CN109492619B

CN109492619B - Variable pesticide application method and system integrating remote sensing, model and algorithm

Info

Publication number: CN109492619B
Application number: CN201811541772.0A
Authority: CN
Inventors: 杨润勃; 李海峰
Original assignee: Beijing Jinghe Big Data Technology Co ltd
Current assignee: Beijing Jinghe Big Data Technology Co ltd
Priority date: 2018-12-17
Filing date: 2018-12-17
Publication date: 2021-07-30
Anticipated expiration: 2038-12-17
Also published as: CN109492619A

Abstract

The application relates to the technical field of data processing, in particular to a variable pesticide application method and system integrating remote sensing, a model and an algorithm, wherein the method comprises the following steps: determining a mapping relation between geographic information of a growing place where a crop is located and actual geographic information of the growing place; dividing the remote sensing image of the growing area where the crop is located into at least one grid according to the mapping relation, wherein each grid corresponds to a working area with a preset area in the growing area; and configuring pesticide dosage matched with the potential disease type and the disease severity of the potential disease of the crop for the operation area with the disease incidence probability of the potential disease larger than a preset value based on the condition of the potential disease of the crop predicted by aiming at each operation area so as to carry out variable pesticide application on the crop at the growing place. The embodiment of the application provides a variable pesticide application method and system integrating remote sensing, a model and an algorithm, the using dosage of pesticides can be controlled, the utilization rate of the pesticides is improved, and further the yield of farmlands is improved.

Description

Variable pesticide application method and system integrating remote sensing, model and algorithm

Technical Field

The application relates to the technical field of data processing, in particular to a variable pesticide application method and system integrating remote sensing, models and algorithms.

Background

At present, in agricultural production, the yield of farmlands is greatly influenced by the problems of plant diseases and insect pests and weeds, taking wheat as an example, the yield and the sowing area of domestic wheat account for 1/3 of grains all over the world, and the wheat is one of three grains all over the world; the weed damage is one of the most threatening factors for the wheat field yield, and can cause the loss of nearly 50 hundred million kilograms in China every year. In agricultural production, pesticides may be used to kill insects, kill bacteria or eliminate weeds in order to ensure and promote the growth of plants and crops.

When the pesticide is used, the pesticide is mostly sprayed or smeared manually, and the dosage of the pesticide cannot be effectively controlled in the process of spraying or smearing the pesticide, so that the abuse of the pesticide is easily caused.

Disclosure of Invention

In view of this, the embodiment of the present application provides a variable pesticide application method and system integrating remote sensing, a model and an algorithm, which can control the dosage of pesticides, improve the utilization rate of pesticides, and further improve the yield of farmlands.

In a first aspect, an embodiment of the present application provides a variable pesticide application method integrating remote sensing, a model and an algorithm, where the method includes:

determining a mapping relation between the geographic information of the growing place where the crop is located and the actual geographic information of the growing place where the crop is located;

dividing a growing area where the crop is located into at least one grid according to the mapping relation, wherein each grid corresponds to a working area with a preset area in the growing area;

and configuring pesticide dosage matched with the potential disease type and the disease severity of the potential disease of the crop for the operation area with the disease incidence probability of the potential disease larger than a preset value based on the condition that the crop is infected with the potential disease and predicted for each operation area so as to carry out variable pesticide application on the crop at the growing place.

In one possible embodiment, the condition of the crop infected with the underlying disease is predicted according to the following steps:

predicting potential diseases of the crops in the current growing season based on at least one of the predicted growing environment characteristics of the crops in the current growing season, the historical morbidity condition of the growing place, the cultivated land condition of the growing place in the current growing season, the capability of the crops in the current growing season to resist the frequent diseases of the growing place, the types of the crops in the growing place in the previous season and the types of the diseases of the crops in the growing place in the previous season;

after the crops are sowed, predicting the condition that the crops in each operation area are infected with the potential diseases within a preset time length in the future based on overground growth environment information, underground growth environment information, soil information, remote sensing indexes, actual growth information of the crops and growth stage information of the crops in each operation area.

In one possible embodiment, the predicting the crop infected with the potential disease in each working area for a preset length of time in the future comprises:

predicting the probability and degree of infection of the crops with the potential diseases in each working area within a preset time length in the future, the yield reduction rate of the crops and the optimal application time.

In one possible embodiment, predicting the growth environment characteristics of the crop in the current growing season comprises:

acquiring historical growing season weather characteristics of the growing area and the growing season weather characteristics in a preset time period before crop sowing;

predicting the growth environment characteristics of the growing area of the crop in the current growing season according to the historical growing season weather characteristics and the growing season weather characteristics in a preset time period before the crop is sown;

predicting overground growth environment information of the crop according to the following steps:

determining a meteorological index representing overground growth environment information of the crops according to the temperature, the humidity and the rainfall of the growing area;

predicting the underground growth environment information of the crop according to the following steps:

determining a daily water stress index and a daily nutrient stress index of the crops for representing the underground growth environment information according to the root depth, the water demand information and the nutrient demand information of the crops in each operation area;

predicting growth stage information for the crop according to the following steps:

and predicting the growth stage information of the crops according to the weather information, the soil information, the farming mode, the previous crop type, the daily water stress index, the daily nutrition stress index, the daily water demand, the daily water supply of the growth environment, the daily nutrition demand, the daily nutrition supply of the growth environment, the current growth season crop planting density, the crop type of the crops and the daily disease stress index of the growing area.

In one possible embodiment, the index of telemetry is determined by:

obtaining a remote sensing image of the crop;

extracting growth characteristic information of the crops on each operation area from the remote sensing image;

and determining the remote sensing index of the crop based on the growth characteristic information.

In one possible embodiment, the configuration of the amount of pesticide matching the type of underlying disease and the severity of the underlying disease of the crop comprises:

and determining the pesticide dosage configured for each working area according to the predicted condition of the potential diseases infected by the crops, the current pesticide price for treating the potential diseases and the predicted crop price.

In a possible embodiment, the method further comprises:

setting a pesticide spraying route and a spraying strategy of the operating equipment for the crops according to the pesticide dosage determined on each operating area and the operating equipment used on each operating area;

and the operation equipment sprays pesticide to crops on the operation area according to the pesticide spraying route, the pesticide dosage matched with the operation area and the spraying strategy of the operation equipment.

In one possible embodiment, after the variable application to the crop at the growing area, the method further comprises:

acquiring actual growth information of crops and/or actual disease information of the crops within a preset time range;

and configuring supplementary pesticide dosage matched with the actual growth information and/or the actual disease information according to the actual growth information and/or the actual disease information, and performing supplementary variable pesticide application on the crops in the growing area under the condition that the supplementary pesticide dosage is larger than zero.

In a possible embodiment, the method further comprises:

acquiring actual disease information, remote sensing indexes and soil information of the crops in each operation area;

judging whether the crops in each operation area have pesticide application value or not according to the actual disease information, the remote sensing index and the soil information;

estimating a disease severity index for the crop in each of the work areas where the crop has application value in the work area;

and determining the pesticide dosage configured for each operation area according to the estimated disease degree index of the crops in each operation area, the current pesticide price of the pesticide for treating the current actual diseases and the predicted crop price.

In a second aspect, an embodiment of the present application further provides a variable pesticide application system integrating remote sensing, models and algorithms, where the system includes: the device comprises a dividing module and a variable pesticide applying module; wherein the content of the first and second substances,

the dividing module is used for dividing the remote sensing image of the growing area where the crop is located into at least one grid, wherein each grid corresponds to a working area with a preset area in the growing area;

the variable pesticide application module is used for configuring pesticide dosage matched with the potential disease type of the crops and the infection severity of the potential diseases for the operation areas with the disease incidence probability of the potential diseases larger than a preset value based on the condition that the crops are infected with the potential diseases and predicted for each operation area so as to carry out variable pesticide application on the crops at the growing places.

In a third aspect, an embodiment of the present application further provides an electronic device, including: the network side device comprises a processor, a memory and a bus, wherein the memory stores machine readable instructions executable by the processor, when the network side device runs, the processor and the memory are communicated through the bus, and the machine readable instructions are executed by the processor to execute the steps of the method.

In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the above method.

By adopting the scheme, the growing area where the crops are located is divided into at least one grid, each grid corresponds to a working area with a preset area in the growing area, the condition that the crops in each working area are infected with the potential diseases is predicted, and the pesticide dosage matched with the potential disease types of the crops and the infection severity of the potential diseases can be configured, so that the pesticide application dosage is effectively controlled, the pesticide utilization rate is improved, and the farmland yield is further improved.

In order to make the aforementioned objects, features and advantages of the embodiments of the present application more comprehensible, embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

FIG. 1 is a basic flow chart of a variable pesticide application method integrating remote sensing, modeling and algorithm provided by the embodiment of the application;

FIG. 2 is a specific flowchart of a variable pesticide application method integrating remote sensing, modeling and algorithm provided by the second embodiment of the present application

FIG. 3 shows a structural diagram of a variable pesticide application system integrating remote sensing, modeling and algorithm provided by the third embodiment of the application;

fig. 4 shows a block diagram of an electronic device according to a fourth embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The following detailed description of the embodiments of the present application is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

In the embodiment of the application, the growing area where the crop is located can be divided into at least one grid, each grid corresponds to the operation area with the preset area in the growing area, the condition that the crop in each operation area is infected with a potential disease is predicted, and the pesticide dosage matched with the potential disease type of the crop and the infection severity of the potential disease can be configured, so that the pesticide application dosage is effectively controlled, the pesticide utilization rate is improved, and the farmland yield is further improved.

The following examples will illustrate the variable dosing process in detail.

Example one

The basic flow of the variable pesticide application method integrating remote sensing, models and algorithms is shown in fig. 1, and the method comprises the following steps:

s101: and determining the mapping relation between the geographic information of the growing place where the crop is located and the actual geographic information of the growing place.

In one possible embodiment, the map of the crop growth area used in the prediction of the condition of the crop infection with the potential disease in the present application, the map of the crop growth area used in the weather prediction model in the growing season in the present application, the map of the crop growth area used in the first crop disease prediction model, the map of the crop growth area used in the second crop disease prediction model, the map of the crop growth area used in the chemical quantity estimation model, or the remote sensing image of the crop growth area may be correlated with the actual geographic information of the growth area. Specifically, for example, the longitude and latitude of the place where the crop is grown are obtained by mapping.

S102: and dividing the growing area of the crop into at least one grid according to the mapping relation, wherein each grid corresponds to a working area with a preset area in the growing area.

In one possible embodiment, for example, after obtaining the longitude and latitude of the growing area of the crop, the growing area of the crop can be divided into at least one grid according to the longitude and latitude of the growing area of the crop. Specifically, the growing area of the crop can be divided into at least one grid of 1 meter by 1 meter according to the longitude and latitude of the growing area of the crop. Of course, the number of 1 m by 1 m is only an example for easy understanding, and the size and shape of the mesh are not specifically limited in the present application, and at least one of the meshes may be a common shape such as a square, a rectangle, a circle, an ellipse, or a trapezoid, and the size of the mesh may be larger or smaller.

For the growing area of the crop, the working area with the preset area in the growing area is divided by using the grids, the range of the working area can be conveniently adjusted, the growing area of the crop is divided into different working areas for analyzing the condition of potential diseases infected by the crop, and the analysis is more accurate. For example, by dividing the growing area of the crop by using a smaller grid, for example, a grid of 1 meter by 1 meter, a working area with a smaller range and higher precision can be obtained, so that the condition of potential diseases of the crop infected in each working area in the growing area of the crop can be more accurately analyzed.

And the operation area with the preset area in the growing area is divided by the grids, so that the application range is wider, and the grids with different sizes can be selected to divide the growth of the crops according to the difference of the sizes of the growing areas of the crops and the difference of pesticide application equipment. For example, in the case of airplane operation, the growth of the crop can be divided into grids with a larger range for rough analysis; under the condition of using unmanned aerial vehicle operation, the grid of dividing the growth of the crop into 1 meter by 1 meter can be more accurately analyzed.

S103: and configuring pesticide dosage matched with the potential disease type and the disease severity of the potential disease of the crop for the operation area with the disease incidence probability of the potential disease larger than a preset value based on the condition that the crop is infected with the potential disease and predicted for each operation area so as to carry out variable pesticide application on the crop at the growing place.

and predicting potential diseases of the crops in the current growing season based on at least one of the predicted growing environment characteristics of the crops in the current growing season, the historical morbidity condition of the growing place, the cultivated land condition of the growing place in the current growing season, the capability of the crops in the current growing season to resist the frequent diseases of the growing place, the types of the crops in the growing place in the previous season and the types of the diseases of the crops in the growing place in the previous season.

In one possible embodiment, predicting the growth environment characteristics of the crop in the current growing season according to the following steps, comprising:

and acquiring historical growing season weather characteristics of the growing area and the growing season weather characteristics in a preset time period before crop sowing.

And predicting the growth environment characteristics of the growing place of the crop in the current growing season according to the historical growing season weather characteristics and the growing season weather characteristics in a preset time period before the crop is sown.

Specifically, the method for predicting the growth environment characteristics of the crops in the current growing season comprises the following steps:

step 1, obtaining the weather characteristics of the historical growing season of the growing place where the crop is located.

Here, the weather characteristics of the historical growing season of the place where the crop is growing may include: el nino-southern oscillation type and degree, growing season weather signature index. The growing season weather characteristic index may include a growing season weather characteristic index for each year over the past 18 years. According to the method, the weather characteristic index of the growing season every year is collected within a longer time range, for example, within 18 years, so that richer and comprehensive historical growing season weather characteristic samples can be obtained, the analysis on the historical growing season weather characteristics is more accurate, and the method is favorable for predicting the weather condition of the current crop growing season by using the historical growing season weather characteristics.

The historical growing season weather characteristics may include: the type and extent of the early-southern oscillation, the effective integrated temperature actually utilized, the maximum temperature of the environment, the minimum temperature of the environment, and the correlation of the season weather signature index with the current early-southern oscillation (ENSO) event.

Wherein, the effective accumulated temperature actually used is the effective accumulated temperature of the growing area of the block multiplied by the available accumulated temperature coefficient; the effective accumulated temperature can be the sum of effective temperatures of crops in a certain growth period or all growth periods, namely the sum of the difference between the daily average temperature and biological zero temperature of the crops in a certain period. Some crops have an upper limit temperature, namely when the temperature rises to a certain limit, the temperature rises again, and the development speed of the crops cannot be obviously accelerated and even can play a role in inhibiting. Within the suitable temperature range (between the upper and lower temperature limits), the crop development speed is linear with temperature.

Hercino-southern oscillations (ENSO), also known as Hercino/Ranina-southern billows, are a type of quasi-cyclic climate that occurs across the Pacific ocean near the equator, approximately once every 5 years. Southern surges refer to sea surface temperatures in the equatorial region of the east pacific (warm in the event of el nino, cool in the event of raney) and variations in sea surface air pressure in the equatorial region of the west pacific. These two variations are interrelated, namely: the warm ocean phase of the east pacific, early nino, is accompanied by high sea surface air pressure of the west pacific; the cold-down phase of the east pacific, raney, is accompanied by a low sea pressure of the west pacific. The el nino events can be divided into 3 basic types: a warm pool, el nino event, occurring in the pacific in the tropics, with weaker intensity; extreme early-nile events, occurring in the tropical east pacific, which are very intense; cold-tongue ernonol events occur in the tropical middle, east pacific, with moderate intensity.

And 2, acquiring the weather characteristics of the growing season in a preset time period before crop sowing.

The growing season weather characteristics may include: early nino-southern oscillation type and extent, effective integrated temperature actually utilized, maximum temperature of the environment, minimum temperature of the environment, and correlation of the growing season weather signature index with the current ENSO event. Wherein the erlinuo-southern oscillation type and extent may include the type and extent of erlinuo-southern oscillation predicted 3 months prior to crop seeding.

And 3, predicting the growth environment characteristics of the growing area of the crop in the current growing season according to the historical growing season weather characteristics and the growing season weather characteristics in a preset time period before the crop is sown.

In the specific implementation process, the historical growing season weather characteristics and the growing season weather characteristics in the preset time period before the crop is sown can be used as parameters for predicting the growing environment information of the growing place where the current growing season crop is located by using a growing season weather prediction model, and after analysis and calculation, the growing environment characteristics for predicting the growing place where the current growing season crop is located can be obtained. The growing environment characteristics of the growing place where the current growing season crop is located may include: the rainfall total amount of the current growing season, the effective accumulated temperature of the current growing season, the highest temperature of the current growing season environment and the lowest temperature of the current growing season environment.

Since the historical growing season weather characteristics of a longer time period, such as the 18-year historical growing season weather characteristics, are acquired and combined with the growing season weather characteristics in the preset time period before the crop is sown, the obtained weather characteristics are used as parameters for predicting the growing environment information of the growing place where the current growing season crop is located, and the prediction of the growing environment characteristics of the growing place where the current growing season crop is located is more accurate.

and determining a meteorological index representing overground growth environment information of the crops according to the temperature, the humidity and the rainfall of the growing area.

Specifically, the overground growth environment information may include: and weather condition information of the growing place, such as ambient temperature (highest daily temperature, lowest daily temperature), ambient humidity (highest daily humidity, lowest daily humidity), rainfall (hourly or daily rainfall), and the like.

and determining the daily water stress index and the daily nutrition stress index of the crops for representing the underground growth environment information according to the root depth, the water demand information and the nutrient demand information of the crops in each operation area.

The underground growth environment information includes: at least one of the root depth of the crop and the soil condition of the growing area may be monitored by a soil tester, for example, the soil depth, soil structure, soil texture, soil moisture change, soil nutrients, pH of the soil surface layer (0-30 cm of the soil layer), pH of the soil bottom layer (30-60 cm of the soil layer), whether there is a soil layer for limiting root growth, the proportion of soil stones, and the depth of the crop root. For example, when planting field corn, a soil tester can be arranged every 30 centimeters in the direction of the soil depth, and the soil environment 1.6 meters underground can be monitored by the soil tester.

The multiple information can be used as input variables of a soil moisture model, and the daily water stress index for representing the condition of the moisture required by the crop and the daily nutrition stress index for representing the condition of the nutrient required by the crop can be obtained after analysis and calculation. Wherein the daily water stress index can be obtained from the daily water demand of crops and the daily water supply of the growing environment of the crops.

Specifically, the above information is analyzed and calculated to determine the growth stage of the crop, such as the androgenesis stage, the black layer stage, etc. The weather information of the crop can comprise the daily accumulated temperature of the current growing season, the daily accumulated temperature of the growing seasons of the past 18 years, the highest daily temperature of the current growing season, the lowest daily temperature of the current growing season, the daily rainfall of the current growing season, the daily humidity of the current growing season, the daily wind speed of the current growing season and the like; the soil information of the crops can comprise soil type, soil texture, soil structure, soil surface layer pH value (0-30 cm), soil bottom layer pH value (30-60 cm), whether the soil has a soil layer for limiting root growth or not and the proportion of soil stones; the cultivation mode of the crops can comprise no-turning, shallow turning (0-40 cm) and deep ploughing (40-70 cm).

According to the method, historical growth season weather characteristics and growth season weather characteristics in a preset time period before crop sowing are comprehensively considered, overground growth environment information and underground growth environment information are comprehensively considered, parameters with abundant types and large quantities are comprehensively considered for calculation, influence factors different in the aspects of crop growth are comprehensively considered, the more the considered factors are, and the more accurate the obtained growth stage estimation result is.

In one possible embodiment, the index of telemetry is determined by:

obtaining a remote sensing image of the crop;

Specifically, the growth characteristic information of the crop may be determined by, for example, obtaining a satellite remote sensing image or a near-field remote sensing image with a height of 15 meters from the crop or with higher precision, extracting growth characteristic information of the crop from the remote sensing image of the crop, such as information of the height of the crop, the color of the crop, the area of a leaf of the crop, and determining a growth index of the crop according to the growth characteristic information of the crop. The growth index of the crop can be a physical quantity for representing the growth of the crop, and the growth characteristic information of the crop can be monitored through the growth index of the crop. For example, the growth index of a crop may be between 0 and 1, indicating that the crop is best grown when the growth index is 1; when the growth index of the crop is 0, the crop is worst in growth; when the growth vigor index of the crop is between 0 and 0.35, the crop is extremely poor in growth vigor (the crop is dead or nearly dead); when the growth vigor index of the crop is between 0.35 and 0.5, indicating the growth vigor difference of the crop; when the growth vigor index of the crop is between 0.5 and 0.75, the growth vigor of the crop is moderate; when the growth vigor index of the crop is between 0.75 and 1, the growth vigor of the crop is moderate or optimal.

According to the method, the remote sensing image of the crop is utilized, the growth characteristic information of the crop in each operation area is extracted from the remote sensing image, the growth characteristic information of the crop can be comprehensively, conveniently and visually acquired, in combination with the step S102, the growing area where the crop is located is divided into at least one grid, the growth characteristic information of the crop is respectively acquired for each operation area with a preset area according to the accuracy of the grid, and the growth of the crop can be more thoroughly evaluated. Especially when the growing area of the crops is large, a large amount of manpower research resources can be saved, the influence of artificial subjective factors is avoided, and the obtained growth evaluation conclusion is more objective, detailed and accurate.

In some embodiments, the obtained field survey item table can be used to confirm the predicted growth stage and growth vigor of the crop, for example, the presence or absence of lodging of corn can be confirmed by a picture of the growth stage of corn.

In one possible embodiment, the predicting the condition of the crop infected with the potential disease in each working area within a preset length of time in the future comprises:

In particular implementations, the potential diseases that may occur during the growth of the current growing season crop can be predicted by a first crop disease prediction model. Specifically, at least one of the information of the growth environment, the historical disease condition of the growing area, the cultivated land condition of the growing area in the current growing season, the crop type of the growing area in the previous season, and the type and degree of the disease occurring in the crop in the growing area in the previous season may be used as a parameter for predicting the potential disease which may occur in the crop, and the first crop disease prediction model may be used to predict the potential disease of the crop in the current growing season according to the at least one information.

The first crop disease prediction model is trained according to historical data of the past year, and is trained by taking at least one of information of growth environment information of a target growth season, historical morbidity condition of a growth place, cultivated land condition of the growth place of the target growth season, crop type of the growth place of the previous season of the target growth season and disease type and degree of crop occurrence of the growth place of the previous season of the target growth season as input features of the first crop disease prediction model, and diseases of the crop occurrence of the target growth season as output features of the first crop disease prediction model.

In the above step, the historical disease condition of the growing area is the historical disease occurrence condition of the growing area, and may include: the type of disease of crops, the disease resistance of crops, the type of disease of crops in the previous season and the severity of infection. For example, the diseases of crops frequently occurring in a growing area in the past 5 years, and the disease resistance (susceptibility, moderate, high, etc.) of crops in the current growing season to the frequently occurring diseases of the growing area, which can be obtained from a questionnaire filled by a grower.

The arable land condition of the growing land in the current growing season may include: the type of plowing and the depth of plowing layer, for example, the type of plowing in the current growing season of the growing land is no tillage or strip tillage, and the depth of plowing layer is deep tillage or shallow tillage.

Here, not only the prediction of the underlying disease but also the probability of onset of the underlying disease, the severity of the underlying disease, and the possible period of onset can be predicted. For example, in a farm in Jilin, Jilin province, corn is moderate in disease resistance when planted, and the incidence of northern leaf blight in 2016 in this growing area is 86%. Moreover, the incidence of northern leaf blight in 2016 of the same growing area as described above has been verified by production practice.

Here, the condition that the crop is infected with the potential disease within a preset time length in the future can be predicted through the second crop disease prediction model after the crop is sown. Specifically, the growth condition of the crop and the growth environment of the crop can be used as parameters for predicting the condition of the crop infected with the potential disease, and the occurrence probability, the infection severity and the crop yield reduction of the potential disease on the growing land can be predicted 7 to 15 days in the future. Similarly, the second crop disease prediction model is obtained by training the growth conditions of the crops and the growth environments of the crops in the target generation season of the past year as inputs and the conditions of the crop infected with diseases as outputs.

In one embodiment, the crop yield reduction may be predicted by:

step 1, predicting the crop yield of the crop in the growing area under the condition of no disease infection, for example, predicting the crop yield of the crop under the condition of no disease infection according to the seeding rate of the crop, the growing environment in the growing process of the crop and other factors.

And 2, predicting the yield reduction percentage caused by the potential disease infection in the future preset time span according to the predicted condition of the potential disease infection of the crop and the disease resistance of the crop in different growth stages.

For example, according to the growth condition and growth environment of the corn, if the area of damaged leaves of the corn is predicted to be larger than 50%, and the yield is reduced by more than 22%, the fact that the corn needs to be sprayed can be determined, and corresponding pesticide and pesticide dosage can be allocated for the corn.

The predicted pesticide amount may be affected by different crop species and different growing areas, that is, the predicted pesticide amount may be different for different crop species, and the predicted pesticide amount may also be different for the same crop species when planted in different growing areas. Thus, in variable application, a crop refers to the same crop and the growing field refers to the same growing field in which the crop is grown.

Specifically, for example, the growing area can be divided into larger grids according to the condition that the crops are infected with the potential diseases, for example, the growing area is divided into operation areas with low pesticide dosage, medium pesticide dosage and high pesticide dosage, and the economic optimized pesticide dosage of the three areas can be calculated according to the current pesticide and crop selling price. Alternatively, the growing area may be divided into grids of 1 meter by 1 meter or higher precision by using a satellite remote sensing image, and the amount of pesticide matched with the condition that the crop on the working area is infected with the potential disease may be allocated to the working area corresponding to each spatial grid.

In a specific embodiment, a pesticide spraying route can be set for each crop according to the condition that the crop on each operation area is infected with the potential diseases, so that the operation equipment sprays pesticide for the crop on the operation area according to the pesticide spraying route and the pesticide dosage matched with the operation area. Here, the work implement may be an airplane, an unmanned aerial vehicle, or a ground pesticide spraying apparatus.

In one possible embodiment, when an aircraft is selected as the working equipment based on the condition of the crop being infected with the potential disease, the growing area is divided into larger grids, for example, the growing area is divided into working areas with low pesticide dosage, medium pesticide dosage and high pesticide dosage, and the economic optimized pesticide dosage for the three areas is calculated based on the current pesticide, the manual input, the predicted yield of the crop and the selling price of the crop. At this time, because the cost of airplane operation is high, it is usually necessary to ensure that the profit of the crop calculated according to the predicted yield of the crop and the selling price of the crop needs to be twice as high as the cost of the current pesticide and manpower input.

In a possible embodiment, when the unmanned aerial vehicle is selected as the working equipment according to the condition that the crop is infected with the potential diseases, the type of the unmanned aerial vehicle used needs to be judged first.

If an unmanned aerial vehicle which can not be manually controlled and has a route and a speed which can not be set in advance is used, the growing area can be divided into larger grids, for example, the growing area is divided into operation areas with low pesticide dosage, medium pesticide dosage and high pesticide dosage, and the economic optimized pesticide dosage of the three areas can be calculated according to the current pesticide, manual input, the predicted yield of crops and the selling price of the crops.

If an automatic drone is used, the route and speed of which can be set in advance, the following steps can be used:

step 1: the locus of the crop may be divided into at least one grid of 1 meter by 1 meter or more.

Step 2: and predicting the condition that the crops in each working area are infected with the potential diseases within a preset time length in the future according to the overground growth environment information, the underground growth environment information, the soil information, the remote sensing index, the actual growth information and the growth stage information of the crops in each working area.

And step 3: distinguishing and marking the operation areas needing pesticide application and the operation areas not needing pesticide application according to the condition that the crops in each operation area are predicted to be infected with the potential diseases within the preset time length in the future; wherein, the operation area without pesticide application includes but is not limited to: seedling lacking areas, lodging crop areas, damaged crop areas, dead crop areas, non-application value crop areas or disease-free crop areas.

And 4, step 4: and (3) allocating pesticide dosage matched with the type of the potential diseases and the infection severity of the potential diseases of the crops to the operation area needing pesticide application.

Inputting the current pesticide price, the predicted crop price, the potential disease type of the crop and the infection severity of the potential disease into a pesticide quantity estimation model to obtain the optimal pesticide dosage of each operation area. In order to simplify the calculation, the severity of the potential disease is generally divided into 10 levels, and the optimal pesticide dosage for each working area is calculated by using a dosage estimation model according to the severity of the potential disease. Here, each working area may be divided into a grid of 1 meter by 1 meter or more in accuracy, and thus, the amount of agricultural chemicals applied to the crop growing area can be accurately controlled.

And 5: and inputting the actual geographic information of the growing area, each operation area and the pesticide dosage corresponding to each operation area into an unmanned aerial vehicle control program, and automatically applying the pesticide to each operation area of the crop growing area through the unmanned aerial vehicle.

In one possible embodiment, when the ground pesticide spraying device is selected as the working equipment according to the condition that the crop is infected with the potential diseases, the following steps are adopted:

step 1: acquiring actual geographic information of a growing place where the crop is located, and dividing the growing place where the crop is located into at least one grid with the accuracy of 1 meter multiplied by 1 meter or higher according to the actual geographic information.

For example, the longitude and latitude of the place where the crop is located are obtained, and the place where the crop is located is divided into at least one grid of 1 meter by 1 meter or higher according to the longitude and latitude of the place where the crop is located.

Here, soil properties of the sampled soil may be acquired, and soil information of each work area may be estimated based on the soil properties.

In particular, soil attributes may be obtained using a soil scanner and/or soil samples may be collected.

When the soil scanner is used for analyzing soil, the soil scanner can be used for acquiring soil information of soil in the primary operation area along one direction at preset intervals.

Specifically, for example, a product MSP3 from veritech corporation or a Ground Penetrating Radar (Ground Penetrating Radar) may be used to scan a predetermined cubic coating, for example, a soil layer of 0-30 cm or 0-90 cm, along one direction and at predetermined intervals, for example, at intervals of 5 m, 8 m or 10 m, to obtain soil information such as conductivity of each scanned point.

When the soil sample is collected to analyze the soil, a plurality of soil samples are collected according to the preset sample quantity and the collection depth, and then soil information such as organic matters, pH value, nitrogen content, phosphorus content, cation exchange capacity, soil sand content, clay content and/or soil structure in the soil samples is detected for the plurality of soil samples.

After soil attributes are acquired by using a soil scanner and/or collecting soil samples, soil information of each working area can be obtained by using a spatial interpolation method.

And classifying the soil of each working area according to the soil information of each working area to obtain the soil type of each working area, and making a soil type graph of each working area.

And 5: and applying pesticide to each operation area of the crop growing area through ground pesticide spraying equipment according to the actual geographic information of the growing area, each operation area and the pesticide dosage corresponding to each operation area.

In a possible embodiment, after the variable application of the pesticide to the crop at the growing place, the method further comprises the following steps:

In one possible embodiment, the method further comprises:

Specifically, in order to simplify the calculation process, the estimated disease degree index of the crop is generally expressed by 1 to 9 natural numbers.

By the variable pesticide applying method integrating remote sensing, models and algorithms, the types and the dosages of the pesticides required by crops can be determined by predicting the conditions of potential diseases, so that the using dosages of the pesticides are effectively controlled, and the utilization rate of the pesticides is improved. In addition, for concrete and convenience of description, a method for describing pesticides is provided in the embodiment of the application, and based on the same design concept, the method can also be used in application scenes of variable application of herbicides, variable fertilization and the like.

Example two

The specific flow of the variable pesticide application method integrating remote sensing, model and algorithm provided by the second embodiment of the application is shown in fig. 2, and may include the following steps:

s201, acquiring historical growing season weather characteristics of the location of the crop by a variable pesticide application system, and predicting growing environment information of the growing area of the crop in the current growing season based on the historical growing season weather characteristics by using a growing area weather prediction model;

here, the historical growing season weather characteristics may include: el nino-southern oscillation type and degree, growing season weather signature index. The growing environment information of the growing place where the current growing season crop is located may include: the total rainfall amount in the current growing season, the effective accumulated temperature utilized in the current growing season, the highest temperature of the environment in the current growing season and the lowest temperature of the environment in the current growing season.

S202, acquiring the disease occurrence condition of historical crops in a growing place by a variable pesticide application system, and predicting the type of potential diseases of the crops in the current growing season and the condition of infecting the potential diseases by utilizing a crop disease prediction model based on the growth environment information of the growing place where the crops in the previous growing season are located and the disease occurrence condition of the historical crops;

here, the historical crop morbidity conditions may include: the type of disease of crops, the disease resistance of crops, the type of disease of crops in the previous season and the severity of infection. Conditions that affect this underlying disease include: the incidence of the underlying disease, the severity of the underlying disease and the likely stage of the disease.

S203, after the crops are sowed, the variable pesticide application system acquires growth environment information, remote sensing images of the crops and field investigation results in the crop growth process;

here, the growth environment information during the growth of the crop may include: overground growth environment information and underground growth environment information. The overground growth environment information may include: weather condition information of the growing place, such as ambient temperature (highest daily temperature, lowest daily temperature), ambient humidity (highest daily humidity, lowest daily humidity), rainfall (hourly or daily rainfall), and the like; the subsurface growth environment information may include: the soil depth, soil structure, soil texture, soil moisture change, soil nutrients, soil surface pH (0-30 cm), soil bottom pH (30-60 cm), whether soil layers for limiting root growth exist, the proportion of soil stones, the depth of crop roots and the like.

The field survey results may include pictures, descriptions, etc. of the growth stage of the crop, such as the presence or absence of lodging of the crop.

S204, determining the meteorological index of the crop growing area by the variable pesticide application system based on the overground growing environment information in the crop growing process by using a meteorological model;

here, the weather index of the growing place of the crop may represent the weather condition of the growing place where the crop is located.

S205, determining the water and nutrient conditions needed by the crops by the variable pesticide application system based on the underground growth environment information in the crop growth process by using a soil water and nutrient model;

here, the water condition required for a crop can be characterized by a daily water stress index, and the nutrient condition required for a crop can be characterized by a daily nutrient stress index.

S206, the variable pesticide application system obtains the growth index of the crop based on the remote sensing image of the crop by using the remote sensing model;

here, the growth vigor index of the crop may characterize the growth vigor of the crop.

S207, determining the growth stage of the crops by the variable pesticide application system based on the growth environment information in the growth process of the crops and the conditions of water and nutrients required by the crops by using a crop growth model;

here, the growth stage of the crop may be characterized by a crop growth stage index, for example, the crop growth stage index may include 10 growth stages, which may be represented by V_NWherein N is a positive integer less than or equal to 10.

And S208, predicting and predicting the condition that the potential diseases are infected by the crops within the future preset time length by the variable pesticide application system by using a crop disease prediction model based on the meteorological index of the crop growing area, the water and nutrient conditions required by the crops, the growth index of the crops, the growth stage of the crops and the field investigation result of the actual area.

Here, the conditions in which the crop is infected with the underlying disease may include: the type of the potential diseases, the incidence probability of the potential diseases, the severity of the potential diseases, the yield reduction of crops and the like.

For example, if the variable delivery system predicts that the corn has a high spot disease probability of 86%, the severity of the infection is low, the area of the damaged leaves of the corn is less than 20%, and the yield is reduced by less than 3%, then the corn can be determined not to need to be sprayed.

S209, the variable pesticide application system predicts the pesticide application period and pesticide dosage of the crops according to the predicted condition that the crops are infected with the potential diseases, divides the growing area into at least one operation area with a preset area, and allocates pesticide dosage matched with the potential disease types and the disease severity of the potential diseases for the operation area.

By the variable pesticide application method integrating remote sensing, models and algorithms, the types and the pesticide dosages of pesticides can be determined by predicting the conditions of potential diseases, so that the pesticide dosage is effectively controlled, and the pesticide utilization rate is improved.

EXAMPLE III

The basic structure of the variable pesticide application system 30 integrating remote sensing, model and algorithm provided by the third embodiment of the present application is shown in fig. 2, and includes: an acquisition module 31, a growth environment prediction module 32, a disease prediction module 33 and a pesticide dosage configuration module 34; wherein the content of the first and second substances,

the obtaining module 31 is configured to obtain weather characteristics of a historical growing season of a growing area where the crop is located;

the growing environment predicting module 32 is configured to predict growing environment information of a growing area where the crop is located in a current growing season according to the historical growing season weather characteristics;

the disease prediction module 33 is configured to predict a potential disease of the crop in the current growing season based on the growing environment information and the historical crop disease conditions of the growing area; and the system is used for predicting the condition that the crops are infected with the potential diseases within a preset time length in the future based on the growth stage information of the crops and the growth environment of the crops after the crops are sown;

the pesticide dosage configuration module 34 is configured to configure a pesticide dosage for treating the potential disease for the crop at the growing area according to the predicted condition that the crop is infected with the potential disease.

Through the variable pesticide application system 30 integrating remote sensing, models and algorithms, the types and the dosages of the pesticides required by crops can be determined through prediction of potential disease conditions, so that the using dosages of the pesticides are effectively controlled, and the utilization rate of the pesticides is improved.

The disease prediction module 33 is specifically configured to predict the potential disease of the crop in the current growing season according to the following steps:

predicting potential diseases of the crops in the current growing season based on at least one of the information of the growing environment, the historical morbidity condition of the growing area, the cultivated land condition of the growing area in the current growing season, the types of the crops in the growing area in the previous season, and the types of diseases occurring in the crops in the growing area in the previous season.

Further, the disease prediction module 33 is specifically configured to determine the growth stage information of the crop according to the following steps:

acquiring growth environment information of the crops, growth characteristic information of the crops and growth stage information of the crops;

and determining the growth stage information of the crop based on the growth environment information of the crop, the growth characteristic information of the crop and the growth stage information of the crop.

Further, the growth environment information includes: overground growth environment information and underground growth environment information;

the overground growth environment information comprises: weather condition information of the growing area;

the underground growth environment information includes: at least one of the root depth of the crop and the soil condition of the growing area.

The disease prediction module 33 is specifically configured to obtain growth characteristic information of the crop according to the following steps:

obtaining a remote sensing image of the crop;

extracting growth characteristic information of the crops from the remote sensing image;

and determining the growth vigor index of the crop based on the growth vigor characteristic information.

Further, the conditions of the crop infected with the underlying disease include: the type of the underlying disease, the incidence of the underlying disease, and the severity of the underlying disease;

the pesticide dosage configuration module 34 is specifically configured to configure the crops of the growing area with pesticide dosage for treating the potential diseases according to the situations of the crops infected with the potential diseases predicted by the following steps:

dividing the growing area into at least one working area with a preset area according to the predicted condition that the crops are infected with the potential diseases;

and aiming at each operation area, if the incidence probability of the potential diseases of the crops on the operation area is greater than a preset value, allocating pesticide dosage matched with the type of the potential diseases of the crops and the infection severity of the potential diseases for the operation area.

Optionally, the variable pesticide application system 30 integrating remote sensing, modeling and algorithm further includes:

and the route setting module 35 is used for setting a pesticide spraying route for the crops according to the condition that the crops on the operation area are infected with the potential diseases, so that the operation equipment sprays pesticides for the crops on the operation area according to the pesticide spraying route and the pesticide dosage matched with the operation area.

The variable pesticide application system 30 integrating remote sensing, models and algorithms can control the pesticide dosage and plan a reasonable operation route for the pesticide spraying operation equipment according to the pesticide dosage, so that the operation equipment can economically spray the pesticide for crops.

EXAMPLE III

Fig. 4 shows a basic structure of an electronic device 40 provided in the third embodiment of the present application, which includes: a processor 41, a memory 42, and a bus 43;

the memory 42 stores machine-readable instructions executable by the processor 41, the processor 41 and the memory 42 communicate via the bus 43 when the network-side device is running, and the machine-readable instructions, when executed by the processor 41, perform the following:

acquiring the weather characteristics of the historical growing season of the growing place where the crop is;

according to the weather characteristics of the historical growing seasons, predicting the growing environment information of the growing area of the crop in the current growing season;

predicting potential diseases of the crops in the current growing season based on the growing environment information and the historical crop morbidity conditions of the growing places;

after the crops are sown, predicting the condition that the crops are infected with the potential diseases within a preset time length in the future based on the growth stage information of the crops and the growth environment of the crops;

configuring a crop of said growing area with an amount of an agricultural agent to treat said underlying disease based on the predicted infestation of said crop by said underlying disease.

In a specific implementation, the processor 41 performs the process of predicting the potential diseases of the crop in the current growing season based on the growing environment information and the historical crop disease conditions of the growing area, including:

In a specific implementation, in the processing performed by the processor 41, the growth stage information of the crop is determined according to the following steps:

In a specific implementation, in the processing performed by the processor 41, the growth environment information includes: overground growth environment information and underground growth environment information;

In a specific implementation, the obtaining of the growth characteristic information of the crop performed by the processor 41 includes:

obtaining a remote sensing image of the crop;

In a specific implementation, the processor 41 performs the processing, wherein the condition that the crop is infected with the potential disease includes: the type of the underlying disease, the incidence of the underlying disease, and the severity of the underlying disease; configuring a crop of said growing area with an amount of an agricultural agent to treat said underlying disease based on said predicted infestation of said crop by said underlying disease, comprising:

In a specific implementation, in the processing performed by the processor 41, the method further includes:

and setting a pesticide spraying route for the crops according to the condition that the crops on the operation area are infected with the potential diseases, so that the operation equipment sprays the pesticides for the crops on the operation area according to the pesticide spraying route and the pesticide dosage matched with the operation area.

Example four

The fourth embodiment of the present application provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the variable pesticide application method integrating remote sensing, models and algorithms are executed.

Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk and the like, and when a computer program on the storage medium is run, the variable pesticide application method integrating remote sensing, a model and an algorithm can be executed, so that the problem of pesticide abuse caused by the fact that the use dosage of the pesticide cannot be effectively controlled at present is solved, the use dosage of the pesticide is controlled, and the utilization rate of the pesticide is improved.

The computer program product of the variable pesticide application method integrating remote sensing, models and algorithms provided by the embodiment of the application comprises a computer readable storage medium storing program codes, instructions included in the program codes can be used for executing the method in the previous method embodiment, specific implementation can be referred to the method embodiment, and details are not repeated herein.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the system described above may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A variable pesticide application method integrating remote sensing, models and algorithms is characterized by comprising the following steps:

determining the grid size of the grid based on the size of the crop generation area, the pesticide application equipment and the condition that the crop is infected with potential diseases;

dividing a growing area of the crop into at least one grid according to the mapping relation and the grid size, wherein each grid corresponds to a working area with a preset area in the growing area;

configuring pesticide dosage matched with the potential disease type and the disease severity of the potential disease of the crop for the operation area with the disease incidence probability of the potential disease larger than a preset value based on the condition that the crop is infected with the potential disease and predicted for each operation area so as to carry out variable pesticide application on the crop at the growing place;

predicting the condition of the crop infected with the potential disease according to the following steps:

the method further comprises the following steps:

setting a pesticide spraying route and a spraying strategy of the operating equipment for the crops according to the pesticide dosage determined on each operating area and whether the route of the operating equipment used on each operating area is set in advance; wherein the working area is divided based on the working implement used;

2. The method of claim 1, wherein the condition of the crop infected with the underlying disease is predicted according to the following steps:

3. The method of claim 2, wherein predicting the crop infection with the underlying disease in each work area for a predetermined length of time in the future comprises:

4. The method of claim 2, wherein predicting the growth environment characteristic of the crop in the current growing season comprises:

5. The method of claim 2, wherein the telemetry index is determined by:

obtaining a remote sensing image of the crop;

6. The method of claim 1, wherein configuring the amount of pesticide to match the type of underlying disease and the severity of the underlying disease of the crop comprises:

7. The method of claim 1, further comprising, after variable application to the crop at the locus, the steps of:

8. The method of claim 1, further comprising:

9. A variable pesticide application system integrating remote sensing, models and algorithms is characterized by comprising the following components: the device comprises a dividing module and a variable pesticide applying module; wherein the content of the first and second substances,

the dividing module is used for determining the grid size of the grid based on the size of the crop generation area, the pesticide application equipment and the condition that the crop is infected with the potential diseases; dividing the remote sensing image of the growing area where the crop is located into at least one grid according to the mapping relation and the grid size, wherein each grid corresponds to a working area with a preset area in the growing area;

the variable pesticide application module is used for configuring pesticide dosage matched with the potential disease type of the crops and the infection severity of the potential diseases for the operation areas with the disease incidence probability of the potential diseases larger than a preset value based on the condition that the crops are infected with the potential diseases and predicted for each operation area so as to carry out variable pesticide application on the crops at the growing places;

the variable dosing module is further to:

the variable pesticide application system integrating remote sensing, models and algorithms further comprises a route setting module, and is used for: