CN111126720A

CN111126720A - Farm risk prediction method, device, equipment and storage medium

Info

Publication number: CN111126720A
Application number: CN202010130757.8A
Authority: CN
Inventors: 邱添羽; 周古月; 吴泽衡; 徐倩; 杨强
Original assignee: WeBank Co Ltd
Current assignee: WeBank Co Ltd
Priority date: 2020-02-28
Filing date: 2020-02-28
Publication date: 2020-05-08
Anticipated expiration: 2040-02-28
Also published as: CN111126720B

Abstract

The invention discloses a risk prediction method, a risk prediction device, risk prediction equipment and a storage medium for an aquaculture farm, wherein the method comprises the following steps: acquiring first position information of a farm to be predicted, and determining a target geographical range corresponding to the farm to be predicted based on the first position information; screening target influence factors falling into the target geographical range from the risk influence factors according to preset second position information of the risk influence factors; and acquiring influence data corresponding to the target influence factors, and inputting the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted. The method and the system realize rapid and accurate prediction of the risk condition of the farm.

Description

Farm risk prediction method, device, equipment and storage medium

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to a method, a device, equipment and a storage medium for predicting risk of a farm.

Background

African swine fever is an acute, febrile and highly contagious animal infectious disease of pigs caused by African swine fever virus, and the related lethality rate is up to 100%. Since the first African swine fever in China is confirmed to be diagnosed by the Chinese animal health and epidemiology center, the African swine fever epidemic situation occurs in part of China and the loss is serious. Once an infectious disease of an animal, which is similar to the African swine fever, is discovered, a prevention and control measure needs to be taken quickly and accurately to avoid the negative influence on all aspects of society caused by the continuous expansion of epidemic situations. In order to take rapid and accurate preventive and control measures, the risk condition of each farm needs to be known rapidly and accurately. However, the current situation is risk assessment of the farm, and the assessment needs to be carried out after an expert inspects the condition of the farm on the spot, or the risk information of the farm is reported by a farmer. However, the scheme of field investigation by experts is obviously low in efficiency and difficult to deal with the outburst and rapidly spread epidemic situation; the active reporting of the farmers can cause the situation that the farmers withhold the breeding risk information, and the farmers are difficult to accurately estimate the risk, so that the risk assessment accuracy is low. The risk condition of the farm cannot be quickly and accurately obtained, and the prevention and control working effect is poor.

Disclosure of Invention

The invention mainly aims to provide a risk prediction method, a device, equipment and a storage medium for a farm, aiming at solving the problem that the prevention and control work effect is poor due to the fact that the risk condition of the farm cannot be rapidly and accurately obtained aiming at the epidemic situation of animals at present.

In order to achieve the above object, the present invention provides a farm risk prediction method, including the steps of:

acquiring first position information of a farm to be predicted, and determining a target geographical range corresponding to the farm to be predicted based on the first position information;

screening target influence factors falling into the target geographical range from the risk influence factors according to preset second position information of the risk influence factors;

and acquiring influence data corresponding to the target influence factors, and inputting the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted.

Optionally, the target influencing factor includes a target water network and a target farm, and the step of obtaining the influencing data corresponding to the target influencing factor includes:

determining a waterway net density characteristic value of the target geographic range according to the second position information of the target waterway net and the target geographic range;

determining a farm density characteristic value of the target geographical range according to the second position information of the target farm and the target geographical range;

and taking the density characteristic value of the waterway network and the density characteristic value of the farm as the influence data.

Optionally, the step of determining a farm density characteristic value of the target geographical range according to the second location information of the target farm and the target geographical range includes:

determining the area of the target farm according to the second position information of the target farm;

determining the total feeding amount of the target farm according to the area of the farm and a preset corresponding relation between the area and the feeding amount;

and calculating to obtain a farm density characteristic value of the target geographical range according to the total feeding amount and the area of the target geographical range.

Optionally, the step of determining the characteristic value of the density of the waterway net in the target geographic range according to the second location information of the target waterway net and the target geographic range includes:

determining the length or area of a waterway network of the target waterway network according to the second position information of the target waterway network;

and calculating to obtain a waterway net density characteristic value of the target geographical range according to the waterway net length and the area of the target geographical range, or calculating to obtain the waterway net density characteristic value of the target geographical range according to the waterway net area and the area of the target geographical range.

Optionally, the risk prediction model includes a preset weight value corresponding to each of the influence data, and the step of inputting the influence data into the preset risk prediction model to obtain a risk prediction result of the farm to be predicted includes:

inputting each influence data into the risk prediction model to call the risk prediction model, and calculating to obtain a risk coefficient of the farm to be predicted based on the weight value and each influence data;

and taking the risk coefficient as a risk prediction result of the farm to be predicted, or taking the risk grade as a risk prediction result of the farm to be predicted after determining the risk grade according to the risk coefficient and a preset corresponding relation between the coefficient and the grade.

Optionally, before the step of screening, according to preset second location information of each risk influencing factor, a target influencing factor falling into the target geographic range from each risk influencing factor, the method further includes:

inputting a map to be identified into an influence factor detection model to obtain the classification category of each pixel point in the map to be identified, wherein the influence factor detection model is obtained by pre-training;

determining target pixel points belonging to risk influence factors from all the pixel points based on the classification category;

and determining the actual position of each target pixel point according to the scale of the map to be identified, and obtaining second position information of each risk influence factor based on the actual position.

Optionally, the risk influencing factor includes a road network, and before the step of inputting the map to be recognized into the influencing factor detection model to obtain the classification category of each pixel point in the map to be recognized, the method further includes:

performing primary training on a semantic segmentation model to be trained by adopting regular training data, wherein the regular training data comprises a plurality of pre-collected remote sensing satellite maps containing a water network and water network annotation data corresponding to each remote sensing satellite;

adjusting the preliminarily trained semantic segmentation model by adopting negative example training data or adopting the negative example training data and the positive example training data, wherein the negative example training data comprise a plurality of remote sensing satellite maps which are collected in advance and do not comprise a water network;

when the adjusted model to be trained is detected to accord with the preset model condition, the adjusted semantic segmentation model is used as the influence factor detection model, otherwise, the step is executed again based on the adjusted semantic segmentation model: and (4) performing primary training on the semantic segmentation model to be trained by adopting the normal training data.

Optionally, after the step of obtaining the influence data corresponding to the target influence factor and inputting the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted, the method further includes:

and determining a target prevention and control scheme of the farm to be predicted from preset prevention and control schemes according to the risk prediction result, and outputting the target prevention and control scheme.

In order to achieve the above object, the present invention also provides a farm risk prediction device, including:

the system comprises an acquisition module, a storage module and a display module, wherein the acquisition module is used for acquiring first position information of a farm to be predicted and determining a target geographical range corresponding to the farm to be predicted based on the first position information;

the screening module is used for screening target influence factors falling into the target geographical range from the risk influence factors according to preset second position information of the risk influence factors;

and the prediction module is used for acquiring influence data corresponding to the target influence factors and inputting the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted.

In order to achieve the above object, the present invention also provides a farm risk prediction apparatus, including: a memory, a processor, and a farm risk prediction program stored on the memory and executable on the processor, the farm risk prediction program when executed by the processor implementing the steps of the farm risk prediction method as described above.

Furthermore, to achieve the above object, the present invention also proposes a computer-readable storage medium having stored thereon a farm risk prediction program that, when executed by a processor, implements the steps of the farm risk prediction method as described above.

According to the method, a target geographical range corresponding to the farm to be predicted is determined according to first position information of the farm to be predicted by obtaining the first position information; screening target influence factors falling into a target geographical range from the risk influence factors according to preset second position information of the risk influence factors; and inputting the influence data of the target influence factors into a risk prediction model to obtain a risk prediction result of the farm to be predicted. According to the method, for the farms with risks to be predicted in various places, influence data of risk influence factors in a target geographical range corresponding to the farms to be predicted are obtained, and a risk prediction model is set to process the influence data, so that a risk prediction result of the farms to be predicted can be obtained; the condition of each farm is not required to be inspected by experts on the spot, so that the risk prediction efficiency of the farm is improved, and the condition that the epidemic situation spreads rapidly can be dealt with; compared with the scheme that the risk condition is actively reported by the farmers, the risk condition of each farm can be more accurately predicted by adopting a unified risk prediction model in the embodiment; based on a rapid and accurate risk prediction result, the epidemic prevention and control work can be rapidly and pertinently carried out, and the epidemic out of control is avoided.

Drawings

FIG. 1 is a schematic diagram of a hardware operating environment according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart illustrating a first embodiment of a risk prediction method for a farm according to the present invention;

FIG. 3 is a block diagram of a functional diagram of a farm risk prediction device according to a preferred embodiment of the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

As shown in fig. 1, fig. 1 is a schematic device structure diagram of a hardware operating environment according to an embodiment of the present invention.

It should be noted that, the plant risk prediction device in the embodiment of the present invention may be a smart phone, a personal computer, a server, and the like, and is not limited herein.

As shown in fig. 1, the farm risk prediction apparatus may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.

Those skilled in the art will appreciate that the configuration of the apparatus shown in fig. 1 does not constitute a limitation of the plant risk prediction apparatus and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.

As shown in fig. 1, a memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and a farm risk prediction program. The operating system is a program for managing and controlling hardware and software resources of the equipment, and supports the operation of a farm risk prediction program and other software or programs.

In the device shown in fig. 1, the user interface 1003 is mainly used for data communication with a client; the network interface 1004 is mainly used for establishing communication connection with a server; and the processor 1001 may be configured to invoke the farm risk prediction program stored in the memory 1005 and perform the following operations:

Further, the target influence factor includes a target water network and a target farm, and the step of obtaining the influence data corresponding to the target influence factor includes:

Further, the step of determining a farm density characteristic value of the target geographical range according to the second location information of the target farm and the target geographical range includes:

Further, the step of determining the characteristic value of the density of the waterway net in the target geographical range according to the second position information of the target waterway net and the target geographical range includes:

Further, the risk prediction model includes a preset weight value corresponding to each of the influence data, and the step of inputting the influence data into the preset risk prediction model to obtain a risk prediction result of the farm to be predicted includes:

Further, before the step of screening the target influencing factors falling into the target geographical range from the risk influencing factors according to the preset second position information of each risk influencing factor, the processor 1001 may be further configured to call a farm risk prediction program stored in the memory 1005, and perform the following operations:

Further, the risk influencing factors include a water network, and before the step of inputting the map to be recognized into the influencing factor detection model to obtain the classification categories of the pixel points in the map to be recognized, the processor 1001 may be further configured to call a farm risk prediction program stored in the memory 1005, and execute the following operations:

Further, after the step of obtaining the influence data corresponding to the target influence factor and inputting the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted, the processor 1001 may be further configured to call a farm risk prediction program stored in the memory 1005, and perform the following operations:

Based on the above structure, various embodiments of a farm risk prediction method are provided.

Referring to fig. 2, fig. 2 is a schematic flow chart of a farm risk prediction method according to a first embodiment of the present invention.

While a logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different than that shown or described herein. The execution subject of each embodiment of the farm risk prediction method of the invention can be a smart phone, a personal computer, a server and other devices, and for convenience of description, the execution subject is omitted in the following embodiments for explanation. In this embodiment, the method for predicting risk of a farm includes:

step S10, acquiring first position information of the farm to be predicted, and determining a target geographical range corresponding to the farm to be predicted based on the first position information;

for a farm needing risk prediction, the position information of the farm can be obtained first, and the farm is taken as the farm to be predicted. The risk may be the degree of influence that the farm may be affected when the infectious disease is transmitted by an animal such as swine fever. There are various ways to obtain the location information of the farm; for example, the method can be obtained from a map, and based on the obtained electronic map data, relevant information of the farm is extracted from the map data, including extracting position information of the farm; the position information of the farm can be extracted from farm data reported by farmers or pre-collected farm data; other methods are also possible, not to mention here. The farm can be seen as a point (the central point of the farm), and the position information is the position information of the point; the farm may be treated as a region, and the location information may be location information of the region. The position information may be longitude and latitude of the farm, or coordinates under a predetermined specific coordinate system, and the like.

And determining a target geographical range corresponding to the farm to be predicted based on the first position information of the farm to be predicted. It should be noted that, for the farm to be predicted, the risk of the farm is to be predicted, the principle adopted in this embodiment is that the relevant data of the area around the farm can be used as the basis for judging the risk of the farm, and the target geographical range can be the determined area around the farm. In particular, the region may be determined in a number of ways; for example, a radius value may be preset, a divided circular area with the farm as a center of a circle and the radius value as a radius may be used as a target geographic range corresponding to the farm, and a position of the target geographic area may be determined according to the position information of the farm and the radius value; the method can also be used for dividing a matrix area as a target geographical range of the farm by taking the farm as the center of a rectangle and presetting a matrix edge length value; there are other feasible methods, and different methods may be adopted to determine the target geographic range according to different actual application scenarios, which are not listed here.

Step S20, screening target influence factors falling into the target geographical range from the risk influence factors according to preset second position information of the risk influence factors;

various risk influencing factor types, such as farms, water networks, road networks, villages, etc., which are different types of influencing factors, may be specified in advance. The water network may refer to rivers, lakes and the like with various sizes, and the road network may refer to various roads, such as national roads, high speed and the like. Hereinafter, the water network and the road network are collectively referred to as a "waterway network", and the water network and the road network may be viewed together or separately. Data of each risk influence factor of a region can be collected in advance, for example, data of risk influence factors of a certain city or a certain region can be collected, that is, data of each farm, data of each water network, data of each road network and the like of the region can be collected; the data may specifically include the location, size, and the like of the risk influencing factor, and the data may be different according to the type of the risk influencing factor, for example, the collected farm data may include location information, area, yield, and the like of a farm, and the collected waterway network data may include location information, length, and the like of a waterway network. The data may be collected by extracting it directly from the electronic map data.

And after the data of each risk influence factor is acquired, screening the target influence factors falling into the target geographical range from each risk influence factor according to the second position information of each risk influence factor. Specifically, for each risk influencing factor, whether the second position information of the risk influencing factor is within the target geographic range or not can be judged, if yes, the risk influencing factor is determined to be within the target geographic range, and if not, the risk influencing factor is determined not to be within the target geographic range; the risk impact determined to be at the target geographic scope is taken as the target impact. That is, it is determined which farms, which water network networks, etc. of a region are within the target geographical range corresponding to the farm to be predicted. For example, when the location information is longitude and latitude, the location of the target geographic range may be composed of a longitude interval and a latitude interval, and it is necessary to determine whether the longitude of each risk influencing factor falls into the longitude interval and whether the latitude falls into the latitude interval, and if both of the longitude and the latitude fall into the latitude interval, it is determined that the risk influencing factor falls into the target geographic range.

And step S30, obtaining influence data corresponding to the target influence factors, and inputting the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted.

Acquiring influence data corresponding to the target influence factors, and inputting the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted. Specifically, a risk prediction model may be preset, the risk prediction model may be a model structure such as a linear model or a neural network model, if the risk prediction model is the neural network model, the neural network model needs to be trained in advance, and the trained model is used for risk prediction. The setting principle of the risk prediction model is as follows: when the more and more target influence factors in the target geographic range are represented by the input data, the more dense the input data are, the higher the risk of the farm to be predicted is represented by the output risk prediction result. Input data of the risk prediction model can be set according to situations, and the input data is called influence data. The influence data of the target influence factors can be different according to different types of the target influence factors; for example, for a farm, the impact data may be the total area of the farm within a target geographic range, and for a water network, the impact data may be the total area of the water network within the target geographic range.

And inputting the influence data into a risk prediction model, processing the influence data through the risk prediction model, and outputting a risk prediction result of the farm to be predicted. The risk prediction result may be in various forms, for example, a score or a grade. The risk levels corresponding to the various scores or grades may be predefined, for example, the score is 0 to 10, and the risk levels become larger from 0 to 10.

In the embodiment, a target geographical range corresponding to the farm to be predicted is determined according to first position information by acquiring the first position information of the farm to be predicted; screening target influence factors falling into a target geographical range from the risk influence factors according to preset second position information of the risk influence factors; and inputting the influence data of the target influence factors into a risk prediction model to obtain a risk prediction result of the farm to be predicted. In the embodiment, for the farm to be subjected to risk prediction in each place, influence data of risk influence factors in a target geographical range corresponding to the farm to be subjected to risk prediction are acquired, and a risk prediction model is set to process the influence data, so that a risk prediction result of the farm to be subjected to risk prediction can be obtained; the condition of each farm is not required to be inspected by experts on the spot, so that the risk prediction efficiency of the farm is improved, and the condition that the epidemic situation spreads rapidly can be dealt with; compared with the scheme that the risk condition is actively reported by the farmers, the risk condition of each farm can be more accurately predicted by adopting a unified risk prediction model in the embodiment; based on a rapid and accurate risk prediction result, the epidemic prevention and control work can be rapidly and pertinently carried out, and the epidemic out of control is avoided.

Further, after the step S30, the method further includes:

and step S40, determining a target prevention and control scheme of the farm to be predicted from preset prevention and control schemes according to the risk prediction result, and outputting the target prevention and control scheme.

Furthermore, different prevention and control schemes can be set in advance according to different risk prediction results, after the risk prediction result of the farm to be predicted is obtained, a proper prevention and control scheme is matched from the various prevention and control schemes according to the risk prediction result, the matched prevention and control scheme is determined as a target prevention and control scheme of the farm to be predicted, and the target prevention and control scheme can be output. The output mode can be output to a display screen for display or output in a voice mode, and the output mode can be set differently according to specific application scenes. The control scheme of the farm to be predicted is directly output and displayed, so that a user can visually know the control scheme, and control work is carried out according to the control scheme. Particularly, for users who need to take prevention and control measures but do not know which prevention and control measures exist or do not know the risk condition of the farm, such as farmers or residents around the farm, the prevention and control measures matched with the risk condition of the farm can be conveniently and directly obtained, so that the development of prevention and control work is facilitated.

Based on the risk prediction method in the embodiment, the risk prediction method can be practically applied in a mobile phone or computer application software mode, so that a user can input relevant information, such as position information, of the farm to be predicted based on an installed application program, and a risk prediction result of the farm to be predicted, even a prevention and control scheme, can be directly obtained.

Further, based on the first embodiment, a second embodiment of the method for predicting risk of a farm according to the present invention is provided, in this embodiment, the target influencing factor includes a target water network and a target farm, and the step of acquiring the influence data corresponding to the target influencing factor in step S30 includes:

step S301, determining a waterway net density characteristic value of the target geographical range according to second position information of the target waterway net and the target geographical range;

further, in this embodiment, the target influencing factors may be the target waterway network and the target farm. That is, the types of risk influencing factors may include the water network and the farm. After the target farm and the target water network falling within the target geographical range are determined, influence data of the target farm and the target water network can be obtained.

Specifically, the waterway net density characteristic value of the target geographic range is determined according to the second position information of the target waterway net and the target geographic range. The method for calculating the density characteristic value of the waterway network may be various, and for example, the result may be obtained by dividing the total area of the target waterway network by the total area of the target geographical range.

Further, step S301 includes:

step S3011, determining the length or area of the waterway network of the target waterway network according to the second position information of the target waterway network;

step S3012, calculating to obtain a waterway net density characteristic value of the target geographical range according to the waterway net length and the area of the target geographical range, or calculating to obtain the waterway net density characteristic value of the target geographical range according to the waterway net area and the area of the target geographical range.

Specifically, the length or the area of the target road network may be determined according to the second position information of the target road network, and the calculation process of the area and the length may be as follows: the total area of the target geographic range can be determined according to the position information of the target geographic range, and if the total area of the target geographic range can be calculated according to the longitude and latitude interval of the target geographic range; determining the length of the target water network according to the second position information of the target water network, for example, one river or one road can be represented by the longitude and latitude of two points, the length of the river or the road is calculated according to the longitude and latitude of the two points, and the lengths of all the rivers or the roads are added to obtain the length of the target water network; the total area of the target water network may also be determined according to the second position information of the target water network, for example, one river or one road may be represented by latitude and longitude intervals, at this time, the area of the river or the road is calculated according to the latitude and longitude intervals, and the total area of the target water network is obtained by adding the areas of all the rivers or the roads.

The characteristic value of the density of the waterway net can be a result obtained by dividing the total area of the target waterway net by the total area of the target geographic range, or can be a result obtained by dividing the length of the target waterway net by the total area of the target geographic range.

Step S302, determining a farm density characteristic value of the target geographical range according to the second position information of the target farm and the target geographical range;

and determining a farm density characteristic value of the target geographical range according to the second position information of the target farm and the target geographical range. Specifically, the farm density characteristic value may be a result of dividing the total area of the target farm by the total area of the target geographical range. The total area calculation method of the target farm may be: each target farm can be represented by a longitude and latitude interval, the area of each target farm is obtained through calculation according to the longitude and latitude intervals, and then the areas are added to obtain the total area of the target farm.

Step S303, using the characteristic value of the density of the waterway network and the characteristic value of the density of the farm as the influence data.

And taking the density characteristic value of the water path network and the density characteristic value of the farm as influence data of a target influence factor, namely inputting the density characteristic value of the water path network and the density characteristic value of the farm into a risk prediction model to obtain a risk prediction result of the farm to be predicted.

In this embodiment, the risk prediction result of the farm to be predicted is calculated by calculating the farm density characteristic value and the waterway network density characteristic value of the target geographical range corresponding to the farm to be predicted, inputting the farm density characteristic value and the waterway network density characteristic value into the risk prediction model, and calculating the risk prediction result of the farm to be predicted, so that the risk of the farm is predicted according to the density characteristics of the farm and the waterway network of the target geographical range, and the actual infectious characteristics of the animal infectious diseases are combined, so that the predicted risk prediction result is more accurate.

Further, the step S302 includes:

step S3021, determining a farm area of the target farm according to the second position information of the target farm;

the farm density characteristic value may also be a result of dividing the total feeding amount of the target farm by the total area of the target geographical range, i.e., the farm density characteristic value may be the feeding density. The total feeding amount of the target farm can be calculated in the following mode:

and determining the area of the target farm according to the second position information of the target farm, namely the total area calculation process of the target farm.

Step S3022, determining the total feeding amount of the target farm according to the area of the farm and a preset corresponding relation between the area and the feeding amount;

the correspondence between the area of the farm and the amount of feed can be set in advance, for example, 1.5 square meters is set to produce one pig. And determining the total feeding amount of the target farm according to the calculated farm area of the target farm and the corresponding relation. For example, if the farm area is 150 square meters, the total feeding amount of the target farm is 100 pigs according to the above example.

And step S3023, calculating a farm density characteristic value of the target geographical range according to the total feeding amount and the area of the target geographical range.

And calculating to obtain the density characteristic value of the farm in the target geographical range according to the total feeding amount and the total area of the target geographical range. Specifically, the total feeding amount may be divided by the total area, and the result is taken as a characteristic value of farm density.

Further, the risk prediction model includes a preset weight value corresponding to each of the influence data, and the step of inputting the influence data into the preset risk prediction model in step S30 to obtain the risk prediction result of the farm to be predicted includes:

step S304, inputting each influence data into the risk prediction model so as to call the risk prediction model to calculate a risk coefficient of the farm to be predicted based on the weight value and each influence data;

further, the risk prediction model may be a linear model, and includes weighted values corresponding to the respective influence data, for example, when the influence data are a farm density characteristic value and a waterway network density characteristic value, a weight corresponding to the farm density characteristic value, a weight corresponding to the waterway network density characteristic value, and a weight corresponding to the waterway network density characteristic value may be set, respectively.

After the influence data of the target influence factors are acquired, inputting each influence data into a risk prediction model, and calling the risk prediction model to calculate the risk coefficient of the farm to be predicted based on each weight value and each influence data. Specifically, each influence data may be multiplied by a corresponding weight value, and then the results obtained by the multiplication are added, and the added result is used as a risk coefficient, that is, the nature of the risk prediction model may be a linear model. It should be noted that each weight value in the risk prediction model may be set according to specific experience.

Step S305, using the risk coefficient as a risk prediction result of the farm to be predicted, or after determining a risk grade according to the risk coefficient and a preset corresponding relation between the coefficient and the grade, using the risk grade as the risk prediction result of the farm to be predicted.

After the risk coefficient is obtained, the risk coefficient can be directly used as a risk prediction result of the farm to be predicted. The corresponding relationship between the risk coefficients and the levels may also be preset, for example, the risk coefficients are sorted from low to high, the lowest risk coefficient 1/4 is divided into a first level, 1/4-1/2 is divided into a second level, 1/2-3/4 is divided into a third level, and the highest risk coefficient 1/4 is divided into a fourth level. It should be noted that the number of levels and the boundary of each level can be selected according to specific experience. And determining which risk grade the risk coefficient corresponds to according to the obtained risk coefficient and the corresponding relation between the coefficient and the grade, and taking the determined risk grade as a risk prediction result of the farm to be predicted. According to different actual requirements of users, risk prediction results can be different, and risk levels are possibly more intuitive for some users, so that the users can make a targeted prevention and control scheme.

Further, based on the first and second embodiments, a third embodiment of the risk prediction method for a farm according to the present invention is provided, and in this embodiment, the risk prediction method for a farm further includes:

step S50, inputting a map to be recognized into an influence factor detection model to obtain the classification category of each pixel point in the map to be recognized, wherein the influence factor detection model is obtained by pre-training;

the data of each risk influencing factor may be collected in advance, and in this embodiment, the collection method may be: an influence detection model for identifying risk influencing factors in a map may be trained in advance. The influence factor detection model can be a multi-classification model and is used for identifying which of multiple risk influence factors each pixel point in the map belongs to; the influence factor detection model can also be a binary model used for identifying whether each pixel point in the map belongs to a certain risk influence factor. The influence factor detection model can adopt a common image target detection model, such as a semantic image segmentation model DeepLab-v3 +.

And taking the map of the area where the farm is to be predicted as the map to be identified, and inputting the map to be identified into the influence factor detection model to obtain the classification category of each pixel point in the map to be identified. The classification type is used for indicating which risk influence factor the corresponding pixel belongs to, or is used for indicating whether the corresponding pixel belongs to a certain risk influence type.

Step S60, determining target pixel points belonging to risk influence factors from all the pixel points based on the classification categories;

and determining target pixel points belonging to risk influence factors from the pixel points according to the classification category of each pixel point. That is, according to the classification category, it is determined which pixel points belong to the risk influencing factor, and if the classification is multi-classified, it is also determined which type of risk influencing factor the pixel points belonging to the risk influencing factor belong to. For example, when the multi-classification includes four categories, which do not belong to risk influencing factors, farms, water networks and road networks, it is determined which pixels of each category are according to the classification category of each pixel.

And step S70, determining the actual position of each target pixel point according to the scale of the map to be identified, and obtaining second position information of each risk influence factor based on the actual position.

And determining the actual position of each target pixel point according to the scale of the map to be identified. Specifically, according to the scale of the map to be identified, the number of actual areas represented by each pixel point can be determined, then according to the longitude and latitude interval of the map to be identified, the longitude and latitude interval of each target pixel point can be determined, and as each target pixel point corresponds to each risk influence factor, the longitude and latitude interval of the target pixel point can be used as the longitude and latitude interval of the corresponding risk influence factor, and the longitude and latitude interval is used as the position information of the risk influence factor.

It should be noted that, for the pixels belonging to the farm and identified in the map to be identified, when the actual corresponding area of one pixel is smaller, it is likely that one pixel represents a breeding house (i.e. a unit smaller than the farm), at this time, a clustering algorithm may be adopted to cluster a plurality of identified pixels, or cluster longitude and latitude intervals corresponding to a plurality of pixels, and cluster a plurality of breeding houses with closer distances into one farm. The clustering algorithm may employ a conventional clustering algorithm, such as a neighbor clustering method. The longitude and latitude of the farm can be the center of the cluster, and the feeding amount of the farm can be the sum of the feeding amounts of all breeding houses in the cluster range.

Further, the risk influencing factors include a water network, and the method for predicting the risk of the farm further includes:

step A10, performing preliminary training on a semantic segmentation model to be trained by adopting regular training data, wherein the regular training data comprises a plurality of pre-collected remote sensing satellite maps containing a water network and water network annotation data corresponding to each remote sensing satellite map;

in this embodiment, the risk influencing factors may include a water network, the influencing factor detection model may be a multi-classification model formed by a semantic segmentation model, and the output result of the preset influencing factor detection model is three categories: and the method does not belong to risk influencing factors, water networks and road networks. The method for training the influence factor detection model can be as follows:

the method comprises the steps of collecting a plurality of remote sensing satellite maps containing water networks in advance, and collecting mark data of the water networks in the remote sensing satellite maps, namely the mark data of the water networks. The waterway net annotation data can be a mask graph corresponding to the remote sensing satellite picture, classification categories corresponding to each pixel point of the remote sensing satellite picture are arranged in the mask graph, if 0 indicates that the pixel point does not belong to risk influence factors, 1 indicates that the pixel point belongs to a waterway net, 2 indicates that the pixel point belongs to a road net, and different colors can be adopted in the mask graph to indicate different categories. And taking a plurality of remote sensing satellite maps containing a water network and water network label data as the training data of the prime example.

Firstly, the semantic segmentation model to be trained is preliminarily trained by adopting the normal training data. The process of preliminary training may employ an existing machine learning model training process.

Step A20, adopting negative example training data or adopting the negative example training data and the positive example training data to adjust the preliminarily trained semantic segmentation model, wherein the negative example training data comprise a plurality of pre-collected remote sensing satellite maps without a water network;

a plurality of remote sensing satellite maps without a water network are collected in advance to serve as negative training data, the proportion of the negative training data to the positive training data can be set according to practical experience, and for example, the negative training data can be one tenth of the positive training data. After the initial training, the semantic segmentation model after the initial training is adjusted by adopting the negative training data and the positive training data together, or the semantic segmentation model after the initial training is adjusted by adopting the negative training data alone, wherein the adjustment at this time can be fine adjustment. Specifically, the fine tuning is trained by using the positive training data and the negative training data together, and the training process can also be the training process of the existing machine learning model. It should be noted that, during fine tuning, the hyper-parameters of the model may be adjusted according to the ratio of the positive training data to the negative training data, and then fine tuning training is performed after the adjustment, for example, when the negative training data may be one tenth of the positive training data, the learning rate in the fine tuning stage may also be set to one tenth of the initial training stage.

Step A30, when detecting that the adjusted model to be trained accords with the preset model condition, taking the adjusted semantic segmentation model as the influence factor detection model, otherwise, executing the steps based on the adjusted semantic segmentation model: and (4) performing primary training on the semantic segmentation model to be trained by adopting the normal training data.

And taking the finely adjusted semantic segmentation model as an influence factor detection model so as to detect risk influence factors by adopting the influence factor detection model in the following.

And detecting whether the adjusted semantic segmentation model meets preset model conditions, and if so, taking the adjusted semantic segmentation model as an influence factor detection model so as to detect risk influence factors by adopting the influence factor detection model. Otherwise, if the preset model condition is not met, performing preliminary training on the adjusted semantic segmentation model by adopting the normal training data, and then performing adjustment training until the preset model condition is detected to be met. The preset model condition may be a condition set in advance according to a performance requirement of the model, for example, a loss function convergence of the model may be used as a condition, or a commonly used objective index for detecting the model performance may be used as a condition, for example, objective indexes such as accuracy, recall rate, and IOU (Intersection over unit, Intersection ratio). The calculation method of each objective index may refer to the existing index calculation method, which is not described in detail herein.

In the embodiment, an influence factor detection model is trained in advance, a map to be recognized is detected and recognized, and data of each risk influence factor in the map to be recognized is obtained, so that the risk influence factors of each region can be intelligently and quickly acquired, and the risk prediction efficiency of a farm to be predicted is improved. Moreover, some data of risk influencing factors which cannot be extracted from the map data can be detected through the influencing factor detection model, so that the data of the risk influencing factors are more comprehensive, the data on which the risk prediction is based is more accurate and comprehensive, and the risk prediction result is more accurate.

And the model is initially trained by adopting the positive training data, and then the model after the initial training is finely adjusted by adopting the positive training data and the negative training data together, namely, the correction is carried out by the negative training data, so that the false detection of the model is reduced. And the model is finely adjusted by combining the positive training data and the negative training data, so that the situation that the training effect of the positive training data is covered by adopting the negative training data is avoided.

In addition, an embodiment of the present invention further provides a farm risk prediction device, and referring to fig. 3, the farm risk prediction device includes:

the system comprises an acquisition module 10, a storage module and a display module, wherein the acquisition module is used for acquiring first position information of a farm to be predicted and determining a target geographical range corresponding to the farm to be predicted based on the first position information;

the screening module 20 is configured to screen target influence factors falling into the target geographic range from the risk influence factors according to preset second location information of the risk influence factors;

and the prediction module 30 is configured to obtain influence data corresponding to the target influence factor, and input the influence data into a preset risk prediction model to obtain a risk prediction result of the farm to be predicted.

Further, the target influencing factors include a target water network and a target farm, and the prediction module 30 includes:

the first determining unit is used for determining a waterway network density characteristic value of the target geographic range according to the second position information of the target waterway network and the target geographic range;

the second determining unit is used for determining a farm density characteristic value of the target geographical range according to second position information of the target farm and the target geographical range;

and the third determining unit is used for taking the waterway net density characteristic value and the farm density characteristic value as the influence data.

Further, the second determination unit includes:

a first determining subunit, configured to determine a farm area of the target farm according to the second location information of the target farm;

the second determining subunit is used for determining the total feeding amount of the target farm according to the area of the farm and a preset corresponding relation between the area and the feeding amount;

and the first calculating subunit is used for calculating to obtain a farm density characteristic value of the target geographical range according to the total feeding amount and the area of the target geographical range.

Further, the first determination unit includes:

the third determining subunit is used for determining the waterway net length or the waterway net area of the target waterway net according to the second position information of the target waterway net;

and the second determining subunit is used for calculating a waterway net density characteristic value of the target geographic range according to the waterway net length and the area of the target geographic range, or calculating the waterway net density characteristic value of the target geographic range according to the waterway net area and the area of the target geographic range.

Further, the risk prediction model includes preset weight values corresponding to each of the influence data, and the prediction module 30 includes:

the input unit is used for inputting each piece of influence data into the risk prediction model so as to call the risk prediction model to calculate a risk coefficient of the farm to be predicted based on the weight value and each piece of influence data;

and the fourth determining unit is used for taking the risk coefficient as a risk prediction result of the farm to be predicted, or taking the risk grade as a risk prediction result of the farm to be predicted after determining a risk grade according to the risk coefficient and a preset corresponding relation between the coefficient and the grade.

Further, the plant risk prediction device further includes:

the map recognition system comprises an input module, a recognition module and a recognition module, wherein the input module is used for inputting a map to be recognized into an influence factor detection model to obtain the classification category of each pixel point in the map to be recognized, and the influence factor detection model is obtained by pre-training;

the first determining module is used for determining target pixel points belonging to risk influence factors from all the pixel points based on the classification categories;

and the second determining module is used for determining the actual position of each target pixel point according to the scale of the map to be identified and obtaining second position information of each risk influence factor based on the actual position.

Further, the risk influencing factor includes a water network, and the plant risk prediction device further includes:

the system comprises a preliminary training module, a semantic segmentation module and a semantic segmentation module, wherein the preliminary training module is used for carrying out preliminary training on a semantic segmentation model to be trained by adopting positive training data, and the positive training data comprises a plurality of pre-collected remote sensing satellite maps containing a water network and water network annotation data;

the adjusting module is used for adjusting the preliminarily trained semantic segmentation model by adopting negative example training data or adopting the negative example training data and the positive example training data, wherein the negative example training data comprise a plurality of pre-collected remote sensing satellite maps without water network;

a third determining module, which takes the adjusted semantic segmentation model as the influence factor detection model when detecting that the adjusted model to be trained accords with the preset model condition, otherwise, executes the steps based on the adjusted semantic segmentation model: and (4) performing primary training on the semantic segmentation model to be trained by adopting the normal training data.

Further, the plant risk prediction device further includes:

and the output module is used for determining a target prevention and control scheme of the farm to be predicted from preset prevention and control schemes according to the risk prediction result and outputting the target prevention and control scheme.

The expansion content of the specific implementation of the plant risk prediction device of the present invention is basically the same as that of each embodiment of the plant risk prediction method, and is not described herein again.

Furthermore, an embodiment of the present invention further provides a computer-readable storage medium, in which a farm risk prediction program is stored, and when executed by a processor, the farm risk prediction program implements the steps of the farm risk prediction method as described below.

For the embodiments of the plant risk prediction device and the computer-readable storage medium of the present invention, reference may be made to the embodiments of the plant risk prediction method of the present invention, and details are not described herein.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. A risk prediction method for a farm, characterized by comprising the steps of:

2. The method of predicting risk of a farm according to claim 1, wherein the target influencing factors include a target water network and a target farm, and the step of obtaining the influence data corresponding to the target influencing factors includes:

3. The method of predicting risk of a farm according to claim 2, wherein the step of determining a farm density characteristic value of the target geographical range based on the second location information of the target farm and the target geographical range comprises:

4. The method of predicting risk in a farm according to claim 2, wherein the step of determining a waterway network density characteristic value for the target geographical area based on the second location information of the target waterway network and the target geographical area comprises:

5. The method of claim 1, wherein the risk prediction model includes a preset weight value corresponding to each of the impact data, and the step of inputting the impact data into the preset risk prediction model to obtain the risk prediction result of the farm to be predicted includes:

6. The method of predicting risk in a farm according to claim 1, wherein the step of selecting a target influencing factor falling within the target geographical range from the risk influencing factors according to the preset second location information of each risk influencing factor further comprises:

7. The method according to claim 6, wherein the risk influencing factors include a water network, and before the step of inputting the map to be identified into the influencing factor detection model and obtaining the classification category of each pixel point in the map to be identified, the method further comprises:

8. The method for predicting the risk of the farm according to any one of claims 1 to 7, wherein after the step of obtaining the influence data corresponding to the target influence factor and inputting the influence data into a preset risk prediction model to obtain the risk prediction result of the farm to be predicted, the method further comprises:

9. A farm risk prediction device, characterized by comprising:

10. A farm risk prediction apparatus, characterized by comprising: a memory, a processor, and a farm risk prediction program stored on the memory and executable on the processor, the farm risk prediction program when executed by the processor implementing the steps of the farm risk prediction method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a farm risk prediction program that, when executed by a processor, implements the steps of the farm risk prediction method according to any one of claims 1 to 8.