CN109492619A

CN109492619A - Integrate the variable pesticide grown method and system of remote sensing, model, algorithm

Info

Publication number: CN109492619A
Application number: CN201811541772.0A
Authority: CN
Inventors: 杨润勃; 李海峰
Original assignee: Beijing He Da Da Data Technology Co Ltd
Current assignee: Beijing He Da Da Data Technology Co Ltd
Priority date: 2018-12-17
Filing date: 2018-12-17
Publication date: 2019-03-19
Anticipated expiration: 2038-12-17
Also published as: CN109492619B

Abstract

This application involves technical field of data processing, more particularly to a kind of variable pesticide grown method and system for integrating remote sensing, model, algorithm, this method comprises: the mapping relations where determining crop between the geography information of growing location and the actual geographic information of the growing location；According to mapping relations, the remote sensing images of growing location where crop are divided at least one grid, wherein each grid corresponds to the operating area of preset area in growing location；The case where infecting potential disease based on the crop for the prediction of each operating area, it is greater than the operating area of preset value for the incidence rate of potential disease, the pesticide dosage that configuration matches with the potential disease type of crop and the severity of catching an illness of potential disease carries out variable farm chemical applying with the crop to place growing location.The embodiment of the present application provides a kind of variable pesticide grown method and system for integrating remote sensing, model, algorithm, can control the dosage of pesticide, improves the utilization rate of pesticide, and then improve farmland yield.

Description

Integrate the variable pesticide grown method and system of remote sensing, model, algorithm

Technical field

This application involves technical field of data processing more particularly to it is a kind of integrate remote sensing, model, algorithm variable apply Pesticide method and system.

Background technique

Currently, in agricultural production, pest and disease damage and weed problem can largely effect on the yield in farmland, by taking wheat as an example, state The yield and sown area of interior wheat account for the 1/3 of whole world grain, are one of three generalized grain of the world；And crop smothering be wheatland yield most Have one of deterrent, nearly 5,000,000,000 kilograms of the loss in China can be caused every year.In agricultural production, for ensure, promote plant and The growth of crop can be used pesticide and come desinsection, sterilization or eliminate weeds.

Pesticide be mostly when in use by manually being sprayed or being smeared, spray or smear pesticide during, can not The effectively dosage of control pesticide, be easy to cause indiscriminate use of pesticide.

Summary of the invention

In view of this, the embodiment of the present application provide it is a kind of integrate remote sensing, model, algorithm variable pesticide grown method and System can control the dosage of pesticide, improve the utilization rate of pesticide, and then improve farmland yield.

In a first aspect, the embodiment of the present application provides a kind of variable pesticide grown side for integrating remote sensing, model, algorithm Method, which comprises

Mapping where the geography information of growing location where determining crop and crop between the actual geographic information of growing location Relationship；

According to the mapping relations, growing location where crop is divided at least one grid, wherein each grid is corresponding The operating area of preset area in the growing location；

The case where infecting potential disease based on the crop for each operating area prediction, is the potential disease The incidence rate of disease is greater than the operating area of preset value, configuration and the potential disease type of the crop and catching an illness for potential disease The pesticide dosage that severity matches carries out variable farm chemical applying with the crop to place growing location.

In a kind of possible embodiment, the case where predicting the crop infection potential disease according to following steps:

The growing environment feature of crop described in current Growing season based on prediction, the growing location history onset state, The arable land situation of growing location described in current Growing season, current Growing season crop resist this place and often send out disease ability, a upper Ji Suoshu At least one of the disease type that the crops of growing location described in the crop type of growing location and a upper season occur information, Predict the potential disease of crops described in current Growing season；

After crop sowing, based on the aerial environmental information of crop described in each operating area, underground Growing environment information, soil information, Indices, crop actual growing situation information and growth phase information, when predicting following default Between the case where crop described in each operating area infects the potential disease in length.

In a kind of possible embodiment, crop described in each operating area in the following predetermined time period of the prediction The case where infecting the potential disease, comprising:

Predict crop described in each operating area in the following predetermined time period infect the potential disease probability and Degree, the underproduction rate of the crop and best spraying time.

In a kind of possible embodiment, according to following steps predict current Growing season described in crop growing environment it is special Sign, comprising:

Obtain the growing location historical growth season weather characteristics and the crop sow before growth in preset time period Season weather characteristics；

Growing season weather before being sowed according to the historical growth season weather characteristics and the crop in preset time period is special Sign, the growing environment feature of growing location where predicting crop described in current Growing season；

The aerial environmental information of the crop is predicted according to following steps:

According to the temperature, humidity and rainfall of the growing location, the aerial environmental information for characterizing the crop is determined Meteorological Index；

The underground growing environment information of the crop is predicted according to following steps:

Deep, water demand information and nutrient demand information according to the root of crop described in each operating area, determination are used for Characterize the day water stress index and day Nutrient Stress index of the crop of the underground growing environment information；

The growth phase information of the crop is predicted according to following steps:

According to the Weather information of the growing location, soil information, tillage system reform, upper season agrotype, the day water side of body Compel that index, day Nutrient Stress index, day water requirement, growing environment day water supply, amount of nutrients, growing environment day be for nutrient day by day Amount, current Growing season density of crop, the agrotype of the crop and day disease coerce index, predict the life of the crop Long session information.

In a kind of possible embodiment, the Indices are determined by following steps:

Obtain the remote sensing images of the crop；

From the growing way characteristic information for extracting the crop on each operating area in the remote sensing images；

Based on the growing way characteristic information, the Indices of the crop are determined.

In a kind of possible embodiment, the configuration and the potential disease type of the crop and catching an illness for potential disease The pesticide dosage that severity matches, comprising:

The agriculture of the case where infecting potential disease according to the crop of each operating area prediction, the treatment potential disease Current pesticide price, the crop price of prediction of medicine are determined as the pesticide dosage of each operating area configuration.

In a kind of possible embodiment, the method also includes:

It is described according to the operating apparatus used on the pesticide dosage and each operating area determined on each operating area The sprinkling strategy of route and the operating apparatus that crop setting is sprayed insecticide；

The operating apparatus is according to the route sprayed insecticide, the pesticide dosage to match with the operating area and institute The sprinkling strategy of operating apparatus is stated, is sprayed insecticide for the crop on the operating area.

In a kind of possible embodiment, after carrying out variable farm chemical applying to the crop of place growing location, further includes:

It obtains in preset time range, the actual growing situation information of crop and/or the practical defect information of crop；

According to the actual growing situation information and/or practical defect information, configuration and the actual growing situation information and/or reality The supplement pesticide dosage that defect information matches, in the case where affiliated supplement pesticide dosage is greater than zero, to place growing location The crop carry out additional variable application.

In a kind of possible embodiment, the method also includes:

Obtain practical defect information, Indices and the soil information of crop described in each operating area；

According to the practical defect information, the Indices and the soil information, each operating area is judged Described in crop whether have application be worth；

In the case that the crop described in the operating area has application value, institute in each operating area is estimated State the disease degree index of crop；

According to the disease degree index of crop described in each operating area of estimation, the pesticide of the currently practical disease for the treatment of Current pesticide price, the crop price of prediction, be determined as the pesticide dosage of each operating area configuration.

Second aspect, the embodiment of the present application also provides a kind of variable pesticide grown systems for integrating remote sensing, model, algorithm System, the system comprises: division module and variable farm chemical applying module；Wherein,

The division module, for the remote sensing images of growing location where crop to be divided at least one grid, wherein every A grid corresponds to the operating area of preset area in the growing location；

The variable farm chemical applying module, for infecting potential disease based on the crop for each operating area prediction The situation of disease is that the incidence rate of the potential disease is greater than the operating area of preset value, potential disease of the configuration with the crop The pesticide dosage that sick type and the severity of catching an illness of potential disease match carries out variable with the crop to place growing location and applies Medicine.

The third aspect, the embodiment of the present application also provides a kind of electronic equipment, comprising: processor, memory and bus, institute State memory and be stored with the executable machine readable instructions of the processor, when network side equipment operation, the processor with By bus communication between the memory, the step of the above method is executed when the machine readable instructions are executed by the processor Suddenly.

Fourth aspect, the embodiment of the present application also provides a kind of computer readable storage medium, the computer-readable storages The step of being stored with computer program on medium, the above method executed when which is run by processor.

Using the above scheme, growing location where crop is divided at least one grid, wherein described in each grid is corresponding The operating area of preset area in growing location predicts crop infection potential disease situation in each operating area, can be with The pesticide dosage that configuration matches with the severity of catching an illness with the potential disease type of the crop and potential disease, to have The control Pesticide use dosage of effect improves the utilization rate of pesticide, and then improves farmland yield.

To enable the above objects, features, and advantages of the embodiment of the present application to be clearer and more comprehensible, below in conjunction with embodiment, and Cooperate appended attached drawing, elaborates.

Detailed description of the invention

Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in the embodiment attached Figure is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore is not construed as pair The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this A little attached drawings obtain other relevant attached drawings.

Fig. 1 shows the variable pesticide grown method for integrating remote sensing, model, algorithm provided by the embodiment of the present application one Basic flow chart；

Fig. 2 shows the variable pesticide grown methods for integrating remote sensing, model, algorithm provided by the embodiment of the present application two Specific flow chart

Fig. 3 shows the variable pesticide grown system for integrating remote sensing, model, algorithm provided by the embodiment of the present application three Structure chart；

Fig. 4 shows the structure chart of electronic equipment provided by the embodiment of the present application four.

Specific embodiment

To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only It is only some embodiments of the present application, instead of all the embodiments.It is not to the detailed description of embodiments herein below It is intended to limit claimed scope of the present application, but is merely representative of the selected embodiment of the application.Reality based on the application Apply example, those skilled in the art's every other embodiment obtained without making creative work belongs to The range of the application protection.

It, can be by the way that growing location where crop be divided at least one grid, wherein each net in the embodiment of the present application Lattice correspond to the operating area of preset area in the growing location, carry out to crop infection potential disease situation in each operating area Prediction can configure the pesticide agent to match with the severity of catching an illness with the potential disease type of the crop and potential disease Amount improves the utilization rate of pesticide, and then improve farmland yield to effectively control Pesticide use dosage.

Following embodiments will elaborate to variable farm chemical applying process.

Embodiment one

The embodiment of the present application one provide it is a kind of integrate remote sensing, model, algorithm variable pesticide grown method it is basic Process is as shown in Figure 1, comprising the following steps:

S101: the mapping where determining crop between the geography information of growing location and the actual geographic information of the growing location Relationship.

In a kind of possible embodiment, it can be predicted by the way that the application is carried out the case where crop infects potential disease The map on plant growth ground used in the map on the plant growth ground of Shi Suoyong, the application Growing season weather prediction model, first Plant growth ground used in the map on the ground of plant growth used in crop diseases prediction model, the second crop diseases prediction model The remote sensing images on the map on plant growth ground or plant growth ground used in map, explosive charge model, with the growing location Actual geographic information is mapped.Specifically for example, the longitude and latitude of growing location where obtaining crop by mapping.

S102: according to the mapping relations, growing location where crop is divided at least one grid, wherein each net Lattice correspond to the operating area of preset area in the growing location.

It, can basis in a kind of possible embodiment, such as after the longitude and latitude of growing location where acquisition crop Growing location where crop is divided at least one grid by the longitude and latitude of growing location where crop.It specifically, can basis Growing location where crop is divided at least one the 1 meter grid for multiplying 1 meter by the longitude and latitude of growing location where crop.Certainly , 1 meter herein multiplies 1 meter only to facilitate understanding the example enumerated, and the application does not do specifically the size and shape of grid Restriction, at least one above-mentioned grid can be square, rectangle, circle, ellipse or the common shape such as trapezoidal, grid Size can be bigger or smaller.

It, can using the operating area of preset area in growing location described in grid dividing for growing location where crop Easily to adjust the range of operating area, it is potential that growing location where crop is divided into different operating area progress crop infection The analysis of the case where disease, and analyze more accurate.For example, using lesser grid, such as 1 meter multiplies 1 meter of grid, divides crop Place growing location, available range is smaller, the higher operating area of precision, thus to each operation in growing location where crop The crop in region infects the more accurate analysis of the case where potential disease progress.

Also, using the operating area of preset area in growing location described in grid dividing, the scope of application is wider, can basis Plant growth size difference, the difference of application equipment, select different size of grid to divide growth where crop. For example, growth where crop can be divided into the biggish grid of range in the case where using aircraft operation and do rough analysis； In the case where using unmanned machine operation, the grid that growth where crop is divided into the grid that 1 meter multiplies 1 meter can be done more accurate Analysis.

S103: the case where infecting potential disease based on the crop for each operating area prediction, is described The incidence rate of potential disease is greater than the operating area of preset value, the potential disease type and potential disease of configuration and the crop The pesticide dosage that matches of severity of catching an illness, variable farm chemical applying is carried out with the crop to place growing location.

In a kind of possible embodiment, the case where crop infects potential disease is predicted according to following steps:

The growing environment feature of crop described in current Growing season based on prediction, the growing location history onset state, The arable land situation of growing location described in current Growing season, current Growing season crop resist this place and often send out disease ability, a upper Ji Suoshu At least one of the disease type that the crops of growing location described in the crop type of growing location and a upper season occur information, Predict the potential disease of crops described in current Growing season.

Obtain the growing location historical growth season weather characteristics and the crop sow before growth in preset time period Season weather characteristics.

Growing season weather before being sowed according to the historical growth season weather characteristics and the crop in preset time period is special Sign, the growing environment feature of growing location where predicting crop described in current Growing season.

Specifically, predicting the growing environment feature of crop described in current Growing season, comprising:

Step 1, the historical growth season weather characteristics of growing location where obtaining crop.

Here, the historical growth season weather characteristics of growing location where crop may include: EI Nino-south oscillation class Type and degree, Growing season weather characteristics index.Growing season weather characteristics index may include Growing season annual in 18 years in the past Weather characteristics index.The application acquired within the scope of the long period, such as in 18 years, annual Growing season weather characteristics index, It is hereby achieved that richer, comprehensive historical growth season weather characteristics sample, more to the analyses of historical growth season weather characteristics Precisely, be conducive to usage history Growing season weather characteristics to predict the weather condition of current Growing Season of Crops.

Historical growth season weather characteristics may include: EI Nino-south type of oscillation and degree, actually utilize it is effective Accumulated temperature, the maximum temperature of environment, the minimum temperature of environment and Growing season weather characteristics index and current EI Nino-south The correlation of (abbreviation ENSO) event of oscillation.

Wherein, the effective accumulated temperature actually utilized=this block growing location effective accumulated temperature × using accumulated temperature coefficient；Effectively product Temperature can be the summation of crop effective temperature within some breeding time or whole breeding times, i.e., crop is in certain a period of time Nei Ping The summation of the difference of equal temperature and biological zero point.There is also ceiling temperatures for some crops, i.e., when temperature rise to certain limit with Afterwards, temperature increases again, and the rate of development of crop can not be made obviously to accelerate, or even can play inhibiting effect.In preference temperature scope Interior (between upper and lower limit temperature), Crop development speed and temperature line relationship.

EI Nino-south oscillation (abbreviation ENSO) is referred to as EI Nino/La Nina-southern oscillation, is to occur Across a kind of Pacific climate type paracycle of equator, occurred every about 5 years primary.Southern oscillation refers to eastern peace The sea of foreign equatorial zone sea surface temperature when Ramsey numbers (warm when EI Nino event, turn cold) and Equatorial Western Pacific region The variation of air pressure on face.Both variations connect each other, it may be assumed that the warm foreign stage of Eastern Pacific, i.e. EI Nino, along with The high sea level pressure of Western Pacific；Eastern Pacific turns cold the stage, i.e. La Nina, along with the low sea level pressure of Western Pacific. EI Nino event can be divided into 3 kinds of fundamental types: occur in tropical Middle Pacific, the weaker warm pool EI Nino thing of intensity Part；Occur in tropical Eastern Pacific, the very strong extreme EI Nino event of intensity；- Eastern Pacific occurs in the torrid zone, intensity is suitable In cold tongue EI Nino event.

Step 2, the Growing season weather characteristics before the crop is sowed in preset time period are obtained.

Growing season weather characteristics may include: EI Nino-south type of oscillation and degree, the effective product actually utilized Temperature, the maximum temperature of environment, the minimum temperature of environment and Growing season weather characteristics index are related to current ENSO event Property.Wherein, EI Nino-south type of oscillation and degree may include the current Growing season strategic point that crop sows prediction in first 3 months That Nino-south type of oscillation and degree.

Step 3, the Growing season before being sowed according to the historical growth season weather characteristics and the crop in preset time period Weather characteristics, the growing environment feature of growing location where predicting crop described in current Growing season.

In the specific implementation process, it can use Growing season weather prediction model, by above-mentioned historical growth season weather characteristics Growing season weather characteristics before being sowed with the crop in preset time period, as growing location where predicting current Growing season crop Growing environment information parameter, by analysis and after calculating, growing location where the current Growing season crop of available prediction Growing environment feature.The growing environment feature of growing location where current Growing season crop may include: the rainfall of current Growing season The minimum temperature of total amount, the effective accumulated temperature of current Growing season, the maximum temperature of current Growing season environment and current Growing season environment.

Due to obtaining the historical growth season weather characteristics of longer period, such as 18 years historical growth season weather spies Sign, and the Growing season weather characteristics before the crop is sowed in preset time period are combined, make as the current Growing season of prediction The parameter of the growing environment information of growing location where object, to the pre- of the growing environment feature of growing location where current Growing season crop It is more accurate to survey.

According to the temperature, humidity and rainfall of the growing location, the aerial environmental information for characterizing the crop is determined Meteorological Index.

Specifically, aerial environmental information may include: the weather conditions information of the growing location, such as environment temperature (max. daily temperature, Daily minimum temperature), ambient humidity (day highest humidity, day minimum humidity), rainfall are (every hour or daily Rainfall) etc..

Deep, water demand information and nutrient demand information according to the root of crop described in each operating area, determination are used for Characterize the day water stress index and day Nutrient Stress index of the crop of the underground growing environment information.

The underground growing environment information include: that the root of the crop is deep and the soil regime of the growing location at least One kind specifically such as can use soil testing instrument and monitor the depth of soil of the crop, soil texture, the soil texture, the soil water Divide variation, soil nutrient, upper soll layer (0-30 centimetres of soil horizon) pH value, soil bottom (30-60 centimetres of soil horizon) soda acid Degree, whether the soil layer of restricted root growth, the ratio of soil stone and the depth of root of the crop etc..For example, in plantation field corn When, underground can be monitored using soil testing instrument on the direction of depth of soil every 30 centimetres of settings, one soil testing instrument 1.6 meters of soil environment.

Above-mentioned much information can be used as the input variable of soil moisture model, by analysis, calculate after, available table The day water of moisture situation needed for levying the crop coerces index, and the day Nutrient Stress of nutrient situation needed for characterizing the crop refers to Number.Wherein, day water stress index can by crop day water requirement, crop growth environment day water supply obtain.

Specifically, above- mentioned information are analyzed and is calculated, can determine the growth phase of the crop, such as tasseling stage, black Layer phase etc..Wherein, the Weather information of the crop may include current Growing season day accumulated temperature, past 18 years Growing season day product Highest degree/day, the minimum degree/day of current Growing season of warm, current Growing season, the daily rainfall of current Growing season work as previous existence Long season day humidity, current Growing season day wind speed etc.；The soil information of the crop may include soil types, the soil texture, Soil texture, upper soll layer pH value (0-30 centimetres), soil bottom pH value (30-60 centimetres), whether restricted, soil The ratio of the soil layer of growth, soil stone；The tillage system reform of the crop may include not ploughing, shallowly ploughing (0~40 centimetre), is deep It ploughs (40~70 centimetres).

The application has comprehensively considered the growth in historical growth season weather characteristics and the preceding preset time period of crop sowing Season weather characteristics, and comprehensively considered aerial environmental information and underground growing environment information, combined that type is abundant, number It measures huge parameter to be calculated, comprehensively considers the influence factor different to the every aspect of plant growth, the factor of consideration is got over More, obtained growth phase estimation results are more accurate.

Obtain the remote sensing images of the crop；

Specifically, the growing way characteristic information of above-mentioned crop can determine in the following manner, specifically such as, available distance 15 meters of height of crop or the higher satellite remote sensing images of precision or near-earth remote sensing images, extracting from the remote sensing images of the crop should The growing way characteristic information of crop, such as plant height, crop color and crop leaf area information, and according to the growing way of the crop Characteristic information determines the growing way index of the crop.The growing way index of crop can be the physical quantity of characterization crop growing state, pass through The growing way index of crop, can be with the growing way characteristic information of monitoring crop.For example, the growing way index of crop can between 0-1, when When growing way index is 1, show that the crop growing state is best；When the growing way index of crop is 0, show that the crop growing state is worst；When When the growing way index of crop is between 0-0.35, show that the crop growing state is very poor (crop dead or close dead)；When crop When growing way index is between 0.35-0.5, show that the crop growing state is poor；When the growing way index of crop is between 0.5-0.75, table The bright crop growing state is medium；When the growing way index of crop is between 0.75-1, show that the crop growing state is medium excellent or optimal.

The application utilizes the remote sensing images of crop, from extracting the crop on each operating area in the remote sensing images Growing way characteristic information will can be made with the intuitive growing way characteristic information for obtaining crop in conjunction in step S102 comprehensively, conveniently Growing location where object is divided at least one grid, divides according to the precision size of grid the operating area of each preset area Not Huo Qu crop growing way characteristic information, more the growing way of crop can be assessed in detail.Especially in the life of crop When long ground area is larger, the application can save a large amount of manpowers investigation resources, and avoid the influence of artificial subjective factor, obtain The growing way assessment result arrived is more objective, detailed and accurate.

It, can also be using the field investigation project table on the spot obtained, to the plant growth rank of prediction in some embodiments Section is confirmed that whether there is or not lodging for example, confirming corn by the picture in corn growth stage with growing way.

In a kind of possible embodiment, make described in each operating area in the following predetermined time period of the prediction Object infects the case where potential disease, comprising:

It in specific implementation, can be by the first crop diseases prediction model to can in current Growing season process of crop growth The potential disease that can occur is predicted.It specifically such as, can be by growing environment information, the history onset state, current of growing location Growing location described in the agrotype of growing location described in the arable land situation of growing location described in Growing season, a upper season and a upper season At least one of the disease type of crop generation and degree information, the ginseng as the potential disease that prediction crop may occur Amount, and utilize the potential of the first crop diseases prediction model crop according to above-mentioned at least one current Growing season of information prediction Disease.

Above-mentioned first crop diseases prediction model be according to former years historical data train come, for former years target growth Season, by the arable land of growing location described in the growing environment information in target growth season, the history onset state of growing location, target growth season Growing location described in the agrotype of growing location described in situation, a upper season in target growth season and a upper season in target growth season The crop disease type and at least one of degree input feature vector of the information as the first crop diseases prediction model that occur, The disease that crop described in target growth season is occurred obtains first as the output feature of the first crop diseases prediction model, training Crop diseases prediction model.

In above-mentioned steps, the history crop onset state of growing location is the history disease generation feelings for growing ground crop Condition, may include: disease type that recurrent crop diseases type, the premunition of crop, upper season crop have occurred and It catches an illness severity.For example, the crop on a certain growing location in the past 5 years for the application form acquisition that farmer fills in can be passed through Recurrent crop diseases, and current Growing season crop to the growing location often send out disease premunition (it is susceptible, in Deng, it is high) etc..

The arable land situation of growing location described in current Growing season may include: plough type and topsoil depth etc., for example, should The type of ploughing of the current Growing season in growing location is that no-tillage or item is ploughed, and topsoil depth is deep ploughing or shallow plowing etc..

Here, potential disease can not only be predicted, can also predicts incidence rate, the potential disease of potential disease Catch an illness and severity and may occur the phase.During for example, Jilin Province, farm, Jilin, when maize planting, the premunition of corn is The incidence rate for the leaf blight of corn occur Deng, growing location in 2016 is 86%.Also, above-mentioned 2016 above-mentioned growing location The incidence rate for the leaf blight of corn occur, which has passed through production practices, confirms its correctness.

Here, it can be somebody's turn to do by the second crop diseases prediction model in the following predetermined time period after crop sowing Crop infects the case where potential disease and is predicted.It specifically such as, can be by the life of the upgrowth situation of the crop and the crop Long environment is predicted to dive on following 7 to 15 days growing locations as the parameter for predicting the case where crop infects the potential disease In the probability of happening of disease, severity of catching an illness and crop cut yield.Similarly, the second crop diseases prediction model be pass through by Former years target generates the upgrowth situation of the crop in season and the growing environment of the crop as input, by the feelings of crop infectious disease Condition is obtained as output training.

In one embodiment, above-mentioned crop cut yield can be predicted by following steps:

Step 1, the crop of this block growing location is predicted in the crop yield being uninfected by under disease event, for example, according to crop The factors such as the growing environment in application rate and process of crop growth predict that the crop makees produce being uninfected by under disease event Amount.

Step 2, disease-resistant in the case where infecting potential disease according to the crop of prediction and crop different stages of growth Power predicts the interior percentage of underproduction due to infecting potential disease of the following predetermined time period.

For example, according to the upgrowth situation and growing environment of corn, it is predicted that the area of the aggrieved leaf of corn is greater than 50%, subtracts It produces and is greater than 22%, then can determine that corn needs to spray, and configure corresponding pesticide and pesticide dosage for corn.

In a kind of possible embodiment, the configuration and the potential disease type of the crop and the dye of potential disease The pesticide dosage that sick severity matches, comprising:

It should be noted that different crop species and different growing locations can there are shadows to the pesticide dosage of prediction It rings, i.e., the pesticide dosage of different crop species predictions may be different, and identical crop species are pre- when different growing locations are planted The pesticide dosage of survey may also can be different.Therefore, in variable farm chemical applying, crop refers to same crop, and growing location refers to The same growing location of the crop-planting.

Specifically for example, being divided into the growing location biggish the case where above-mentioned potential disease being infected according to the crop Grid, such as the growing location is divided into the operating area of low pesticide dosage, middle peasant's pharmaceutical quantities and high pesticide dosage, and according to working as Preceding pesticide and crop price calculate the economical optimization pesticide dosage in this three areas.Alternatively, can use satellite remote sensing images, The growing location is divided into 1 meter of grid for multiplying 1 meter or higher precision, and for the corresponding operating area configuration of each space lattice with The pesticide dosage that the case where crop infects the potential disease on the operating area matches.

In a specific embodiment, the case where potential disease can also being infected according to crop on each operating area, For the crop, the route sprayed insecticide is set, so as to operating apparatus according to the route sprayed insecticide and with the operating area phase The pesticide dosage matched is sprayed insecticide for the crop on the operating area.Here, operating apparatus can for aircraft, unmanned plane or Person ground pesticide dispersal equipment.

In a kind of possible embodiment, when according to the crop infect above-mentioned potential disease the case where, select aircraft for When operating apparatus, which is divided into biggish grid, such as the growing location is divided into low pesticide dosage, middle pesticide agent The operating area of amount and high pesticide dosage, and calculated according to current pesticide, artificial investment, the forecast production of crop and crop price The economical optimization pesticide dosage in this three areas out.At this point, due to the higher cost of aircraft operation, it usually needs guarantee, according to The income for the crop that the forecast production and crop price of crop are calculated, the cost for needing to reach current pesticide and manually putting into Twice.

In a kind of possible embodiment, when according to the crop infect above-mentioned potential disease the case where, select unmanned plane When for operating apparatus, the type for the unmanned plane for needing first to judge to use.

It, can be by the growing location if using the unmanned plane that the manual control of route and speed cannot be arranged in advance It is divided into biggish grid, such as the growing location is divided into the operation of low pesticide dosage, middle peasant's pharmaceutical quantities and high pesticide dosage Region, and it is economical optimal according to current pesticide, artificial investment, the forecast production of crop and crop price to calculate this three areas Change pesticide dosage.

If using the automatic unmanned plane that route and speed can be arranged in advance, following step progress can be used:

Step 1: growing location where crop can be divided at least one 1 meter multiplied by 1 meter or the grid of higher precision.

Step 2: according to the aerial environmental information of crop described in each operating area, underground growing environment information, Soil information, Indices, crop actual growing situation information and growth phase information predict each work in the following predetermined time period Crop described in industry region infects the case where potential disease.

Step 3: the potential disease is infected according to crop described in each operating area in the following predetermined time period of prediction The situation of disease, is distinguished and label needs the operating area being administered and do not need the operating area of application；Wherein, application is not needed Operating area, including but not limited to: crop belts, dead crop belts have been damaged, without application valence by the area that is short of seedling, laid crop area It is worth crop belts or disease-free crop belts.

Step 4: the operating area that needs are administered, configuration and the potential disease type of the crop and the dye of potential disease The pesticide dosage that sick severity matches.

Catching an illness for current pesticide price, the crop price of prediction, the potential disease type of crop and potential disease is serious Degree inputs explosive charge model, obtains the optimal pesticide dosage of each operating area.It is calculated to simplify, it usually will be potential The severity of catching an illness of disease is divided into 10 ranks, according to the severity level of catching an illness of potential disease, using explosive charge model, Calculate the optimal pesticide dosage of each operating area.Here, each operating area can be divided into 1 meter and multiply 1 meter or more high-precision The grid of degree, therefore, the pesticide dosage that can to plant growth apply are accurately controlled.

Step 5: by the actual geographic information of the growing location, the corresponding pesticide in each operating area and each operating area Dosage is input in unmanned aerial vehicle (UAV) control program, passes through unmanned plane from trend plant growth each operating area applying pesticides.

In a kind of possible embodiment, when according to the crop infect above-mentioned potential disease the case where, select ground agriculture When medicine spray appliance is operating apparatus, carried out using following step:

Step 1: the actual geographic information of growing location where obtaining crop, according to above-mentioned actual geographic information, by crop institute At least one 1 meter is divided into multiplied by 1 meter or the grid of higher precision in growing location.

For example, the longitude and latitude of growing location where obtaining crop, and according to the longitude and latitude of growing location where crop, Growing location where crop is divided at least one 1 meter multiplied by 1 meter or the grid of higher precision.

Here it is possible to obtain the soil attribute of sampling soil, the soil of each operating area is estimated according to above-mentioned soil attribute Earth information.

Specifically, soil scanner can be used and obtain soil attribute, and/or acquisition soil sample obtains soil attribute.

When analyzing using soil scanner soil, soil scanner can be used in one direction, every default Distance obtains the soil information of one-stop operation regional soil.

Specifically it is, for example, possible to use the MSP3 product of Veristech company or Ground Penetrating Radar (Ground Penetrating Radar), in one direction, every pre-determined distance, such as every 5 meters, 8 meters or 10 meters, default cube of scanning Coating, such as 0-30 centimetres or 0-90 centimetres of soil layer of scanning obtains the soil informations such as the conductivity of each scanning element.

When acquisition soil sample analyzes soil, according to preset sample size and sampling depth, several are acquired Soil sample detects the organic matter in soil sample, pH value, nitrogen content, phosphorus and contains later to several above-mentioned soil samples The soil informations such as amount, cation exchange capacity (CEC), soil content, amount containing clay and/or soil texture.

After obtaining soil attribute using soil scanner and/or acquisition soil sample, the side of space interpolation can be used Method obtains the soil information of each operating area.

And classified according to soil of the soil information of each operating area to each operating area, obtain each operation The soil types in region, and make the soil type map of each operating area.

Step 5: according to the actual geographic information of the growing location, the corresponding agriculture in each operating area and each operating area Pharmaceutical quantities, by ground pesticide dispersal equipment to each operating area applying pesticides on plant growth ground.

In a kind of possible embodiment, after the crop progress variable farm chemical applying of place growing location, further includes:

In a kind of possible embodiment, the method also includes:

Specifically, in order to simplify calculating process, the disease degree index of the crop of estimation usually uses 1-9 natural number table Show.

A variable pesticide grown method for integrating remote sensing, model, algorithm provided through the foregoing embodiment, can pass through The prediction of the case where to potential disease, and then the type and dosage of pesticide needed for determining crop, so that effectively control pesticide makes With dosage, the utilization rate of pesticide is improved.In addition, in order to illustrate specific and conveniently, the method that the embodiment of the present application one is directed to, for Pesticide is illustrated, and is based on identical design concept, and method described herein can be also used for variable application herbicide, variable In the application scenarios such as fertilising.

Embodiment two

The embodiment of the present application two provide integrate remote sensing, model, algorithm variable pesticide grown method detailed process As shown in Fig. 2, may comprise steps of:

S201, variable farm chemical applying system obtains the historical growth season weather characteristics in crop location, and utilizes growing location weather Prediction model is based on historical growth season weather characteristics, the growing environment information of growing location where predicting current Growing season crop；

Here, historical growth season weather characteristics may include: EI Nino-south type of oscillation and degree, Growing season day Gas characteristic index.The growing environment information of growing location where current Growing season crop may include: current Growing season quantum of rainfall, Effective accumulated temperature, the maximum temperature of current Growing season environment and the minimum temperature of current Growing season environment that current Growing season utilizes.

S202, variable farm chemical applying system obtains the history crop onset state of growing location, and utilizes crop diseases prediction model Based on the growing environment information and history crop onset state of growing location where preceding Growing season crop, current Growing season crop is predicted Potential disease type and the case where infect the potential disease；

Here, history crop onset state may include: recurrent crop diseases type, the premunition of crop, on One season the disease type that has occurred of crop and severity of catching an illness.The case where infecting the potential disease includes: the hair of potential disease Severity and may occur for sick probability, catching an illness for potential disease the phase.

S203, growing environment information, crop after crop sowing, in variable farm chemical applying system acquisition process of crop growth Remote sensing images and field investigation result on the spot；

Here, the growing environment information in process of crop growth may include: aerial environmental information and underground growth Environmental information.Aerial environmental information may include: the weather conditions information of the growing location, such as environment temperature (day highest Temperature, Daily minimum temperature), ambient humidity (day highest humidity, day minimum humidity), rainfall (rainfall every hour or daily) Deng；The underground growing environment information may include: the depth of soil, soil texture, the soil texture, soil moisture content transformation, soil Earth nutrient, upper soll layer pH value (0-30 centimetres), soil bottom pH value (30-60 centimetres), if restricted root growth Soil layer, the ratio of soil stone and depth of root of the crop etc..

Field investigation result may include the picture of crop growth stage, description etc. on the spot, as whether there is or not lodging situations for crop Deng.

S204, variable farm chemical applying system is using meteorologic model based on the ground in the growing environment information in process of crop growth Growing environment information determines the Meteorological Index on plant growth ground；

Here, the weather conditions of growing location where the Meteorological Index on plant growth ground can characterize crop.

S205, variable farm chemical applying system utilizes soil moisture and nutrient model, based on the growing environment in process of crop growth Underground growing environment information in information, moisture needed for determining crop and nutrient situation；

Here, moisture situation needed for crop can coerce index by day water and be characterized, nutrient situation needed for crop It can be characterized by day Nutrient Stress index.

S206, variable farm chemical applying system utilize Remote Sensing Model, and the remote sensing images based on crop obtain the growing way index of crop；

Here, the growing way index of crop can characterize the growing way of crop.

S207, variable farm chemical applying system utilizes crop growth model, based on the growing environment information in process of crop growth, with And moisture needed for crop and nutrient situation, determine the growth phase of crop；

Here, the growth phase of crop can be characterized by crop growth stage index, such as crop growth stage refers to Number may include 10 growth phases, can be by V_NIt is indicated, wherein N is less than or equal to 10 positive integers.

S208, variable farm chemical applying system utilize crop diseases prediction model, Meteorological Index, crop institute based on plant growth ground The moisture and nutrient situation, the growing way index of crop, the growth phase of crop and field investigation on the spot needed is as a result, prediction is predicted not Carry out the case where crop infects the potential disease in predetermined time period.

Here, it may include: that potential disease type, the morbidity of potential disease are general that crop, which infects the case where potential disease, Rate, the catch an illness severity and crop cut yield etc. of potential disease.

For example, variable farm chemical applying system prediction is 86% to the probability that leaf blight occurs in corn, severity of catching an illness is low, jade The area of the aggrieved leaf of rice is less than 20%, and output reduction can then determine that corn does not need to spray less than 3%.

S209, variable farm chemical applying system according to the crop of prediction infect potential disease the case where, predict crop the application phase and Pesticide dosage, and growing location is divided into the operating area of at least one preset area, for operating area configuration and crop The pesticide dosage that potential disease type and the severity of catching an illness of potential disease match.

The two variable pesticide grown method for integrating remote sensing, model, algorithm provided through the foregoing embodiment, can pass through The prediction of the case where potential disease can determine the type and pesticide dosage of Pesticide use, to effectively control Pesticide use Dosage improves the utilization rate of pesticide.

Embodiment three

The embodiment of the present application three provide integrate remote sensing, model, algorithm variable pesticide grown system 30 basic knot Structure is as shown in Figure 2, comprising: obtains module 31, growing environment prediction module 32, disease forecasting module 33 and pesticide dosage and configures mould Block 34；Wherein,

The acquisition module 31, the historical growth season weather characteristics for growing location where obtaining crop；

The growing environment prediction module 32, for predicting current Growing season according to the historical growth season weather characteristics The growing environment information of growing location where the crop；

The disease forecasting module 33, for the history crop morbidity based on the growing environment information and the growing location Situation predicts the potential disease of crop described in current Growing season；And for being based on the crop after crop sowing Growth phase information and the crop growing environment, predict that the crop infection is described potential in the following predetermined time period The case where disease；

The pesticide dosage configuration module 34, for according to the crop of prediction infect the potential disease the case where, The pesticide dosage of the potential disease is treated for the crop disposition of the growing location.

It, can be by potential disease by the above-mentioned variable pesticide grown system 30 for integrating remote sensing, model, algorithm The prediction of situation, and then the type and dosage of pesticide needed for determining crop improve agriculture to effectively control Pesticide use dosage The utilization rate of medicine.

The disease forecasting module 33, specifically for according to following steps predict current Growing season described in crop potential disease Disease:

Based on growing location described in the growing environment information, the history onset state of the growing location, current Growing season The disease type that the crop of growing location described in the agrotype of growing location described in arable land situation, a upper season and a upper season occurs At least one of information, predict the potential disease of crop described in current Growing season.

Further, the disease forecasting module 33, specifically for determining the growth step of the crop according to following steps Segment information:

Obtain the growth step of the growing environment information of the crop, the growing way characteristic information of the crop and the crop Segment information；

The growing way characteristic information of growing environment information, the crop based on the crop and the growth step of the crop Segment information determines the growth phase information of the crop.

Further, the growing environment information includes: aerial environmental information and underground growing environment information；

The aerial environmental information includes: the weather conditions information of the growing location；

The underground growing environment information include: that the root of the crop is deep and the soil regime of the growing location at least It is a kind of.

The disease forecasting module 33, specifically for obtaining the growing way characteristic information of the crop according to following steps:

Obtain the remote sensing images of the crop；

The growing way characteristic information of the crop is extracted from the remote sensing images；

Based on the growing way characteristic information, the growing way index of the crop is determined.

Further, the case where crop infects the potential disease includes: the hair of potential disease type, potential disease The severity of catching an illness of sick probability and potential disease；

The pesticide dosage configuration module 34 infects described potential specifically for the crop predicted according to following steps The case where disease, treats the pesticide dosage of the potential disease for the crop disposition of the growing location:

The case where infecting the potential disease according to the crop of prediction, it is pre- to be divided at least one for the growing location If the operating area of area；

For each operating area, if the incidence rate of the potential disease of crop is greater than preset value on the operating area, it is The pesticide agent that operating area configuration matches with the potential disease type of the crop and the severity of catching an illness of potential disease Amount.

Optionally, the variable pesticide grown system 30 for integrating remote sensing, model, algorithm further include:

Route setup module 35 is used for the case where infecting the potential disease according to crop on the operating area, for institute State crop and the route sprayed insecticide be set, so as to operating apparatus according to the route sprayed insecticide and with the operating area phase Matched pesticide dosage is sprayed insecticide for the crop on the operating area.

The above-mentioned variable pesticide grown system 30 for integrating remote sensing, model, algorithm, not only can control Pesticide use agent Amount, can also be that the operating equipment sprayed insecticide plans reasonable work route according to pesticide dosage, so that operating equipment can be with Economy is sprayed insecticide for crop.

Embodiment three

The basic structure of a kind of electronic equipment 40 provided by the embodiment of the present application three is as shown in Figure 4, comprising: processor 41, memory 42 and bus 43；

The machine readable instructions that the memory storage 42 has the processor 41 executable, when network side equipment is run When, it is communicated between the processor 41 and the memory 42 by bus 43, the machine readable instructions are by the processor 41 execute following processing when executing:

The historical growth season weather characteristics of growing location where obtaining crop；

According to the historical growth season weather characteristics, the growing environment of growing location where predicting crop described in current Growing season Information；

History crop onset state based on the growing environment information and the growing location, is predicted described in current Growing season The potential disease of crop；

After crop sowing, the growing environment of growth phase information and the crop based on the crop, in advance Survey the case where crop infects the potential disease in the following predetermined time period；

The case where infecting the potential disease according to the crop of prediction treats institute for the crop disposition of the growing location State the pesticide dosage of potential disease.

In specific implementation, described based on the growing environment information and described in the processing that above-mentioned processor 41 executes The history crop onset state of growing location, predicts the potential disease of crop described in current Growing season, comprising:

In specific implementation, described to determine the crop according to following steps in the processing that above-mentioned processor 41 executes Growth phase information:

In specific implementation, in the processing that above-mentioned processor 41 executes, the growing environment information includes: aerial ring Border information and underground growing environment information；

In specific implementation, in the processing that above-mentioned processor 41 executes, the growing way characteristic information for obtaining the crop, Include:

Obtain the remote sensing images of the crop；

In specific implementation, in the processing that above-mentioned processor 41 executes, the case where crop infects the potential disease It include: the severity of catching an illness of potential disease type, the incidence rate of potential disease and potential disease；The institute according to prediction The case where crop infects the potential disease is stated, the pesticide agent of the potential disease is treated for the crop disposition of the growing location Amount, comprising:

In specific implementation, in the processing that above-mentioned processor 41 executes, the method also includes:

The case where infecting the potential disease according to crop on the operating area, the crop setting are sprayed insecticide Route, so that operating apparatus is according to the route sprayed insecticide and the pesticide dosage to match with the operating area, for institute The crop stated on operating area is sprayed insecticide.

Example IV

The embodiment of the present application four provides a kind of computer readable storage medium, stores on the computer readable storage medium Have computer program, the computer program executed when being run by processor it is above-mentioned integrate remote sensing, model, algorithm variable apply The step of pesticide method.

Specifically, which can be general storage medium, such as mobile disk, hard disk, on the storage medium Computer program when being run, be able to carry out the above-mentioned variable pesticide grown method for integrating remote sensing, model, algorithm, thus Solve the problems, such as the dosage that can not effectively control pesticide at present and caused by pesticide abuse, control the dosage of pesticide, mention The utilization rate of high pesticide.

Integrate provided by the embodiment of the present application remote sensing, model, algorithm variable pesticide grown method computer journey Sequence product, the computer readable storage medium including storing program code, before the instruction that program code includes can be used for execution Method in the embodiment of the method for face, specific implementation can be found in embodiment of the method, and details are not described herein.

It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description Specific work process, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.

If function is realized in the form of SFU software functional unit and when sold or used as an independent product, can store In a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words to existing Having the part for the part or the technical solution that technology contributes can be embodied in the form of software products, the computer Software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be personal meter Calculation machine, server or network equipment etc.) execute each embodiment method of the application all or part of the steps.And it is above-mentioned Storage medium includes: USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory The various media that can store program code such as (Random Access Memory, RAM), magnetic or disk.

More than, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, and it is any to be familiar with Those skilled in the art within the technical scope of the present application, can easily think of the change or the replacement, and should all cover Within the protection scope of the application.Therefore, the protection scope of the application should be subject to the protection scope in claims.

Claims

1. a kind of variable pesticide grown method for integrating remote sensing, model, algorithm, which is characterized in that the described method includes:

Mapping relations where the geography information of growing location where determining crop and crop between the actual geographic information of growing location；

According to the mapping relations, growing location where crop is divided at least one grid, wherein described in each grid is corresponding The operating area of preset area in growing location；

The case where infecting potential disease based on the crop for each operating area prediction, is the potential disease Incidence rate is greater than the operating area of preset value, configuration and the potential disease type of the crop and catching an illness seriously for potential disease The pesticide dosage that degree matches carries out variable farm chemical applying with the crop to place growing location.

2. the method according to claim 1, wherein predicting that the crop infects potential disease according to following steps The case where:

The growing environment feature of crop described in current Growing season based on prediction, the growing location history onset state, current The arable land situation of growing location described in Growing season, current Growing season crop resist this place and often sent out growth described in disease ability, a upper season At least one of the disease type that the crops of growing location described in the crop type on ground and a upper season occur information, prediction The potential disease of crops described in current Growing season；

After crop sowing, aerial environmental information, underground growth based on crop described in each operating area Environmental information, soil information, Indices, crop actual growing situation information and growth phase information predict that the following preset time is long Crop described in each operating area infects the case where potential disease in spending.

3. according to the method described in claim 2, it is characterized in that, the prediction future predetermined time period in each operation area Crop described in domain infects the case where potential disease, comprising:

4. according to the method described in claim 2, it is characterized in that, according to following steps predict current Growing season described in crop Growing environment feature, comprising:

Obtain the growing location historical growth season weather characteristics and the crop sow before Growing season day in preset time period Gas feature；

Growing season weather characteristics before being sowed according to the historical growth season weather characteristics and the crop in preset time period, in advance The growing environment feature of growing location where surveying crop described in current Growing season；

According to the temperature, humidity and rainfall of the growing location, the gas for characterizing the aerial environmental information of the crop is determined As index；

Deep, water demand information and nutrient demand information according to the root of crop described in each operating area, determine for characterizing The day water stress index and day Nutrient Stress index of the crop of the underground growing environment information；

Referred to according to the Weather information of the growing location, soil information, tillage system reform, upper season agrotype, day water stress Number, day Nutrient Stress index, day water requirement, growing environment day water supply, day by day amount of nutrients, growing environment day for amount of nutrients, when Preceding Growing season density of crop, the agrotype of the crop and day disease coerce index, predict the growth step of the crop Segment information.

5. according to the method described in claim 2, it is characterized in that, determining the Indices by following steps:

Obtain the remote sensing images of the crop；

6. the method according to claim 1, wherein the configuration and the potential disease type of the crop and latent In the pesticide dosage that the severity of catching an illness of disease matches, comprising:

The case where infecting potential disease according to the crop of each operating area prediction, the treatment potential disease pesticide Current pesticide price, the crop price of prediction are determined as the pesticide dosage of each operating area configuration.

7. method according to claim 1 or 6, which is characterized in that the method also includes:

It is the crop according to the operating apparatus used on the pesticide dosage and each operating area determined on each operating area The sprinkling strategy of the route sprayed insecticide and the operating apparatus is set；

The operating apparatus is according to the route sprayed insecticide, the pesticide dosage and the work that match with the operating area The sprinkling strategy of industry instrument is sprayed insecticide for the crop on the operating area.

8. the method according to claim 1, wherein after carrying out variable farm chemical applying to the crop of place growing location, Further include:

According to the actual growing situation information and/or practical defect information, configuration and the actual growing situation information and/or practical disease The supplement pesticide dosage that information matches, in the case where affiliated supplement pesticide dosage is greater than zero, with the institute to place growing location It states crop and carries out additional variable application.

9. the method according to claim 1, wherein the method also includes:

According to the practical defect information, the Indices and the soil information, institute in each operating area is judged State whether crop there is application to be worth；

In the case that the crop described in the operating area has application value, estimates and make described in each operating area The disease degree index of object；

Worked as according to the pesticide of the disease degree index of crop described in each operating area of estimation, treatment currently practical disease Preceding pesticide price, the crop price of prediction are determined as the pesticide dosage of each operating area configuration.

10. a kind of variable pesticide grown system for integrating remote sensing, model, algorithm, which is characterized in that the system comprises: it draws Sub-module and variable farm chemical applying module；Wherein,

The division module, for the remote sensing images of growing location where crop to be divided at least one grid, wherein each net Lattice correspond to the operating area of preset area in the growing location；

The variable farm chemical applying module, for based on the crop infection potential disease for each operating area prediction Situation is that the incidence rate of the potential disease is greater than the operating area of preset value, potential disease class of the configuration with the crop The pesticide dosage that the severity of catching an illness of type and potential disease matches carries out variable farm chemical applying with the crop to place growing location.