Farm site selection method, device, equipment and computer-readable storage medium

Abstract

The present invention discloses a method, device, equipment and computer-readable storage medium for selecting a site for a breeding farm. The method obtains remote sensing satellite map data of a site selection area, extracts site selection consideration factor information in the site selection area based on the remote sensing satellite map data; determines an alternative location from the site selection area based on preset breeding farm site selection rules and the site selection consideration factor information; and determines the final site selection location of the breeding farm based on the alternative location. The present invention fully considers various factors that affect the site selection of the breeding farm, thereby scientifically planning and designing the address of the breeding farm, and can automatically screen out qualified alternative locations based on computer algorithms, which not only reduces labor costs but also improves the overall site selection efficiency.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and computer-readable storage medium for selecting a site for a breeding farm. Background Art

[0002] With the rapid development of agriculture, high quality and high efficiency are being pursued in all aspects to realize smart agriculture. For example, the breeding industry is gradually becoming intelligent and automated. At present, the site selection method of farms needs to be improved. The existing site selection for farms for breeding poultry and livestock is usually based on human autonomy. The selection of a comprehensive and optimal farm address requires consideration of many factors, resulting in the inefficiency of the existing method of selecting the address of the farm, and it is often impossible to select an address for a farm with excellent results. Summary of the invention

[0003] The main purpose of the present invention is to provide a method, device, equipment and computer-readable storage medium for selecting a breeding farm site, aiming to solve the technical problem that the existing breeding farm site selection method cannot select a comprehensive and optimal breeding farm address and has low site selection efficiency.

[0004] To achieve the above object, the present invention provides a method for selecting a site for a breeding farm, the method comprising the following steps:

[0005] Acquire remote sensing satellite map data of the area to be selected, and extract information on site selection considerations in the area to be selected based on the remote sensing satellite map data;

[0006] Based on the preset farm site selection rules and the site selection consideration information, determine the candidate location from the area to be selected;

[0007] The final site selection location of the breeding farm is determined based on the candidate locations.

[0008] Optionally, the step of extracting the information of site selection consideration factors in the area to be selected based on the remote sensing satellite map data includes:

[0009] Inputting the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain a classification result of each pixel point in the remote sensing satellite map data, wherein the classification result is used to indicate the type of site selection consideration factor to which the corresponding pixel point belongs;

[0010] The site selection consideration factor information within the area to be selected is determined based on the classification result.

[0011] Optionally, the site selection consideration information includes public places and waterway networks.

[0012] Before the step of inputting the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain the classification result of each pixel point in the remote sensing satellite map data, the step further includes:

[0013] The semantic segmentation model to be trained is initially trained using positive training data, wherein the positive training data includes a plurality of pre-collected remote sensing satellite maps containing public places or waterway networks, and waterway network annotation data corresponding to each remote sensing map satellite;

[0014] Adopting negative training data, or using the negative training data and the positive training data to adjust the semantic segmentation model after preliminary training, wherein the negative training data includes a plurality of pre-collected remote sensing satellite maps that do not include public places and waterway networks;

[0015] When it is detected that the adjusted semantic segmentation model does not meet the preset model conditions, the steps are performed again based on the adjusted semantic segmentation model: using positive example training data to perform preliminary training on the semantic segmentation model to be trained;

[0016] When it is detected that the adjusted semantic segmentation model meets the preset model condition, the adjusted semantic segmentation model is used as the consideration factor detection model.

[0017] Optionally, the farm site selection rule includes distance intervals corresponding to various site selection considerations.

[0018] The step of determining the candidate location from the area to be selected based on the preset farm site selection rules and the site selection consideration information includes:

[0019] According to each of the distance intervals and the location information of each location consideration factor in the location consideration factor information, a selectable location range is divided from the to-be-selected location area;

[0020] An alternative position is determined from the range of selectable positions.

[0021] Optionally, the step of determining an alternative position from within the optional position range includes:

[0022] Obtain the aquaculture species of the aquaculture farm to be sited, and determine the aquaculture environment requirements corresponding to the aquaculture species;

[0023] A position that meets the breeding environment requirements is selected from the optional position range as an alternative position.

[0024] Optionally, when there are multiple candidate locations, the step of determining the final site selection location of the farm based on the candidate locations includes:

[0025] Obtain information on factors affecting the spread of epidemic disasters within a preset range around the candidate location;

[0026] Inputting the information of the propagation influencing factors into a preset risk assessment model to obtain the risk level of the candidate location;

[0027] The final site selection location of the breeding farm is determined from the candidate locations based on the risk level of each candidate location.

[0028] Optionally, the information on the factors influencing the spread includes waterway network information and aquaculture information, and the step of inputting the information on the factors influencing the spread into a preset risk assessment model to obtain the risk level of the candidate location includes:

[0029] Determine a waterway network density characteristic value within the preset range according to the waterway network information and the preset range;

[0030] Determine a farm density characteristic value within the preset range according to the breeding information and the preset range;

[0031] The waterway network density characteristic value and the farm density characteristic value are input into the risk assessment model to obtain the risk level of the candidate location.

[0032] In order to achieve the above object, the present invention further provides a farm site selection device, the farm site selection device comprising:

[0033] An acquisition module is used to acquire remote sensing satellite map data of the area to be selected, and extract the site selection consideration factor information in the area to be selected based on the remote sensing satellite map data;

[0034] A determination module is used to determine an alternative location from the area to be selected based on a preset farm site selection rule and the site selection consideration factor information; and to determine the final site selection location of the farm based on the alternative location.

[0035] To achieve the above-mentioned purpose, the present invention also provides a farm site selection device, which includes: a memory, a processor, and a farm site selection program stored in the memory and executable on the processor, and when the farm site selection program is executed by the processor, the steps of the farm site selection method as described above are implemented.

[0036] In addition, to achieve the above-mentioned purpose, the present invention also proposes a computer-readable storage medium, on which a farm site selection program is stored. When the farm site selection program is executed by a processor, the steps of the farm site selection method as described above are implemented.

[0037] In the present invention, by acquiring remote sensing satellite map data of the area to be selected, the site selection consideration factor information in the area to be selected is extracted based on the remote sensing satellite map data; based on the preset farm site selection rules and the site selection consideration factor information, the candidate location is determined from the area to be selected; based on the candidate location, the final site selection location of the farm is determined. The present invention is based on extracting relevant information that has an impact on the breeding of poultry and livestock from the satellite map of the area to be selected as the address consideration factor information of the farm, and then combining the preset farm site selection rules and the address consideration factor information to calculate the various candidate locations suitable as the farm in the area to be selected, and finally determine the final site selection location of the farm from the candidate locations, so as to fully consider the factors affecting the various site selection of the farm, thereby scientifically planning and designing the address selection of the farm, and can automatically screen out qualified candidate locations based on computer algorithms, which not only reduces labor costs but also improves the overall site selection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the structure of the hardware operating environment involved in the embodiment of the present invention;

[0039] Figure 2 It is a schematic diagram of the process of the first embodiment of the method for selecting a breeding farm site according to the present invention;

[0040] Figure 3 It is a functional module schematic diagram of a preferred embodiment of the farm site selection device of the present invention.

[0041] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0042] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0043] like Figure 1 As shown, Figure 1 It is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0044] It should be noted that the farm site selection device in the embodiment of the present invention can be a smart phone, a personal computer, a server and other devices, and no specific limitation is made here.

[0045] like Figure 1As shown, the farm site selection device may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and 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 and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0046] Those skilled in the art will understand that Figure 1 The equipment structure shown in the figure does not constitute a limitation on the farm site selection equipment, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0047] like Figure 1 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module and a farm site selection program. The operating system is a program for managing and controlling the hardware and software resources of the device, and supports the operation of the farm site selection program and other software or programs.

[0048] exist Figure 1 In the device shown, the user interface 1003 is mainly used for data communication with the client; the network interface 1004 is mainly used for establishing a communication connection with the server; and the processor 1001 can be used to call the farm site selection program stored in the memory 1005 and perform the following operations:

[0049] Acquire remote sensing satellite map data of the area to be selected, and extract information on site selection considerations in the area to be selected based on the remote sensing satellite map data;

[0050] Based on the preset farm site selection rules and the site selection consideration information, determine the candidate location from the area to be selected;

[0051] The final site selection location of the breeding farm is determined based on the candidate locations.

[0052] Furthermore, the step of extracting the information of the site selection consideration factors in the area to be selected based on the remote sensing satellite map data includes:

[0053] Inputting the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain a classification result of each pixel point in the remote sensing satellite map data, wherein the classification result is used to indicate the type of site selection consideration factor to which the corresponding pixel point belongs;

[0054] The site selection consideration factor information within the area to be selected is determined based on the classification result.

[0055] Furthermore, the site selection consideration information includes public places and waterway networks.

[0056] Before the step of inputting the remote sensing satellite map data into the pre-trained consideration factor detection model to obtain the classification result of each pixel point in the remote sensing satellite map data, the processor 1001 can also be used to call the farm site selection program stored in the memory 1005 to perform the following operations:

[0057] The semantic segmentation model to be trained is initially trained using positive training data, wherein the positive training data includes a plurality of pre-collected remote sensing satellite maps containing public places or waterway networks, and waterway network annotation data corresponding to each remote sensing map satellite;

[0058] Adopting negative training data, or using the negative training data and the positive training data to adjust the semantic segmentation model after preliminary training, wherein the negative training data includes a plurality of pre-collected remote sensing satellite maps that do not include public places and waterway networks;

[0059] When it is detected that the adjusted semantic segmentation model does not meet the preset model conditions, the steps are performed again based on the adjusted semantic segmentation model: using positive example training data to perform preliminary training on the semantic segmentation model to be trained;

[0060] When it is detected that the adjusted semantic segmentation model meets the preset model condition, the adjusted semantic segmentation model is used as the consideration factor detection model.

[0061] Furthermore, the farm site selection rules include distance intervals corresponding to various site selection considerations.

[0062] The step of determining the candidate location from the area to be selected based on the preset farm site selection rules and the site selection consideration information includes:

[0063] According to each of the distance intervals and the location information of each location consideration factor in the location consideration factor information, a selectable location range is divided from the to-be-selected location area;

[0064] An alternative position is determined from the range of selectable positions.

[0065] Furthermore, the step of determining an alternative position from within the optional position range includes:

[0066] Obtain the aquaculture species of the aquaculture farm to be sited, and determine the aquaculture environment requirements corresponding to the aquaculture species;

[0067] A position that meets the breeding environment requirements is selected from the optional position range as an alternative position.

[0068] Furthermore, when there are multiple candidate locations, the step of determining the final site selection location of the breeding farm based on the candidate locations includes:

[0069] Obtain information on factors affecting the spread of epidemic disasters within a preset range around the candidate location;

[0070] Inputting the information of the propagation influencing factors into a preset risk assessment model to obtain the risk level of the candidate location;

[0071] The final site selection location of the breeding farm is determined from the candidate locations based on the risk level of each candidate location.

[0072] Furthermore, the information on the factors influencing the spread includes waterway network information and aquaculture information, and the step of inputting the information on the factors influencing the spread into a preset risk assessment model to obtain the risk level of the candidate location includes:

[0073] Determine a waterway network density characteristic value within the preset range according to the waterway network information and the preset range;

[0074] Determine a farm density characteristic value within the preset range according to the breeding information and the preset range;

[0075] The waterway network density characteristic value and the farm density characteristic value are input into the risk assessment model to obtain the risk level of the candidate location.

[0076] Based on the above structure, various embodiments of the method for selecting a site for a breeding farm are proposed.

[0077] Reference Figure 2 , Figure 2 This is a schematic flow chart of the first embodiment of the method for selecting a site for a breeding farm according to the present invention.

[0078] The embodiments of the present invention provide embodiments of the method for selecting a site for a farm. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here. The execution subjects of each embodiment of the method for selecting a site for a farm of the present invention may be devices such as smart phones, personal computers, and servers. For ease of description, the execution subjects are omitted in the following embodiments. In this embodiment, the method for selecting a site for a farm includes:

[0079] Step S10, acquiring remote sensing satellite map data of the area to be selected, and extracting information on site selection considerations in the area to be selected based on the remote sensing satellite map data;

[0080] In this embodiment, the address of a farm for raising livestock can be selected in an area. The area can be a city or a district, etc., and is not specifically limited. The area is used as a site selection area for the farm.

[0081] Acquire remote sensing satellite map data of the area to be selected, and extract information on site selection factors that should be considered when selecting a farm in the area to be selected from the remote sensing satellite map data in combination with the actual conditions of the production characteristics, production scale, feeding management methods, epidemic prevention needs, and degree of production intensity of the required poultry and livestock.

[0082] Specifically, for example, the staff who are currently performing the site selection operation for the breeding farm predetermines a certain city as the area to be selected for the breeding farm, and then obtains the remote sensing satellite map data of the city from the existing commonly used map software (such as the map image of the city or the data of each entity in the map), and then combines the actual conditions of various aspects such as the production characteristics, production scale, feeding management methods, epidemic prevention needs and degree of production intensity of the poultry and livestock that need to be raised in the current breeding farm, and combines the pre-trained consideration factor detection model to determine and extract the locations of public places such as residential areas, schools, hospitals, rivers and lakes, and roads (including railways, highways, national roads, provincial roads, county roads, and village roads) that should be considered for breeding farm site selection in the city from the map image of the city or the data of each entity in the map, and selects the information that needs to be considered for the location of the breeding farm for breeding poultry and livestock.

[0083] In another embodiment, if the factors such as the regional topography, topography, soil quality, water source, location of settlements, transportation, electricity, material supply and regional climatic conditions that should be considered for site selection of breeding farms in the city are predetermined, the factor information can be directly obtained through the background of commonly used map software. For example, by entering "settlement" in a certain map software, the location information of all settlements in the city can be obtained from the map software index.

[0084] Step S20, determining an alternative location from the area to be selected based on the preset farm site selection rules and the site selection consideration information;

[0085] The site selection rules for the breeding farm are set in advance according to the types of poultry and livestock to be raised. Based on the pre-set site selection rules for the breeding farm and the information on site selection factors that should be considered for site selection in the area to be selected, various alternative locations that can be used as breeding farms for the required poultry and livestock are determined from the area to be selected.

[0086] It should be noted that, in the present embodiment, the preset farm site selection rules can be specifically set by the staff performing the farm site selection operation, based on the type and scale of the poultry and livestock to be raised, and the distance range rules of different factors can be manually set. Specifically, considering that the poultry and livestock currently raised in the farm need to ensure convenient transportation but prevent the spread of the epidemic through roads, the distance from the alternative location of the farm can be set to 1000 meters to 5000 meters from the road, or, considering the impact of the farm on public places, the distance from the alternative location to the public place can be set to greater than 1000 meters, or, considering the dependence of the farm on water sources but preventing the spread of the epidemic through water sources, the distance from the alternative location to rivers and lakes can be set to 500 meters to 5000 meters. Other distance rule settings are formulated according to the specific poultry and livestock raised.

[0087] Specifically, for example, the locations of public places such as residential areas, schools, hospitals, rivers and lakes, and roads (including railways, highways, national roads, provincial roads, county roads, and village roads) that should be considered for site selection of farms in the city are determined and extracted from a map of the area to be selected - a certain city, or the data of each entity in the map. The locations are input into a computer together with the pre-established site selection rules for farms: the distance from the road is 1,000 to 5,000 meters, the distance from public places is greater than 1,000 meters, and the distance from rivers and lakes is 500 to 5,000 meters. The computer then calculates the locations in the map that satisfy each distance interval in the above-mentioned site selection rules for farms one by one, and then uses the calculated locations that satisfy each distance interval in the above-mentioned site selection rules for farms as candidate locations for breeding this type of livestock.

[0088] Step S30, determining the final site selection location of the breeding farm based on the candidate locations.

[0089] According to the environmental characteristics of each candidate location determined by the test and combined with the field investigation, the final site selection location of the farm for raising poultry and livestock in the area to be selected is determined from the candidate locations. Specifically, it can be detected whether the soil, air and other factors at each candidate location meet the requirements, and then the staff who are performing the current farm site selection operation will conduct a field investigation at the candidate location that meets the requirements, and finally manually determine the site selection location for the farm for raising poultry and livestock.

[0090] In this embodiment, by acquiring remote sensing satellite map data of the determined area to be sited, combined with the actual conditions of the production characteristics, production scale, feeding management methods, epidemic prevention needs and degree of production intensity of the required livestock and poultry, the site selection considerations that should be considered for site selection of the farm in the area to be sited are extracted from the acquired remote sensing satellite map data, and the farm site selection rules are set in advance according to the types of livestock and poultry to be raised. According to the pre-set farm site selection rules and the acquired site selection considerations, various alternative locations that can be used as farms for the required livestock and poultry are determined from the area to be sited, and the final site selection location of the farm for raising livestock and poultry in the area to be sited is determined from among the various alternative locations based on the environmental characteristics of each alternative location determined by detection and combined with field investigations. Based on the farm site selection method in this embodiment, it is possible to fully consider various factors that affect the site selection of the farm, so as to scientifically plan and design the address selection of the farm, and automatically screen out qualified alternative locations based on computer algorithms, which not only reduces labor costs but also improves the overall site selection efficiency and makes the site selection results more appropriate.

[0091] Further, based on the above first embodiment, a second embodiment of the method for selecting a site for a breeding farm of the present invention is proposed. In this embodiment, the above step S10 includes:

[0092] Step S101, inputting the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain a classification result of each pixel point in the remote sensing satellite map data, wherein the classification result is used to indicate the type of site selection consideration factor to which the corresponding pixel point belongs;

[0093] In this embodiment, the method for extracting the information of the site selection consideration factors from the acquired remote sensing satellite map data of the area to be selected may be: a consideration factor detection model may be pre-trained, and the consideration factor detection model is used to identify the site selection consideration factor information in the map. The consideration factor detection model may be a multi-classification model, which is used to identify which of the multiple site selection consideration factor information each pixel point in the map belongs to; the consideration factor detection model may also be multiple binary classification models, which are used to identify whether each pixel point in the map belongs to a certain site selection consideration factor information. The consideration factor detection model may adopt a commonly used image target detection model, such as the semantic image segmentation model DeepLab-v3+.

[0094] The acquired remote sensing satellite map data of the area to be selected is input into the influencing factor detection model to obtain the classification results of each pixel point in the remote sensing satellite map. The classification results are used to indicate the type of site selection consideration factor to which the corresponding pixel point belongs.

[0095] Step S102: determining the site selection consideration factor information in the area to be selected based on the classification result.

[0096] According to the classification results of each pixel point, the target pixel point belonging to the site selection consideration factor information is determined from each pixel point. That is, according to the classification results, it is determined which pixel points belong to the site selection consideration factor information.

[0097] Furthermore, the site selection factor information includes but is not limited to public places and waterway networks. Before the above step S101, it also includes:

[0098] Step S40, using positive training data to perform preliminary training on the semantic segmentation model to be trained, wherein the positive training data includes a plurality of pre-collected remote sensing satellite maps containing public places or waterway networks, and waterway network annotation data corresponding to each remote sensing map satellite;

[0099] In this embodiment, the site selection factor information includes but is not limited to public places and waterway networks. The factor detection model can be a multi-classification model composed of a semantic segmentation model. The output results of the influencing factor detection model are preset to be three categories: not belonging to the site selection factor information, public places, and waterway networks. The training method of the factor detection model can be as follows:

[0100] Collect multiple remote sensing satellite maps containing public places and waterway networks in advance, and collect the annotation data of public places and waterway networks in the remote sensing satellite maps, that is, the waterway network annotation data and the public place annotation data. The public place annotation data and the waterway network annotation data can be a mask map corresponding to the remote sensing satellite image. The mask map contains the classification category corresponding to each pixel point of the remote sensing satellite image, such as 0 indicates that the pixel point does not belong to the site selection factor information, 1 indicates that the pixel point belongs to the public place, and 2 indicates that the pixel point belongs to the waterway network. The mask map can use different colors to represent different categories. Use multiple remote sensing satellite maps containing public places and waterway networks, as well as the waterway network annotation data and public place annotation data as positive training data.

[0101] First, the positive training data is used to perform preliminary training on the semantic segmentation model to be trained. The preliminary training process can adopt the existing machine learning model training process.

[0102] Step S50, using negative training data, or using the negative training data and the positive training data to adjust the semantic segmentation model after preliminary training, wherein the negative training data includes a plurality of pre-collected remote sensing satellite maps that do not include public places and waterway networks;

[0103] Collect multiple remote sensing satellite maps that do not contain public places and waterway networks in advance as negative training data. The ratio of negative training data to positive training data can be set according to actual experience. For example, the negative training data can be one-tenth of the positive training data. After the preliminary training, the negative training data and the above-mentioned positive training data are used together to adjust the semantic segmentation model after preliminary training, or the negative training data is used alone to adjust the semantic segmentation model after preliminary training. The adjustment at this time can be fine-tuning. Specifically, fine-tuning is to use positive training data and negative training data together for training, and the training process can also use the training process of the existing machine learning model. It should be noted that during fine-tuning, the hyperparameters of the model can be adjusted according to the ratio of positive training data to negative training data, and fine-tuning training can be performed after the adjustment. For example, when the negative training data can be one-tenth of the positive training data, the learning rate in the fine-tuning stage can also be set to one-tenth of the preliminary training stage.

[0104] Step S60, when it is detected that the adjusted semantic segmentation model does not meet the preset model condition, the steps are executed again based on the adjusted semantic segmentation model: using positive training data to perform preliminary training on the semantic segmentation model to be trained;

[0105] Step S70, when it is detected that the adjusted semantic segmentation model meets the preset model condition, the adjusted semantic segmentation model is used as the consideration factor detection model.

[0106] Check whether the adjusted semantic segmentation model meets the preset model conditions. If it meets the preset model conditions, the adjusted semantic segmentation model is used as a consideration factor detection model so that the consideration factor detection model can be used to extract the site selection consideration factor information later. Otherwise, if it does not meet the preset model conditions, the adjusted semantic segmentation model is preliminarily trained with positive example training data, and then adjusted training is performed until it is detected that it meets the preset model conditions. Among them, the preset model conditions can be conditions set in advance according to the performance requirements of the model, such as the convergence of the model's loss function can be used as a condition, or commonly used objective indicators of detection model performance can be used as conditions, such as accuracy, recall rate and IOU (Intersection over Union, intersection ratio) and other objective indicators. Among them, the calculation method of each objective indicator can refer to the existing indicator calculation method, which will not be described in detail here.

[0107] In this embodiment, by pre-training a consideration factor detection model, the remote sensing satellite map data of the area to be selected is detected and identified, and the information of each location consideration factor in the remote sensing satellite map data is extracted, so that the information of each location consideration factor of the area to be selected can be intelligently and quickly extracted, thereby improving the efficiency of farm site selection. In addition, for some data of location consideration factor information that cannot be extracted from the map data, it can be detected by the consideration factor detection model, so that the data of risk influencing factors is more comprehensive, and then the location of the farm is more accurate and comprehensive, and the final location of the farm is more comprehensive.

[0108] In addition, by first using positive training data to preliminarily train the model, and then using both positive training data and negative training data to fine-tune the preliminarily trained model, that is, using negative examples to correct the deviation and reduce the false detection of the model. In addition, by fine-tuning the model with both positive training data and negative training data, it is avoided that the training effect of the positive training data is covered by using only negative training data.

[0109] Further, based on the above first embodiment, a third embodiment of the method for selecting a site for a breeding farm of the present invention is proposed. In this embodiment, the above step S20 includes:

[0110] Step S201, dividing the selectable location range from the to-be-selected location area according to the distance intervals and the location information of each location consideration factor in the location consideration factor information;

[0111] Step S202: determining an alternative position from the optional position range.

[0112] It should be noted that, in this embodiment, the pre-established farm site selection rules include distance intervals corresponding to various site selection considerations.

[0113] By combining the information of various site selection considerations extracted from the remote sensing satellite image data of the area to be selected, and the distance intervals corresponding to the various site selection considerations in the pre-set farm site selection rules, the overlapping areas of the distance intervals corresponding to the various site selection considerations in the area to be selected are determined, and the overlapping areas are used as the optional location range for farms that can be used for breeding poultry and livestock. Then, from the optional location range, various alternative locations that can be used as farms for breeding poultry and livestock are determined.

[0114] Further, step S202 includes:

[0115] Step S2021, obtaining the breeding type of the breeding farm to be selected, and determining the breeding environment requirements corresponding to the breeding type;

[0116] Step S2022: Select a position that meets the breeding environment requirements from the optional position range as an alternative position.

[0117] Specifically, for example, the staff currently performing the site selection operation for a breeding farm determines that the type of poultry and livestock that currently needs to be bred is "fish". The detection determines that the environmental requirements for breeding "fish" are: healthy water quality, sufficient sunlight time, and product transportation routes, etc. According to the determined environmental requirements for breeding "fish", from the overlapping areas of the distance intervals corresponding to the site selection considerations such as various settlements, schools, hospitals and other public places, rivers and lakes, and roads, various locations that meet the above-mentioned environmental requirements such as "healthy water quality, sufficient sunlight time, and product transportation routes" are selected as the various alternative locations for the breeding farm for the current poultry and livestock "fish" that need to be bred.

[0118] In this embodiment, by combining the information of each site selection consideration factor extracted from the remote sensing satellite image data of the area to be selected, and the distance interval corresponding to each site selection consideration factor in the preset farm site selection rules, the overlapping area of ​​the distance interval corresponding to each site selection consideration factor in the area to be selected is determined, and the overlapping area is used as the optional location range of the farm that can be used for breeding poultry and livestock, and then according to the breeding environment requirements corresponding to the breeding type of the farm to be selected, the various alternative locations that meet the corresponding breeding environment requirements and can be used as the breeding farm for the required breeding of poultry and livestock are determined from the determined optional location range. It is realized that according to the site selection consideration factors such as various public places, rivers and lakes, and roads, the corresponding distance interval in the preset farm site selection rules, and the breeding environment requirements corresponding to the specific types of livestock and poultry, the various alternative locations that can be used as the farm are selected, which ensures the accuracy and comprehensiveness of the final farm site selection and improves the efficiency of farm site selection.

[0119] Further, based on the first, second and third embodiments above, a fourth embodiment of the method for selecting a site for a breeding farm of the present invention is proposed. In this embodiment, when there are multiple candidate locations, the step S30 includes:

[0120] Step S301, obtaining information on factors affecting the spread of epidemic disasters within a preset range around the candidate location;

[0121] It should be noted that, in this embodiment, a preset range around the farm to be predicted can be determined first. There are many ways to determine the preset range; for example, a radius value can be set in advance, and a circular area can be divided with the candidate position as the center and the radius value as the radius, which can be used as the preset range of the candidate position. According to the actual geographical location of the candidate position and the radius value, the actual geographical location of the preset range can be determined; it can also be that the candidate position is used as the center of the rectangle, and the matrix side length value is set in advance, so as to divide a matrix area as the preset range of the candidate position; there are other feasible methods. According to different actual application scenarios, different methods can be adopted to determine the preset range, which are not listed here one by one.

[0122] According to the actual geographical location within a preset range around each determined alternative location and the pre-defined information on factors affecting the spread of epidemic disasters, the information on factors affecting the spread of epidemic disasters contained in the preset range around each determined alternative location is detected.

[0123] Specifically, for example, when the actual geographical location of the preset range of the determined alternative locations is expressed in longitude and latitude (the location of the preset range can be composed of a longitude interval and a latitude interval), it is necessary to determine whether the longitude of the defined factors affecting the spread of the epidemic disaster falls within the longitude interval, and whether the latitude falls within the latitude interval. If both fall within, it is determined that the preset range of the alternative locations includes the factors affecting the spread of the current epidemic disaster.

[0124] It should be noted that in this embodiment, various factors affecting the spread of epidemic disasters can be pre-specified, such as farms, water networks, road networks, and villages, etc. These are different types of factors affecting the spread of epidemic disasters. Among them, the water network can refer to rivers and lakes of various sizes, etc., and the road network can refer to various types of roads, such as national highways and expressways.

[0125] Step S302, inputting the information of the propagation influencing factors into a preset risk assessment model to obtain the risk level of the candidate location;

[0126] Step S303: determining the final site selection location of the breeding farm from the candidate locations based on the risk level of each candidate location.

[0127] Specifically, for example, a risk assessment model can be set in advance. The risk assessment model can be a model structure such as a linear model or a neural network model. If it is a neural network model, the neural network model needs to be trained in advance, and the model obtained after training is used for risk assessment. The training process of the neural network model can refer to the existing neural network model training process, which will not be described in detail here. The setting principle of the risk assessment model is: when the input data represents the information of the factors affecting the spread of the epidemic disaster within the preset range around the candidate location, the more and denser the information, the higher the risk of the farm to be predicted. The input data of the risk assessment model can be set according to the situation. For example, the input data can be set to the total area of ​​the farm and the total area of ​​the waterway network within the preset range around the candidate location. The information on the factors affecting the spread of the epidemic disaster can be processed into a form that matches the input data of the risk assessment model, and then the processed data is input into the risk assessment model. After the risk assessment model processes the data, the risk level of the candidate location is output. The risk level corresponding to various risk levels can be pre-specified. For example, the level is 0 to 10, and the risk level gradually increases from 0 to 10. Based on the output risk levels of the candidate locations, the candidate location with the lowest risk level is selected from the candidate locations as the final location of the breeding farm for raising poultry and livestock in the area to be selected.

[0128] Further, in this embodiment, the information of factors affecting the spread of epidemic disasters includes but is not limited to waterway network information and aquaculture information. The above step S302 includes:

[0129] Step S3021, determining a waterway network density characteristic value within the preset range according to the waterway network information and the preset range;

[0130] Further, in this embodiment, the information of factors affecting the spread of epidemic disasters may be waterway network information. Then, the risk of the farm to be predicted is predicted based on the breeding information and waterway network information within a preset range around the candidate location.

[0131] Specifically, the waterway network density characteristic value of the preset range around the candidate position is determined according to the waterway network information and the preset range. The waterway network density characteristic value may be the result obtained by dividing the total length of the waterway network within the preset range by the total area of ​​the preset range, or may be the result obtained by dividing the total area of ​​the waterway network within the preset range by the total area of ​​the preset range. The calculation process of the area and length may be as follows: the total area of ​​the preset range may be determined according to the location information of the preset range, such as the total area of ​​the preset range may be calculated according to the longitude and latitude interval of the preset range; the length of each waterway network may be determined according to the location information of each waterway network in the waterway network information, for example, a river or a road may be represented by the longitude and latitude of two points, in which case the length of the river or road may be calculated according to the longitude and latitude of the two points, and the lengths of all rivers or roads may be added to obtain the length of the target waterway network; the total area of ​​the waterway network may also be determined according to the location information of each waterway network in the waterway network information, for example, a river or a road may be represented by a longitude and latitude interval, in which case the area of ​​the river or road may be calculated according to the longitude and latitude interval, and the areas of all rivers or roads may be added to obtain the total area of ​​the waterway network.

[0132] Step S3022, determining a farm density characteristic value within the preset range according to the breeding information and the preset range;

[0133] The farm density characteristic value of the preset range is determined according to the breeding information and the preset range around the candidate location. Specifically, the farm density characteristic value can be the result obtained by dividing the total area of ​​the farms in the preset range by the total area of ​​the preset range. The total area of ​​the farms in the preset range can be calculated by: each farm can be represented by a longitude and latitude interval, and the area of ​​each farm is calculated according to the longitude and latitude interval, and then the areas are added together to obtain the total area of ​​the target farm.

[0134] The farm density characteristic value may also be the result obtained by dividing the total feeding amount of farms in a preset range by the total area of ​​the preset range, that is, the farm density characteristic value may be the feeding density. The total feeding amount of the target farm may be calculated by determining the area of ​​each farm according to the location information of each farm in the breeding information, that is, the above-mentioned total area calculation process of the farm. The corresponding relationship between the area of ​​the farm and the feeding amount may be set in advance, for example, it is set that one pig is produced in 1.5 square meters. Based on the calculated area of ​​each farm and the corresponding relationship, the total feeding amount of each farm is determined. For example, if the area of ​​the farm is 150 square meters, according to the above example, the total feeding amount of each farm is 100 pigs; the total feeding amount is divided by the total area, and the result obtained is used as the farm density characteristic value.

[0135] Step S3023: input the waterway network density characteristic value and the farm density characteristic value into the risk assessment model to obtain the risk level of the candidate location.

[0136] The characteristic values ​​of waterway network density and farm density are input into the risk assessment model to obtain the risk level of the alternative locations.

[0137] In this embodiment, the farm density characteristic values ​​and waterway network density characteristic values ​​within a preset range around the candidate farm site selection are calculated, the farm density characteristic values ​​and waterway network density characteristic values ​​are input into the risk level, and the risk level of the alternative location is calculated. The risk of the candidate farm site selection location is evaluated based on the density characteristics of the farms and waterway networks within a preset range around the alternative location. The actual infectious characteristics of infectious diseases of farmed livestock are combined to comprehensively consider the site selection of farms for breeding poultry and livestock, further increasing the comprehensiveness of farm site selection and improving the efficiency of farm site selection.

[0138] In addition, the present invention also provides a farm site selection device, referring to Figure 3 , the farm site selection device comprises:

[0139] An acquisition module 10 is used to acquire remote sensing satellite map data of the area to be selected, and extract information on site selection considerations in the area to be selected based on the remote sensing satellite map data;

[0140] The determination module 20 is used to determine the candidate locations from the to-be-selected site area based on the preset farm site selection rules and the site selection consideration factor information; and determine the final site selection location of the farm based on the candidate locations.

[0141] Furthermore, the confirmation module 20 includes:

[0142] A first computing unit is used to input the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain a classification result of each pixel point in the remote sensing satellite map data, wherein the classification result is used to indicate the type of site selection consideration factor to which the corresponding pixel point belongs;

[0143] The first determining unit is used to determine the site selection consideration factor information in the to-be-selected site area based on the classification result.

[0144] Furthermore, the site selection factor information includes public places and waterway networks, and the confirmation module 20 also includes:

[0145] A first training unit is used to perform preliminary training on the semantic segmentation model to be trained using positive training data, wherein the positive training data includes a plurality of pre-collected remote sensing satellite maps containing public places or waterway networks, and waterway network annotation data corresponding to each remote sensing map satellite;

[0146] A second training unit is used to adjust the semantic segmentation model after preliminary training by using negative training data, or by using the negative training data and the positive training data, wherein the negative training data includes a plurality of pre-collected remote sensing satellite maps that do not include public places and waterway networks;

[0147] The loop unit is used to execute the steps again based on the adjusted semantic segmentation model when it is detected that the adjusted semantic segmentation model does not meet the preset model conditions: using positive example training data to perform preliminary training on the semantic segmentation model to be trained.

[0148] The model confirmation unit is used to use the adjusted semantic segmentation model as the consideration factor detection model when it is detected that the adjusted semantic segmentation model meets the preset model condition.

[0149] Furthermore, the farm site selection rule includes distance intervals corresponding to various site selection considerations, and the confirmation module 20 also includes:

[0150] A division unit, configured to divide the selectable location range from the to-be-selected location area according to each of the distance intervals and the location information of each of the location consideration factors in the location consideration factor information;

[0151] The second determining unit is configured to determine an alternative position from within the optional position range.

[0152] Further, the second determining unit includes:

[0153] The first acquisition unit is used to acquire the breeding type of the breeding farm to be selected and determine the breeding environment requirements corresponding to the breeding type;

[0154] A selection unit is used to select a position that meets the breeding environment requirements from the optional position range as an alternative position.

[0155] Furthermore, when there are multiple candidate positions, the confirmation module 20 further includes:

[0156] A second acquisition unit is used to acquire information on factors affecting the spread of epidemic disasters within a preset range around the candidate location;

[0157] A second calculation unit is used to input the information of the propagation influencing factors into a preset risk assessment model to obtain the risk level of the candidate location;

[0158] The third determining unit is used to determine the final site selection location of the breeding farm from the candidate locations based on the risk level of each candidate location.

[0159] Furthermore, the propagation influencing factor information includes waterway network information and aquaculture information, and the second calculation unit includes:

[0160] A first determining subunit, configured to determine a waterway network density characteristic value within the preset range according to the waterway network information and the preset range;

[0161] A second determining subunit is used to determine a farm density characteristic value within the preset range according to the breeding information and the preset range;

[0162] An input subunit is used to input the waterway network density characteristic value and the farm density characteristic value into the risk assessment model to obtain the risk level of the candidate location.

[0163] The expanded contents of the specific implementation manner of the breeding farm site selection device of the present invention are basically the same as the various embodiments of the above-mentioned breeding farm site selection method, and will not be elaborated here.

[0164] In addition, an embodiment of the present invention further proposes a computer-readable storage medium, on which a farm site selection program is stored. When the farm site selection program is executed by a processor, the steps of the farm site selection method described below are implemented.

[0165] The various embodiments of the breeding farm site selection device and computer-readable storage medium of the present invention can all refer to the various embodiments of the breeding farm site selection method of the present invention, and will not be repeated here.

[0166] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0167] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0168] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0169] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for selecting a site for a breeding farm, characterized in that: The method for selecting a site for a breeding farm comprises the following steps: Acquire remote sensing satellite map data of the area to be selected, and extract information on site selection considerations in the area to be selected based on the remote sensing satellite map data; Based on the preset farm site selection rules and the site selection consideration information, determine the candidate location from the area to be selected; Determining the final site location of the breeding farm based on the candidate locations; The step of extracting the information of the site selection consideration factors in the area to be selected based on the remote sensing satellite map data comprises: Inputting the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain a classification result of each pixel point in the remote sensing satellite map data, wherein the classification result is used to indicate the type of site selection consideration factor to which the corresponding pixel point belongs; Determine the site selection consideration factor information within the to-be-selected site area based on the classification result; The site selection consideration factor information includes public places and waterway networks. Before the step of inputting the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain the classification result of each pixel point in the remote sensing satellite map data, the step further includes: The semantic segmentation model to be trained is initially trained using positive training data, wherein the positive training data includes a plurality of pre-collected remote sensing satellite maps containing public places or waterway networks, and waterway network annotation data corresponding to each remote sensing map satellite; Adopting negative training data, or using the negative training data and the positive training data to adjust the semantic segmentation model after preliminary training, wherein the negative training data includes a plurality of pre-collected remote sensing satellite maps that do not include public places and waterway networks; When it is detected that the adjusted semantic segmentation model does not meet the preset model conditions, the steps are performed again based on the adjusted semantic segmentation model: using positive example training data to perform preliminary training on the semantic segmentation model to be trained; When it is detected that the adjusted semantic segmentation model meets the preset model condition, the adjusted semantic segmentation model is used as the consideration factor detection model.

2. The method for selecting a breeding farm site according to claim 1, characterized in that: The farm site selection rules include the distance intervals corresponding to the site selection factors. The step of determining the candidate location from the area to be selected based on the preset farm site selection rules and the site selection consideration information includes: According to each of the distance intervals and the location information of each location consideration factor in the location consideration factor information, a selectable location range is divided from the to-be-selected location area; An alternative position is determined from the range of selectable positions.

3. The method for selecting a breeding farm site according to claim 2, characterized in that: The step of determining an alternative position from within the range of optional positions comprises: Obtain the aquaculture species of the aquaculture farm to be sited, and determine the aquaculture environment requirements corresponding to the aquaculture species; A position that meets the breeding environment requirements is selected from the optional position range as an alternative position.

4. The method for selecting a breeding farm site according to any one of claims 1 to 3, characterized in that: When there are multiple candidate locations, the step of determining the final site selection location of the farm based on the candidate locations includes: Obtain information on factors affecting the spread of epidemic disasters within a preset range around the candidate location; Inputting the information of the propagation influencing factors into a preset risk assessment model to obtain the risk level of the candidate location; The final site selection location of the breeding farm is determined from the candidate locations based on the risk level of each candidate location.

5. The method for selecting a breeding farm site according to claim 4, characterized in that: The information of the factors influencing the spread includes waterway network information and aquaculture information. The step of inputting the information of the factors influencing the spread into a preset risk assessment model to obtain the risk level of the candidate location includes: Determine a waterway network density characteristic value within the preset range according to the waterway network information and the preset range; Determine a farm density characteristic value within the preset range according to the breeding information and the preset range; The waterway network density characteristic value and the farm density characteristic value are input into the risk assessment model to obtain the risk level of the candidate location.

6. A farm site selection device, characterized in that: The breeding farm site selection device comprises: An acquisition module is used to acquire remote sensing satellite map data of the area to be selected, and extract information on site selection considerations in the area to be selected based on the remote sensing satellite map data; A determination module, used to determine candidate locations from the to-be-selected site area based on preset farm site selection rules and the site selection consideration information; and determine the final site selection location of the farm based on the candidate locations; The determination module comprises: A first computing unit is used to input the remote sensing satellite map data into a pre-trained consideration factor detection model to obtain a classification result of each pixel point in the remote sensing satellite map data, wherein the classification result is used to indicate the type of site selection consideration factor to which the corresponding pixel point belongs; A first determining unit, configured to determine the site selection consideration factor information in the to-be-selected site area based on the classification result; The site selection factor information includes public places and waterway networks, and the determination module also includes: A first training unit is used to perform preliminary training on the semantic segmentation model to be trained using positive training data, wherein the positive training data includes a plurality of pre-collected remote sensing satellite maps containing public places or waterway networks, and waterway network annotation data corresponding to each remote sensing map satellite; A second training unit is used to adjust the semantic segmentation model after preliminary training by using negative training data, or by using the negative training data and the positive training data, wherein the negative training data includes a plurality of pre-collected remote sensing satellite maps that do not include public places and waterway networks; A loop unit, for performing the steps of: performing preliminary training on the semantic segmentation model to be trained using positive training data, based on the adjusted semantic segmentation model when it is detected that the adjusted semantic segmentation model does not meet the preset model conditions; The model confirmation unit is used to use the adjusted semantic segmentation model as the consideration factor detection model when it is detected that the adjusted semantic segmentation model meets the preset model condition.

7. A farm site selection device, characterized in that: The farm site selection device comprises: a memory, a processor and a farm site selection program stored in the memory and executable on the processor. When the farm site selection program is executed by the processor, the steps of the farm site selection method as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a farm site selection program, which, when executed by a processor, implements the steps of the farm site selection method according to any one of claims 1 to 5.

Images