of the to-be-purchased fertilizer product is calculated.

4. A nitrogen fertilization system for rice, characterized by, The rice nitrogen fertilizer application system comprises: A data acquisition module is configured to acquire a to-be-detected rice photo of a to-be-fertilized area. A prediction module is configured to input the to-be-detected rice photo into a trained classification neural network model to obtain a predicted growth period and a predicted growth grade of the to-be-detected rice photo. A first nitrogen application interval calculation module is configured to, when the predicted growth grade is less than a preset grade, calculate a first nitrogen application interval according to the predicted growth period, specifically as follows: A historical SPAD value and a historical nitrogen fertilizer application amount of a historical rice photo are acquired. A fitting equation corresponding to each preset growth period is constructed according to the historical SPAD value and the historical nitrogen fertilizer application amount. When the predicted growth grade is less than the preset grade, an interval calculation is performed according to the predicted growth period and the corresponding fitting equation to obtain the first nitrogen application interval, specifically as follows: A first SPAD value interval corresponding to the predicted growth period is obtained by performing data matching according to the predicted growth period and a preset growth period corresponding SPAD table. A second SPAD value interval is obtained by performing SPAD value calculation according to the first SPAD value interval and a preset growth period corresponding preset minimum SPAD value. The first nitrogen application interval is obtained by performing interval calculation according to the second SPAD value interval and the corresponding fitting equation. A second nitrogen application interval calculation module is configured to calculate a second nitrogen application interval according to the predicted growth period and preset periods corresponding weights, specifically as follows: A yield range of the rice under an optimal nitrogen application level and a nitrogen application amount of a previous period of the predicted growth period are acquired. A total nitrogen application amount is calculated according to the yield range of the rice under the optimal nitrogen application level. A predicted nitrogen application interval of the predicted growth period is calculated according to the total nitrogen application amount and the preset periods corresponding weights. The second nitrogen application interval is obtained by performing nitrogen application amount calculation according to the predicted nitrogen application interval and the nitrogen application amount of the previous period. A rice nitrogen fertilizer application amount calculation module is configured to calculate a rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo according to the first nitrogen application interval and the second nitrogen application interval, specifically as follows: A minimum value of the first nitrogen application interval and a minimum value of the second nitrogen application interval are acquired, and a maximum value of the two minimum values is taken as a minimum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo. A maximum value of the first nitrogen application interval and a maximum value of the second nitrogen application interval are acquired, and a minimum value of the two maximum values is taken as a maximum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo. The rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo is obtained according to the minimum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo and the maximum value of the rice nitrogen fertilizer application amount corresponding to the to-be-detected rice photo.

5. A rice nitrogen fertilization apparatus characterized by, The method comprises at least one control processor and a memory connected to the at least one control processor in communication; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to perform the method for nitrogen fertilizer fertilization of rice according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that: The computer readable storage medium stores computer executable instructions for causing a computer to perform the method for nitrogen fertilizer fertilization of rice according to any one of claims 1 to 3.

Images