Ecological orchard planting management method based on symbiotic mechanism of natural enemy insects and nectar plants

Abstract

The present application relates to the technical field of ecological orchard planting, and particularly to an ecological orchard planting management method based on the symbiotic mechanism of natural enemy insects and nectar plants. The size of insect pests in the global planting area of the ecological orchard and the required natural enemy insect species are obtained, a real-time insect pest distribution hotspot map of the global planting area is constructed, the hotspot color condition of the insect pest size that reaches the expected control and extermination rate in the real-time insect pest distribution hotspot map is analyzed to determine the termination magnitude of the natural enemy insects, a corresponding structural feature dislocation matrix is established in the variation nectar plant distribution map according to the characteristic dislocation amplitude, the variation probability of causing the structural variation of the nectar plants in the planting process is calculated based on the structural feature dislocation matrix and a plurality of planting data, and the ecological orchard is planted and managed through the variation probability. The present application can greatly improve the planting quality and income of the nectar plants by reasonably planting and managing the nectar plants in the ecological orchard based on the symbiotic mechanism between the natural enemy insects and the nectar plants.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological orchard planting, and in particular to an ecological orchard planting management method based on the symbiotic mechanism of natural enemy insects and nectar-producing plants. Background Art

[0002] Natural enemy insects, such as predatory ladybugs and parasitic wasps, are important biological control resources in orchard ecosystems. However, their survival and reproduction often rely on resources such as nectar and pollen provided by nectar-producing plants. Nectar-producing plants not only provide food for natural enemy insects but also improve the orchard's ecological environment through their diverse plant structures and functions. Introducing nectar-producing plants into orchards can promote the colonization and spread of natural enemy insects, enhancing their ability to control pests while reducing the use of chemical pesticides and improving the orchard's biodiversity and ecological services.

[0003] Traditional orchard management relies primarily on chemical pest control, which, while effective in the short term, is unsustainable in the long term. Consequently, nectar plants are planted in fixed quantities in orchards. However, existing orchard planting and management systems are often irrational in their management of nectar plants, causing them to mutate and significantly reducing their production of nectar, pollen, and other nectar-producing resources. This, in turn, attracts a large number of parasitic pests. Furthermore, natural enemy insects introduced to combat these pests struggle to meet their energy needs to survive and inhabit these nectar plants, making it difficult to achieve the basic symbiotic relationship between natural enemy insects and nectar plants. Consequently, the survival rate of nectar plants in orchards plummets, making it impossible to maintain high-quality orchard production. Summary of the Invention

[0004] The present invention overcomes the deficiencies of the prior art and provides an ecological orchard planting and management method based on the symbiotic mechanism of natural enemy insects and nectar-producing plants.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] The first aspect of the present invention provides an ecological orchard planting and management method based on the symbiotic mechanism between natural enemy insects and nectar plants, comprising the following steps:

[0007] S102: Obtain the scale of pests and the required natural enemy insect species in the global planting area of ​​the ecological orchard, construct a real-time pest distribution hotspot map for the global planting area, analyze the color of hotspots within the real-time pest distribution hotspot map when the pest control scale reaches the expected pest control rate, and determine the natural enemy insect termination level;

[0008] S104: Practically planting nectar-producing plants according to the termination level of natural enemy insects to construct a multidimensional spatial density model of natural enemy insects, and calculating and analyzing whether the distribution density of natural enemy insects in the global planting area during the practical pest control meets the ideal distribution density based on the multidimensional spatial density model, thereby obtaining first-day and second-day natural enemy insect distribution analysis results;

[0009] S106: When a sensitive planting area exists in the global planting area of ​​the ecological orchard, a minimum nectar source concentration generation interval of the nectar source plant is set according to the expected intake concentration of the natural enemy insect termination level in the sensitive planting area, and the real-time nectar source concentration generation interval of the part of interest is obtained. The minimum nectar source concentration generation interval and the real-time nectar source concentration generation interval are overlapped and judged by a concentration bubble model to determine whether the structural variation of the nectar source plant occurs;

[0010] S108: Obtain a distribution map of variant nectar plants in sensitive planting areas through image data comparison and calculation, establish a corresponding structural feature dislocation matrix in the distribution map of variant nectar plants according to the feature dislocation amplitude, calculate the variation probability of the structural variation of nectar plants during the planting process based on the structural feature dislocation matrix and a number of planting data, and manage the ecological orchard through the variation probability to obtain a planting management strategy for the ecological orchard.

[0011] More specifically, it is characterized in that the step S102 specifically includes the following steps:

[0012] Obtain the global planting area of ​​the ecological orchard, and use the Internet of Things to control the planting monitoring cameras in the ecological orchard to capture and shoot the global planting area in real time, so as to obtain real-time image data of the surfaces of all nectar-producing plants in the global planting area;

[0013] Introducing the ORB image extraction algorithm to perform feature calculation and extraction on the real-time image data corresponding to each nectar-producing plant in the global planting area, thereby obtaining a real-time pest feature vector existing on the surface of each nectar-producing plant in the global planting area;

[0014] Quantitatively counting the number of insect pests corresponding to the real-time insect pest feature vectors existing on the surface of each nectar-producing plant in the global planting area to obtain the scale of insect pests in the global planting area;

[0015] Obtaining a garden pest control knowledge graph based on a big data network, identifying real-time pest feature vectors through the garden pest control knowledge graph, and outputting the natural enemy insect species required to control the pests;

[0016] Constructing a hotspot map, performing a fitting operation on the pest scale of the global planting area in the hotspot map to generate a real-time pest distribution hotspot map of the global planting area, obtaining a planting distribution pattern diagram of nectar-producing plants in the global planting area, and dividing the real-time pest distribution hotspot map into N uniform sub-hotspot map areas based on the planting distribution pattern diagram;

[0017] Obtain the first hotspot color index displayed in each sub-hotspot map area, create a hotspot color rule query table for the natural enemy insect control required for different pest scales, obtain the second hotspot color index representing the necessary pest scale through the hotspot color rule query table, and determine the termination level of the natural enemy insect control for the pest scale of the global planting area based on the difference between the first hotspot color index and the second hotspot color index.

[0018] More specifically, it is characterized in that in the step S102, a first hot spot color index representing the display in each sub-hot spot map area is obtained, a hot spot color rule query table for natural enemy insect control required for different pest scales is created, a second hot spot color index representing the necessary pest scale is obtained through the hot spot color rule query table, and a termination level of natural enemy insect control for the pest scale of the global planting area is determined based on the difference between the first hot spot color index and the second hot spot color index, which specifically includes the following steps:

[0019] Obtaining the hotspot color index displayed in each sub-hotspot map area, defined as the first hotspot color index, and extracting the real-time environmental parameters of each sub-hotspot map area;

[0020] Preset the initial control level of natural enemy insect species based on the scale of pests in the global planting area, and obtain the historical control and extermination rates of natural enemy insect species under different combinations of environmental parameters and different pest scales based on the initial control level of natural enemy insect species based on the big data network;

[0021] Based on the historical pest control rates of the natural enemy insect species at different pest scales under different combinations of environmental parameter conditions, an association rule is used to create a hotspot color rule query table for natural enemy insect control required for different pest scales.

[0022] Obtain the established planting requirements of the ecological orchard, obtain the expected pest control rate of the global planting area based on the established planting requirements, and traverse the search in the hotspot color rule query table to obtain the hotspot color range that meets the expected pest control rate under the real-time environmental parameter conditions of each sub-hotspot map area;

[0023] Obtaining the hotspot color of each sub-hotspot map region in the hotspot color domain, wherein the hotspot color represents the initial control level of the natural enemy insect species and the pest scale required to achieve the desired control and pest elimination rate when killing the pests under the corresponding real-time environmental parameters in the sub-hotspot map region, and is defined as the required pest scale;

[0024] Extract the hotspot color index of the necessary pest scale and define it as the second hotspot color index. Calculate the difference between the first hotspot color index and the corresponding second hotspot color index one by one to obtain the hotspot color index difference of each sub-hotspot map area.

[0025] The initial control level of natural enemy insect species is adjusted by the difference in hotspot color index of all sub-hotspot map areas and a weighted sum is performed to finally determine the termination level of natural enemy insects for controlling the scale of pests in the global planting area.

[0026] More specifically, it is characterized in that the step S104 specifically includes the following steps:

[0027] Pest control was carried out in the global planting area of ​​the ecological orchard by measuring the termination level of natural enemy insects, and a multidimensional density model of the global planting area was constructed;

[0028] In the process of pest control, the spatial density distribution of pests killed and inhabited by natural enemy insects in the global planting area is recorded to obtain spatial density distribution data of natural enemy insects. The spatial density distribution data is fitted by a multidimensional density model to obtain a multidimensional spatial density model of natural enemy insects.

[0029] Extracting a spatial density distribution model diagram of natural enemy insects within a preset prevention and control period from the multidimensional spatial density model, and dividing the spatial density distribution model diagram into a plurality of sub-model diagrams based on a schematic diagram of the planting distribution pattern;

[0030] Obtaining a practical density kernel function expressing the corresponding spatial density distribution of natural enemy insects in each sub-model graph, introducing a trade-off algorithm, calculating the gain weight of each sub-model graph in the trade-off algorithm based on the scale of the pest, taking the expected pest control rate as the allocation premise, and allocating the pest control rate to each sub-model graph according to the gain weight under the allocation premise to obtain the sub-expected pest control rate of each sub-model graph;

[0031] Based on the sub-expected pest control rate, an ideal density kernel function range of the distribution of natural enemy insects for the practical pest control of nectar plants in each sub-model diagram is preset, and whether the practical density kernel function in each sub-model diagram is within the corresponding ideal density kernel function range is determined one by one;

[0032] If the practical density kernel function in the sub-model graph is within the range of the corresponding ideal density kernel function, the nectar plant planting area corresponding to the sub-model graph is calibrated as the accurate planting area, and the first day enemy insect distribution analysis result is obtained;

[0033] If the practical density kernel function in the sub-model graph is not within the range of the corresponding ideal density kernel function, the nectar plant planting area corresponding to the sub-model graph is calibrated as a sensitive planting area, and the enemy insect distribution analysis results of the second day are obtained.

[0034] More specifically, it is characterized in that the step S106 specifically includes the following steps:

[0035] When there are sensitive planting areas in the global planting area of ​​the ecological orchard, the symbiotic mechanism between natural enemy insect species and nectar-producing plant species can be obtained through big data;

[0036] Based on the analysis of the symbiotic mechanism between natural enemy insect species and nectar plant species, the natural enemy insect termination level is obtained, and the nectar intake concentration provided by the nectar plant in the process of killing pests to achieve the expected control and pest control rate under the corresponding real-time environmental parameters in the sensitive planting area is obtained, and multiple key nectar intake concentrations are obtained;

[0037] Constructing a nectar source intake concentration gradient for the process of killing natural enemy insects in real time within the global planting area of ​​the ecological orchard based on the distribution of the multiple key nectar source intake concentrations, and setting a minimum nectar source concentration generation interval for nectar plants based on the nectar source intake concentration gradient;

[0038] Creating a concentration bubble model, calculating and fitting the minimum nectar source concentration generation interval in the concentration bubble model to obtain a minimum nectar source concentration bubble model, which is marked as a first nectar source concentration bubble model;

[0039] Obtaining the variety information of the nectar-producing plants, searching the big data network for the parts of the nectar-producing plants that the natural enemy insects are interested in when feeding on the nectar-producing plants based on the natural enemy insect species and the variety information of the nectar-producing plants, and simultaneously obtaining the real-time characteristics of the plant structure of the parts of interest;

[0040] Through the comprehensive plant knowledge graph, the real-time feature recognition query of the plant structure is performed to obtain the nectar source concentration generation interval generated when the part of interest is under the real-time environmental parameter conditions corresponding to the sensitive planting area, which is defined as the real-time nectar source concentration generation interval;

[0041] Calculating and fitting the real-time nectar source concentration generation interval in the concentration bubble model to obtain a real-time nectar source concentration bubble model, which is marked as a second nectar source concentration bubble model;

[0042] If the second nectar source concentration bubble model can completely cover the first nectar source concentration bubble model, the real-time plant structure feature corresponding to the second nectar source concentration bubble model is calibrated to be a normal plant structure feature;

[0043] If the second nectar source concentration bubble model cannot completely cover the first nectar source concentration bubble model, the real-time plant structure feature corresponding to the second nectar source concentration bubble model is calibrated to be a variant plant structure feature.

[0044] More specifically, it is characterized in that the step S108 specifically includes the following steps:

[0045] By controlling the planting monitoring camera to shoot the nectar plants in the sensitive planting area, the plant structure image data of each nectar plant is obtained, and the feature extraction of the plant structure image data is performed to obtain the actual plant structure features of each nectar plant in the sensitive planting area;

[0046] Comparing the degree of consistency between the actual plant structural features and the variant plant structural features one by one, if the degree of consistency is greater than a preset degree of consistency, marking the nectar plants corresponding to the sensitive planting areas where the degree of consistency is greater than the preset degree of consistency, and obtaining a distribution map of the variant nectar plants;

[0047] Introducing a feature dislocation algorithm, calculating a dislocation function between the features of the variant plant structural features of each variant nectar plant in the variant nectar plant distribution map compared with the normal plant structural features through the feature dislocation algorithm, and determining the dislocation amplitude of the structural features of each variant nectar plant according to the dislocation function;

[0048] Performing matrix description processing on each variant nectar plant in the sensitive planting area based on the structural feature dislocation amplitude of each variant nectar plant to generate a structural feature dislocation matrix of the variant nectar plant;

[0049] Obtaining a planting log of nectar plants in the ecological orchard and a preset planting management strategy implemented therein, and extracting a number of planting data outputted during the planting process of the nectar plants in accordance with the preset planting management strategy through the planting log;

[0050] Constructing a Bayesian probability distribution network for the generation of variant nectar plants during the planting process of nectar plants implementing a preset planting management strategy, defining the variational lower bound of the Bayesian probability distribution network based on the constructed characteristic dislocation matrix, and establishing a variational distribution family based on a number of planting data;

[0051] Preset a maximum lower boundary likelihood threshold, adjust the variational distribution family to maximize the variational lower boundary, and obtain a lower boundary likelihood value. If the lower boundary likelihood value is greater than or equal to the maximum lower boundary likelihood threshold, then output the adjusted current variational distribution family. Determine the probability of mutation of the nectar source plant during the planting process of the nectar source plant under the preset planting management strategy based on the current variational distribution family.

[0052] If the mutation probability is greater than the preset mutation probability, the nectar plants in the ecological orchard are planted and managed with characteristic mutations and optimized; if the mutation probability is less than the preset mutation probability, the nectar plants in the ecological orchard are planted and managed with non-characteristic mutations and optimized to obtain the planting management strategy of the ecological orchard.

[0053] The second aspect of the present invention provides an ecological orchard planting management system based on the symbiotic mechanism of natural enemy insects and nectar plants. The ecological orchard planting management system includes a memory and a processor. The memory stores an ecological orchard planting management method program based on the symbiotic mechanism of natural enemy insects and nectar plants. When the ecological orchard planting management method program is executed by the processor, any one of the steps of the ecological orchard planting management method is implemented.

[0054] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are:

[0055] The scale of insect pests and the required natural enemy insect species in the global planting area of ​​the ecological orchard are obtained, and a real-time insect pest distribution hotspot map of the global planting area is constructed. The color status of the hotspots within the real-time insect pest distribution hotspot map where the pest killing scale reaches the expected pest control rate is analyzed to determine the termination level of natural enemy insects. When sensitive planting areas exist in the global planting area of ​​the ecological orchard, the minimum nectar source concentration generation interval of the nectar source plants is set according to the expected intake concentration of natural enemy insects in the sensitive planting area. The real-time nectar source concentration generation interval of the part of interest is obtained. The concentration bubble model is used to determine whether the minimum nectar source concentration generation interval and the real-time nectar source concentration generation interval overlap to determine whether the nectar source plant structure has varied. The distribution map of variant nectar source plants in the sensitive planting area is obtained through image data comparison and calculation. The corresponding structural feature dislocation matrix is ​​established in the variant nectar source plant distribution map according to the feature dislocation amplitude. The variation probability that leads to structural variation of nectar source plants during the planting process is calculated based on the structural feature dislocation matrix and several planting data. The ecological orchard is planted and managed according to the variation probability, and the planting management strategy of the ecological orchard is obtained. The present invention can carry out reasonable and accurate planting management of nectar plants in ecological orchards based on the dependent symbiotic mechanism between natural enemy insects and nectar plants, ensuring that natural enemy insects can efficiently kill existing insect pests during the planting process of nectar plants, while ensuring that nectar plants can provide high-quality habitats and living conditions for natural enemy insects, thereby improving the planting quality and benefits of nectar plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying any creative work, they can also obtain drawings of other embodiments based on these drawings.

[0057] Figure 1 The process of ecological orchard planting management method based on the symbiotic mechanism of natural enemy insects and nectar plants is shown Figure 1 ;

[0058] Figure 2 The process of ecological orchard planting management method based on the symbiotic mechanism of natural enemy insects and nectar plants is shown Figure 2 ;

[0059] Figure 3 The system framework diagram of the ecological orchard planting management system based on the symbiotic mechanism of natural enemy insects and nectar plants is shown. DETAILED DESCRIPTION

[0060] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0061] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0062] The first aspect of the present invention provides an ecological orchard planting and management method based on the symbiotic mechanism between natural enemy insects and nectar plants, such as Figure 1 As shown, the following steps are included:

[0063] S102: Obtain the scale of pests and the required natural enemy insect species in the global planting area of ​​the ecological orchard, construct a real-time pest distribution hotspot map for the global planting area, analyze the color of hotspots within the real-time pest distribution hotspot map when the pest control scale reaches the expected pest control rate, and determine the natural enemy insect termination level;

[0064] S104: Practically planting nectar-producing plants according to the termination level of natural enemy insects to construct a multidimensional spatial density model of natural enemy insects, and calculating and analyzing whether the distribution density of natural enemy insects in the global planting area during the practical pest control meets the ideal distribution density based on the multidimensional spatial density model, thereby obtaining first-day and second-day natural enemy insect distribution analysis results;

[0065] S106: When a sensitive planting area exists in the global planting area of ​​the ecological orchard, a minimum nectar source concentration generation interval of the nectar source plant is set according to the expected intake concentration of the natural enemy insect termination level in the sensitive planting area, and the real-time nectar source concentration generation interval of the part of interest is obtained. The minimum nectar source concentration generation interval and the real-time nectar source concentration generation interval are overlapped and judged by a concentration bubble model to determine whether the structural variation of the nectar source plant occurs;

[0066] S108: Obtain a distribution map of variant nectar plants in sensitive planting areas through image data comparison and calculation, establish a corresponding structural feature dislocation matrix in the distribution map of variant nectar plants according to the feature dislocation amplitude, calculate the variation probability of the structural variation of nectar plants during the planting process based on the structural feature dislocation matrix and a number of planting data, and manage the ecological orchard through the variation probability to obtain a planting management strategy for the ecological orchard.

[0067] More specifically, the step S102 is as follows: Figure 2 As shown, specifically including the following steps:

[0068] S202: Acquire the global planting area of ​​the ecological orchard, and use the Internet of Things to control the planting monitoring camera in the ecological orchard to capture and shoot the global planting area in real time, so as to obtain real-time image data of the surfaces of all nectar-producing plants in the global planting area;

[0069] S204: Introducing an ORB image extraction algorithm to perform feature calculation and extraction on the real-time image data corresponding to each nectar-producing plant in the global planting area, to obtain a real-time insect pest feature vector existing on the surface of each nectar-producing plant in the global planting area;

[0070] S206: Quantitatively counting the number of insect pests corresponding to the real-time insect pest feature vectors existing on the surface of each nectar-producing plant in the global planting area to obtain the scale of insect pests in the global planting area;

[0071] S208: Obtaining a garden pest control knowledge graph based on a big data network, identifying real-time pest feature vectors using the garden pest control knowledge graph, and outputting natural enemy insect species required for controlling the pests;

[0072] S210: Constructing a hotspot map, performing a fitting operation on the pest scale of the global planting area in the hotspot map to generate a real-time pest distribution hotspot map of the global planting area, obtaining a planting distribution pattern diagram of nectar-producing plants in the global planting area, and dividing the real-time pest distribution hotspot map into N uniform sub-hotspot map areas based on the planting distribution pattern diagram;

[0073] S212: Obtain a first hotspot color index representing the display in each sub-hotspot map area, create a hotspot color rule query table for natural enemy insect control required for different pest scales, obtain a second hotspot color index representing the necessary pest scale through the hotspot color rule query table, and determine the termination level of natural enemy insect control for the pest scale of the global planting area based on the difference between the first hotspot color index and the second hotspot color index.

[0074] It's important to note that the magnitude of natural enemy insects encompasses both their numbers and the growth stages of their individuals. Natural enemy insects and nectar-producing plants share a symbiotic relationship, enabling them to eliminate various plant pests through ingestion or parasitism. Therefore, to ensure high-quality, rational cultivation and management of ecological orchards, natural enemy insects can be introduced to control and eliminate pests that arise during the cultivation process. However, the introduction of natural enemy insects is not unlimited. Excessive numbers may make it difficult to effectively eliminate pests, and may even cause some damage to the orchard's ecosystem. Therefore, this method extracts the features of the pests that need to be controlled from the real-time image data of the surface of the nectar plants, so that it can further determine the species of natural enemy insects needed to control the pests; because the planting scale of nectar plants in ecological orchards is usually large, and the number of pests on each nectar plant is also large, the distribution of the scale of pests in the planting area will form a scattered distribution pattern. Therefore, in order to efficiently and accurately calculate the number of natural enemy insects and their growth stages that need to be introduced to control the scale of pests of nectar plants in the planting area, this method extracts the features of the pests that need to be controlled from the real-time image data of the surface of the nectar plants, so as to further determine the species of natural enemy insects needed to control the pests; because the planting scale of nectar plants in ecological orchards is usually large, and the number of pests on each nectar plant is also large, the distribution of the scale of pests in the planting area will form a scattered distribution pattern. Therefore, in order to efficiently and accurately calculate the number of natural enemy insects and their growth stages that need to be introduced to control the scale of pests of nectar plants in the planting area, this method extracts the features of the natural enemy insects that need to be controlled from the real-time image data of the surface of the nectar plants, so as to further determine the species of natural enemy insects that need to be controlled The distribution of pests on nectar-producing plants is visualized by constructing a real-time hotspot map. This map uses hotspot values, numbers, and their colors to express the density and distribution of pests. This provides an accurate and reliable analytical basis for subsequent natural enemy insect magnitude location, while reducing the tedious steps of traditional manual field data collection and recording, saving time and effort. Furthermore, the real-time pest distribution hotspot map is divided into N uniform sub-hotspot map regions based on the planting distribution pattern diagram for partitioning operations, improving the precision of pest distribution calculations and avoiding omissions in natural enemy insect magnitude planning. This method can output the pest distribution scale of nectar-producing plants in ecological orchards as a hotspot map, further enabling the calculation of the number of natural enemy insects required for pest control and their individual growth stages. This method replaces the traditional manual field survey steps and the resulting errors in manual calculations, improving the accuracy and reliability of nectar plant planting management in ecological orchards.

[0075] It should be noted that the ORB image extraction algorithm is an efficient and fast feature point detection and description algorithm with excellent performance in feature matching and key point extraction. It is particularly suitable for real-time applications. It can ensure the accuracy of real-time pest feature vector extraction, reduce the extraction noise of real-time pest feature vectors, and enhance the robustness of image feature extraction.

[0076] More specifically, in step S102, a first hotspot color index representing each sub-hotspot map area is obtained, a hotspot color rule query table for natural enemy insect control required for different pest scales is created, a second hotspot color index representing the necessary pest scale is obtained through the hotspot color rule query table, and a termination level of natural enemy insect control for the pest scale of the global planting area is determined based on the difference between the first hotspot color index and the second hotspot color index, specifically comprising the following steps:

[0077] Obtaining the hotspot color index displayed in each sub-hotspot map area, defined as the first hotspot color index, and extracting the real-time environmental parameters of each sub-hotspot map area;

[0078] Preset the initial control level of natural enemy insect species based on the scale of pests in the global planting area, and obtain the historical control and extermination rates of natural enemy insect species under different combinations of environmental parameters and different pest scales based on the initial control level of natural enemy insect species based on the big data network;

[0079] Based on the historical pest control rates of the natural enemy insect species at different pest scales under different combinations of environmental parameter conditions, an association rule is used to create a hotspot color rule query table for natural enemy insect control required for different pest scales.

[0080] Obtain the established planting requirements of the ecological orchard, obtain the expected pest control rate of the global planting area based on the established planting requirements, and traverse the search in the hotspot color rule query table to obtain the hotspot color range that meets the expected pest control rate under the real-time environmental parameter conditions of each sub-hotspot map area;

[0081] Obtaining the hotspot color of each sub-hotspot map region in the hotspot color domain, wherein the hotspot color represents the initial control level of the natural enemy insect species and the pest scale required to achieve the desired control and pest elimination rate when killing the pests under the corresponding real-time environmental parameters in the sub-hotspot map region, and is defined as the required pest scale;

[0082] Extract the hotspot color index of the necessary pest scale and define it as the second hotspot color index. Calculate the difference between the first hotspot color index and the corresponding second hotspot color index one by one to obtain the hotspot color index difference of each sub-hotspot map area.

[0083] The initial control level of natural enemy insect species is adjusted by the difference in hotspot color index of all sub-hotspot map areas and a weighted sum is performed to finally determine the termination level of natural enemy insects for controlling the scale of pests in the global planting area.

[0084] It should be noted that, since the hotspot map can express the distribution density and scale of pests through hotspot colors, this method first obtains the first hotspot color index displayed in each sub-hotspot map area. The first hotspot color index is a fuzzy and concrete value representing the scale of pests. Moreover, the pest control ability of natural enemy insects is closely affected by the surrounding environment and the scale of pests. Therefore, a hotspot color query rule can be formulated based on the historical control and pest control rates of natural enemy insect species under different combinations of environmental parameter conditions and different pest scales, and a hotspot color rule query table can be obtained. Based on the expected pest control rate, the real-time environmental parameters of the existing sub-hotspot map area, and the scale of insect pests, a hotspot color range that can be adjusted for the initial control level of natural enemy insects is further queried, namely the hotspot color range. Through this query table, the scale of insect pests that can be controlled by the formulated initial control level of natural enemy insects can be quickly queried based on the expected control target, and the rationality of the planned initial level of natural enemy insects can be confirmed, thereby providing a stable and reliable basis for the subsequent hotspot color analysis and calculation between the actual pest scale and the controllable scale under the ideal target, greatly improving the planning accuracy of the natural enemy insect level. Thus, it is possible to obtain the pest scale required to achieve the desired pest control rate when killing pests under the corresponding real-time environmental parameters in the sub-hotspot map area after adjusting the initial control level of natural enemy insects, that is, the required pest scale. By calculating the difference between the second hotspot color index corresponding to the required pest scale and the first hotspot color index of the actual pest scale in the sub-hotspot map area, the hotspot color index difference is the final adjustment range of the natural enemy insect scale of each sub-hotspot map area. Therefore, the weighted summation of the adjusted natural enemy insect scale of each area can be used to determine the natural enemy insect termination level for controlling the pest scale in the global planting area. Through this method, the required number of natural enemy insects and individual growth stages for pest control of nectar plants in ecological orchards can be quickly and accurately located and calculated in the form of pest scale hotspot color analysis, thereby improving the stability and rationality of the symbiotic control of natural enemy insects on nectar plants. At the same time, it can greatly improve the accuracy of nectar plant pest control, reduce the situation of ecological damage caused by the traditional unreasonable introduction of natural enemy insects, reduce the use of pesticides, enhance pest population control performance, and improve the biodiversity and long-term ecological stability of the orchard.

[0085] More specifically, the step S104 includes the following steps:

[0086] Pest control was carried out in the global planting area of ​​the ecological orchard by measuring the termination level of natural enemy insects, and a multidimensional density model of the global planting area was constructed;

[0087] In the process of pest control, the spatial density distribution of pests killed and inhabited by natural enemy insects in the global planting area is recorded to obtain spatial density distribution data of natural enemy insects. The spatial density distribution data is fitted by a multidimensional density model to obtain a multidimensional spatial density model of natural enemy insects.

[0088] Extracting a spatial density distribution model diagram of natural enemy insects within a preset prevention and control period from the multidimensional spatial density model, and dividing the spatial density distribution model diagram into a plurality of sub-model diagrams based on a schematic diagram of the planting distribution pattern;

[0089] Obtaining a practical density kernel function expressing the corresponding spatial density distribution of natural enemy insects in each sub-model graph, introducing a trade-off algorithm, calculating the gain weight of each sub-model graph in the trade-off algorithm based on the scale of the pest, taking the expected pest control rate as the allocation premise, and allocating the pest control rate to each sub-model graph according to the gain weight under the allocation premise to obtain the sub-expected pest control rate of each sub-model graph;

[0090] Based on the sub-expected pest control rate, an ideal density kernel function range of the distribution of natural enemy insects for the practical pest control of nectar plants in each sub-model diagram is preset, and whether the practical density kernel function in each sub-model diagram is within the corresponding ideal density kernel function range is determined one by one;

[0091] If the practical density kernel function in the sub-model graph is within the range of the corresponding ideal density kernel function, the nectar plant planting area corresponding to the sub-model graph is calibrated as the accurate planting area, and the first day enemy insect distribution analysis result is obtained;

[0092] If the practical density kernel function in the sub-model graph is not within the range of the corresponding ideal density kernel function, the nectar plant planting area corresponding to the sub-model graph is calibrated as a sensitive planting area, and the enemy insect distribution analysis results of the second day are obtained.

[0093] It should be noted that after determining the natural enemy insect termination level for pest control within the global planting area, due to the symbiotic dependence between natural enemy insects and nectar plants, that is, when natural enemy insects control pests on nectar plants, they also need to feed on the nectar and sap of nectar plants to provide energy for their own habitat. However, when nectar plants are planted improperly, the nectar production concentration of nectar plants may decrease, causing natural enemy insects to not use them as energy sources to maintain their own survival. Therefore, this method uses the determined natural enemy insect termination level to implement practical pest control on nectar plants in the ecological orchard, thereby obtaining a preliminary spatial distribution of natural enemy insects within the global planting area. The spatial density distribution of natural enemy insects on nectar plants can reflect whether natural enemy insects rely on the nectar of a particular nectar plant for survival energy. In other words, nectar plants in the planting area with a higher spatial density distribution are the nectar plants that natural enemy insects have chosen as symbiotic energy sources, indicating that the nectar energy supply of these selected nectar plants is normal and can ensure the survival and habitat of natural enemy insects, and thus indicates that the nectar plant planting in the area is relatively reasonable. The nectar plants that were not selected may have nectar supply that does not meet the survival needs of natural enemy insects due to some planting management errors and mistakes, and tend to be planted irrationally. Therefore, the spatial density distribution in the planting area is relatively small. Therefore, this method analyzes the target selection of natural enemy insects for the survival energy supply of nectar plants by constructing a multidimensional spatial density model of natural enemy insects, and analyzes and determines whether the planting of nectar plants represented by each sub-model is reasonable based on the practical density kernel function and the ideal density kernel function range allocated based on the expected control and pest control rate. If the practical density kernel function in the sub-model diagram is within the corresponding ideal density kernel function range, it means that the practical natural enemy insect symbiosis distribution in the planting area is consistent with the ideal natural enemy insect symbiosis distribution under the premise of the expected control and pest control rate, that is, the nectar energy supply of the nectar plants in the planting area can reasonably meet the survival needs of the natural enemy insects, so the nectar plant planting area corresponding to the sub-model diagram is calibrated as the accurate planting area; otherwise, it means that the nectar energy supply of the nectar plants in the planting area cannot reasonably meet the survival needs of the natural enemy insects, so the nectar plant planting area corresponding to the sub-model diagram is calibrated as the sensitive planting area.

[0094] It should be noted that this method can further analyze whether nectar plants can meet the nectar survival energy supply needs of natural enemy insects based on the distribution density of symbiotic dependence of natural enemy insects on nectar plants during the practical pest control in ecological orchards according to the determined termination level of natural enemy insects. In this way, it is possible to quickly and efficiently preliminarily determine whether the planting of each nectar plant planting area in the ecological orchard is reasonable or not, which can greatly improve the management accuracy and rationality of the planting of nectar plants in ecological orchards, avoid imposing unreasonable planting management steps on normally planted nectar plants, and avoid omissions and errors in the planting management of abnormally planted nectar plants.

[0095] More specifically, the step S106 includes the following steps:

[0096] When there are sensitive planting areas in the global planting area of ​​the ecological orchard, the symbiotic mechanism between natural enemy insect species and nectar-producing plant species can be obtained through big data;

[0097] Based on the analysis of the symbiotic mechanism between natural enemy insect species and nectar plant species, the natural enemy insect termination level is obtained, and the nectar intake concentration provided by the nectar plant in the process of killing pests to achieve the expected control and pest control rate under the corresponding real-time environmental parameters in the sensitive planting area is obtained, and multiple key nectar intake concentrations are obtained;

[0098] Constructing a nectar source intake concentration gradient for the process of killing natural enemy insects in real time within the global planting area of ​​the ecological orchard based on the distribution of the multiple key nectar source intake concentrations, and setting a minimum nectar source concentration generation interval for nectar plants based on the nectar source intake concentration gradient;

[0099] Creating a concentration bubble model, calculating and fitting the minimum nectar source concentration generation interval in the concentration bubble model to obtain a minimum nectar source concentration bubble model, which is marked as a first nectar source concentration bubble model;

[0100] Obtaining the variety information of the nectar-producing plants, searching the big data network for the parts of the nectar-producing plants that the natural enemy insects are interested in when feeding on the nectar-producing plants based on the natural enemy insect species and the variety information of the nectar-producing plants, and simultaneously obtaining the real-time characteristics of the plant structure of the parts of interest;

[0101] Through the comprehensive plant knowledge graph, the real-time feature recognition query of the plant structure is performed to obtain the nectar source concentration generation interval generated when the part of interest is under the real-time environmental parameter conditions corresponding to the sensitive planting area, which is defined as the real-time nectar source concentration generation interval;

[0102] Calculating and fitting the real-time nectar source concentration generation interval in the concentration bubble model to obtain a real-time nectar source concentration bubble model, which is marked as a second nectar source concentration bubble model;

[0103] If the second nectar source concentration bubble model can completely cover the first nectar source concentration bubble model, the real-time plant structure feature corresponding to the second nectar source concentration bubble model is calibrated to be a normal plant structure feature;

[0104] If the second nectar source concentration bubble model cannot completely cover the first nectar source concentration bubble model, the real-time plant structure feature corresponding to the second nectar source concentration bubble model is calibrated to be a variant plant structure feature.

[0105] It should be noted that when sensitive planting areas exist within the global planting area of ​​an ecological orchard, it indicates that the nectar plants planted in the corresponding areas of the sensitive spatial density distribution model do not meet the symbiotic requirements of natural enemy insects. When errors or errors occur in the planting of nectar plants in an ecological orchard, such as improper fertilizer use, excessive irrigation, unsuitable temperature and light conditions, and excessive planting density, these abnormalities in the planting process can easily cause the nectar-producing parts of the nectar plants that provide nectar to natural enemy insects to mutate, thereby significantly reducing the concentration of nectar they can produce, resulting in the nectar plants being unable to meet the energy needs of natural enemy insects for survival and habitat. Therefore, this method uses multiple planting research cases to analyze the key nectar intake concentration that the nectar plants in the ecological orchard need to provide when the nectar plants reach the termination level of natural enemy insect control, thereby setting a minimum nectar concentration generation interval that the nectar plants need to provide in the process of natural enemy insect termination level control. The minimum nectar concentration generation interval is an actual demand range of natural enemy insects for nectar; since the nectar of nectar plants is mainly a certain cell fluid or secretion generated by the action of their internal plant structure, the variation of nectar plants mainly occurs in their plant structure, and different natural enemy insects have special feeding parts, that is, parts of interest, for different nectar plants. Therefore, the real-time characteristics of the plant structure of the parts of interest to natural enemy insects when feeding on nectar plants are obtained, and the real-time nectar concentration generation interval that can be produced under current environmental conditions is obtained based on the real-time characteristics of the plant structure. The real-time nectar concentration generation interval is an actual supply range of nectar concentration that the nectar plants can generate under the current planting environment.

[0106] It should be noted that in order to improve the analysis efficiency and reduce the analysis error rate of whether the nectar concentration meets the energy supply demand of natural enemy insects, this method uses a bubble model to fit the two nectar concentration generation intervals and then performs coverage comparison; if the second nectar concentration bubble model can completely cover the first nectar concentration bubble model, it means that the nectar concentration supply generated by the interesting part of the nectar plant can maximize the actual nectar concentration demand during the natural enemy insect extermination process, indicating that the structural characteristics of the interesting part of the nectar plant have not been mutated due to planting errors. Therefore, the plant structural real-time characteristics corresponding to the second nectar concentration bubble model are calibrated to be normal plant structural characteristics; if the second nectar concentration bubble model cannot completely cover the first nectar concentration bubble model, it means that the nectar concentration supply generated by the interesting part of the nectar plant does not meet the actual nectar concentration demand during the natural enemy insect extermination process, indicating that the characteristics of the interesting part of the nectar plant have been mutated due to planting errors. Therefore, the plant structural real-time characteristics corresponding to the second nectar concentration bubble model are calibrated to be variant plant structural characteristics. This method can be used to calculate and analyze whether the amount of nectar provided by nectar plants can meet the energy requirements of natural enemy insects. This can further determine whether the structure of the nectar plants has mutated during the planting process, providing a strong and feasible management basis for the planting of nectar plants in ecological orchards, improving the management accuracy of ecological nectar plants, and ensuring the growth quality and health level of nectar plants.

[0107] More specifically, the step S108 includes the following steps:

[0108] By controlling the planting monitoring camera to shoot the nectar plants in the sensitive planting area, the plant structure image data of each nectar plant is obtained, and the feature extraction of the plant structure image data is performed to obtain the actual plant structure features of each nectar plant in the sensitive planting area;

[0109] Comparing the degree of consistency between the actual plant structural features and the variant plant structural features one by one, if the degree of consistency is greater than a preset degree of consistency, marking the nectar plants corresponding to the sensitive planting areas where the degree of consistency is greater than the preset degree of consistency, and obtaining a distribution map of the variant nectar plants;

[0110] Introducing a feature dislocation algorithm, calculating a dislocation function between the features of the variant plant structural features of each variant nectar plant in the variant nectar plant distribution map compared with the normal plant structural features through the feature dislocation algorithm, and determining the dislocation amplitude of the structural features of each variant nectar plant according to the dislocation function;

[0111] Performing matrix description processing on each variant nectar plant in the sensitive planting area based on the structural feature dislocation amplitude of each variant nectar plant to generate a structural feature dislocation matrix of the variant nectar plant;

[0112] Obtaining a planting log of nectar plants in the ecological orchard and a preset planting management strategy implemented therein, and extracting a number of planting data outputted during the planting process of the nectar plants in accordance with the preset planting management strategy through the planting log;

[0113] Constructing a Bayesian probability distribution network for the generation of variant nectar plants during the planting process of nectar plants implementing a preset planting management strategy, defining the variational lower bound of the Bayesian probability distribution network based on the constructed characteristic dislocation matrix, and establishing a variational distribution family based on a number of planting data;

[0114] Preset a maximum lower boundary likelihood threshold, adjust the variational distribution family to maximize the variational lower boundary, and obtain a lower boundary likelihood value. If the lower boundary likelihood value is greater than or equal to the maximum lower boundary likelihood threshold, then output the adjusted current variational distribution family. Determine the probability of mutation of the nectar source plant during the planting process of the nectar source plant under the preset planting management strategy based on the current variational distribution family.

[0115] If the mutation probability is greater than the preset mutation probability, the nectar plants in the ecological orchard are planted and managed with characteristic mutations and optimized; if the mutation probability is less than the preset mutation probability, the nectar plants in the ecological orchard are planted and managed with non-characteristic mutations and optimized to obtain the planting management strategy of the ecological orchard.

[0116] It needs to be explained that due to the errors and mistakes of part of the planting steps or processes in the planting process, the variation phenomenon of the nectar source plant structure in the local planting area is generated, and the planting situation of the normal nectar source plant and the variation nectar source plant is formed. It is difficult to guarantee the survival and habitat demand of the mutual beneficial natural enemy insects of the nectar source plant, reduce the survival rate of the natural enemy insects introduced for pest control, and greatly destroy the symbiotic mechanism between the natural enemy insects and the nectar source plant. To this end, the steps or processes of the nectar source plant about the characteristic variation elimination need to be effectively and reasonably planted. First, for each nectar source plant planting area, it is necessary to judge whether there is a variation plant structure characteristic nectar source plant. This method is realized in the form of image feature extraction comparison. If the coincidence degree is greater than the preset coincidence degree, it means that the plant structure characteristics of the interest part of the nectar source plant coincide with the variation plant structure characteristics, so the nectar source plant corresponding to the coincidence degree greater than the preset coincidence degree in the sensitive planting area is further marked, so that a schematic diagram describing the distribution of the variation nectar source plant in the sensitive planting area can be obtained. The variation plant structure is formed by a certain degree of mutation caused by planting errors based on the normal plant structure, so the dislocation function between the two can express the planting error degree of the nectar source plant, and the dislocation matrix of the variation nectar source plant generated by the matrix description can reflect the planting error degree of the variation nectar source plant in the whole sensitive planting area. This is an important parameter prerequisite for calculating the variation probability, which determines whether the calculation of the variation probability is accurate or not. This method constructs a Bayesian probability distribution network of the variation nectar source plant generated in the planting process of the nectar source plant implementing the preset planting management strategy, which expresses the joint distribution of the planting data between the variation nectar source plants. Then the variation distribution family established by adjusting a plurality of planting data is used to maximize the variation lower boundary of the Bayesian probability distribution network defined by the structure characteristic dislocation matrix, so that the randomness of the nectar source plant leading to the variation of the nectar source plant in the planting process of the nectar source plant implementing the preset planting management strategy is more close to the most accurate posterior probability distribution, so that the obtained variation probability is more accurate and reliable, and the planting management of the nectar source plant is greatly optimized.

[0117] It should be noted that if the mutation probability is greater than the preset mutation probability, it means that the mutation phenomenon of the mutant nectar plant is most likely caused by the unreasonable operation of a certain step or process in the planting process, such as improper use of fertilizers, improper use of plant hormones, or improper soil acidification management, etc. Therefore, it is necessary to manage and optimize the planting of these characteristic mutations of the nectar plants in the ecological orchard; otherwise, it means that the mutation phenomenon of the mutant nectar plant is not caused by the unreasonable operation of a certain step or process in the planting process, so it is only necessary to manage and optimize the planting of non-characteristic mutations of the nectar plants in the ecological orchard, such as pest and disease management, excessive planting density, or unsuitable temperature and light. This method can calculate and analyze the probability of mutation that causes nectar plants to mutate during the planting process of nectar plants based on planting data and the characteristic dislocation matrix of mutant nectar plants in the planting area, so as to carry out targeted situation-by-case management for the process operations or steps of characteristic mutation and non-characteristic mutation of nectar plants during the planting process, improve the rationality of planting nectar plants in ecological orchards, avoid the phenomenon of plant structural characteristic mutation with a high probability in nectar plants during the planting process, effectively improve the energy supply demand for survival of natural enemy insects and the quality of the living environment, optimize the symbiotic mechanism between natural enemy insects and nectar plants, and increase the productivity of orchards and the planting quality of nectar plants.

[0118] The second aspect of the present invention provides an ecological orchard planting management system based on the symbiotic mechanism of natural enemy insects and nectar plants, such as Figure 3 As shown, the ecological orchard planting management system includes a memory 31 and a processor 32. The memory 31 stores an ecological orchard planting management method program based on the symbiotic mechanism of natural enemy insects and nectar plants. When the ecological orchard planting management method program is executed by the processor 32, any one of the steps of the ecological orchard planting management method is implemented.

[0119] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. An ecological orchard planting and management method based on the symbiotic mechanism between natural enemy insects and nectar plants, characterized in that: The following steps are involved: S102: Obtain the scale of pests and the required natural enemy insect species in the global planting area of ​​the ecological orchard, construct a real-time pest distribution hotspot map for the global planting area, analyze the color of hotspots within the real-time pest distribution hotspot map when the pest control scale reaches the expected pest control rate, and determine the natural enemy insect termination level; S104: Practically planting nectar-producing plants according to the termination level of natural enemy insects to construct a multidimensional spatial density model of natural enemy insects, and calculating and analyzing whether the distribution density of natural enemy insects in the global planting area during the practical pest control meets the ideal distribution density based on the multidimensional spatial density model, thereby obtaining first-day and second-day natural enemy insect distribution analysis results; S106: When a sensitive planting area exists in the global planting area of ​​the ecological orchard, a minimum nectar source concentration generation interval of the nectar source plant is set according to the expected intake concentration of the natural enemy insect termination level in the sensitive planting area, and the real-time nectar source concentration generation interval of the part of interest is obtained. The minimum nectar source concentration generation interval and the real-time nectar source concentration generation interval are overlapped and judged by a concentration bubble model to determine whether the structural variation of the nectar source plant occurs; S108: Obtaining a distribution map of variant nectar plants in the sensitive planting area through image data comparison and calculation, establishing a corresponding structural feature dislocation matrix in the variant nectar plant distribution map based on the feature dislocation amplitude, calculating the probability of variation that causes structural variation of the nectar plants during the planting process based on the structural feature dislocation matrix and a number of planting data, and managing the planting of the ecological orchard based on the variation probability to obtain a planting management strategy for the ecological orchard; The step S104 specifically includes the following steps: Pest control was carried out in the global planting area of ​​the ecological orchard by measuring the termination level of natural enemy insects, and a multidimensional density model of the global planting area was constructed; In the process of pest control, the spatial density distribution of pests killed and inhabited by natural enemy insects in the global planting area is recorded to obtain spatial density distribution data of natural enemy insects. The spatial density distribution data is fitted by a multidimensional density model to obtain a multidimensional spatial density model of natural enemy insects. Extracting a spatial density distribution model diagram of natural enemy insects within a preset prevention and control period from the multidimensional spatial density model, and dividing the spatial density distribution model diagram into a plurality of sub-model diagrams based on a schematic diagram of the planting distribution pattern; Obtaining a practical density kernel function expressing the corresponding spatial density distribution of natural enemy insects in each sub-model graph, introducing a trade-off algorithm, calculating the gain weight of each sub-model graph in the trade-off algorithm based on the scale of the pest, taking the expected pest control rate as the allocation premise, and allocating the pest control rate to each sub-model graph according to the gain weight under the allocation premise to obtain the sub-expected pest control rate of each sub-model graph; Based on the sub-expected pest control rate, an ideal density kernel function range of the distribution of natural enemy insects for the practical pest control of nectar plants in each sub-model diagram is preset, and whether the practical density kernel function in each sub-model diagram is within the corresponding ideal density kernel function range is determined one by one; If the practical density kernel function in the sub-model graph is within the range of the corresponding ideal density kernel function, the nectar plant planting area corresponding to the sub-model graph is calibrated as the accurate planting area, and the first day enemy insect distribution analysis result is obtained; If the practical density kernel function in the sub-model graph is not within the range of the corresponding ideal density kernel function, the nectar plant planting area corresponding to the sub-model graph is calibrated as a sensitive planting area, and the enemy insect distribution analysis results of the second day are obtained.

2. The ecological orchard planting and management method based on the symbiotic mechanism of natural enemy insects and nectar plants according to claim 1, characterized in that: The step S102 specifically comprises the following steps: Obtain the global planting area of ​​the ecological orchard, and use the Internet of Things to control the planting monitoring cameras in the ecological orchard to capture and shoot the global planting area in real time, so as to obtain real-time image data of the surfaces of all nectar-producing plants in the global planting area; Introducing the ORB image extraction algorithm to perform feature calculation and extraction on the real-time image data corresponding to each nectar-producing plant in the global planting area, thereby obtaining a real-time pest feature vector existing on the surface of each nectar-producing plant in the global planting area; Quantitatively counting the number of insect pests corresponding to the real-time insect pest feature vectors existing on the surface of each nectar-producing plant in the global planting area to obtain the scale of insect pests in the global planting area; Obtaining a garden pest control knowledge graph based on a big data network, identifying real-time pest feature vectors through the garden pest control knowledge graph, and outputting the natural enemy insect species required to control the pests; Constructing a hotspot map, performing a fitting operation on the pest scale of the global planting area in the hotspot map to generate a real-time pest distribution hotspot map of the global planting area, obtaining a planting distribution pattern diagram of nectar-producing plants in the global planting area, and dividing the real-time pest distribution hotspot map into N uniform sub-hotspot map areas based on the planting distribution pattern diagram; Obtain the first hotspot color index displayed in each sub-hotspot map area, create a hotspot color rule query table for the natural enemy insect control required for different pest scales, obtain the second hotspot color index representing the necessary pest scale through the hotspot color rule query table, and determine the termination level of the natural enemy insect control for the pest scale of the global planting area based on the difference between the first hotspot color index and the second hotspot color index.

3. The ecological orchard planting management method based on the symbiotic mechanism of natural enemy insects and nectar plants according to claim 2 is characterized in that, In the step S102, a first hotspot color index representing each sub-hotspot map area is obtained, a hotspot color rule query table for natural enemy insect control required for different pest scales is created, a second hotspot color index representing the necessary pest scale is obtained through the hotspot color rule query table, and a termination level of natural enemy insect control for the pest scale of the global planting area is determined based on the difference between the first hotspot color index and the second hotspot color index, specifically comprising the following steps: Obtaining the hotspot color index displayed in each sub-hotspot map area, defined as the first hotspot color index, and extracting the real-time environmental parameters of each sub-hotspot map area; Preset the initial control level of natural enemy insect species based on the scale of pests in the global planting area, and obtain the historical control and extermination rates of natural enemy insect species under different combinations of environmental parameters and different pest scales based on the initial control level of natural enemy insect species based on the big data network; Based on the historical pest control rates of the natural enemy insect species at different pest scales under different combinations of environmental parameter conditions, an association rule is used to create a hotspot color rule query table for natural enemy insect control required for different pest scales. Obtain the established planting requirements of the ecological orchard, obtain the expected pest control rate of the global planting area based on the established planting requirements, and traverse the search in the hotspot color rule query table to obtain the hotspot color range that meets the expected pest control rate under the real-time environmental parameter conditions of each sub-hotspot map area; Obtaining the hotspot color of each sub-hotspot map region in the hotspot color domain, wherein the hotspot color represents the initial control level of the natural enemy insect species and the pest scale required to achieve the desired control and pest elimination rate when killing the pests under the corresponding real-time environmental parameters in the sub-hotspot map region, and is defined as the required pest scale; Extract the hotspot color index of the necessary pest scale and define it as the second hotspot color index. Calculate the difference between the first hotspot color index and the corresponding second hotspot color index one by one to obtain the hotspot color index difference of each sub-hotspot map area. The initial control level of natural enemy insect species is adjusted by the difference in hotspot color index of all sub-hotspot map areas and a weighted sum is performed to finally determine the termination level of natural enemy insects for controlling the scale of pests in the global planting area.

4. The ecological orchard planting and management method based on the symbiotic mechanism of natural enemy insects and nectar plants according to claim 1 is characterized in that: The S106 step specifically includes the following steps: When there are sensitive planting areas in the global planting area of ​​the ecological orchard, the symbiotic mechanism between natural enemy insect species and nectar-producing plant species can be obtained through big data; Based on the analysis of the symbiotic mechanism between natural enemy insect species and nectar plant species, the natural enemy insect termination level is obtained, and the nectar intake concentration provided by the nectar plant in the process of killing pests to achieve the expected control and pest control rate under the corresponding real-time environmental parameters in the sensitive planting area is obtained, and multiple key nectar intake concentrations are obtained; Constructing a nectar source intake concentration gradient for the process of killing natural enemy insects in real time within the global planting area of ​​the ecological orchard based on the distribution of the multiple key nectar source intake concentrations, and setting a minimum nectar source concentration generation interval for nectar plants based on the nectar source intake concentration gradient; Creating a concentration bubble model, calculating and fitting the minimum nectar source concentration generation interval in the concentration bubble model to obtain a minimum nectar source concentration bubble model, which is marked as a first nectar source concentration bubble model; Obtaining the variety information of the nectar-producing plants, searching the big data network for the parts of the nectar-producing plants that the natural enemies of the insects are interested in when feeding on the nectar-producing plants based on the natural enemy insect species and the variety information of the nectar-producing plants, and simultaneously obtaining the real-time characteristics of the plant structure of the parts of interest; Through the comprehensive plant knowledge graph, the real-time feature recognition query of the plant structure is performed to obtain the nectar source concentration generation interval generated when the part of interest is under the real-time environmental parameter conditions corresponding to the sensitive planting area, which is defined as the real-time nectar source concentration generation interval; Calculating and fitting the real-time nectar source concentration generation interval in the concentration bubble model to obtain a real-time nectar source concentration bubble model, which is marked as a second nectar source concentration bubble model; If the second nectar source concentration bubble model can completely cover the first nectar source concentration bubble model, the real-time plant structure feature corresponding to the second nectar source concentration bubble model is calibrated to be a normal plant structure feature; If the second nectar source concentration bubble model cannot completely cover the first nectar source concentration bubble model, the real-time plant structure feature corresponding to the second nectar source concentration bubble model is calibrated to be a variant plant structure feature.

5. The ecological orchard planting and management method based on the symbiotic mechanism of natural enemy insects and nectar plants according to claim 1 is characterized in that: The step S108 specifically includes the following steps: By controlling the planting monitoring camera to shoot the nectar plants in the sensitive planting area, the plant structure image data of each nectar plant is obtained, and the feature extraction of the plant structure image data is performed to obtain the actual plant structure features of each nectar plant in the sensitive planting area; Comparing the degree of consistency between the actual plant structural features and the variant plant structural features one by one, if the degree of consistency is greater than a preset degree of consistency, marking the nectar plants corresponding to the sensitive planting areas where the degree of consistency is greater than the preset degree of consistency, and obtaining a distribution map of the variant nectar plants; Introducing a feature dislocation algorithm, calculating a dislocation function between the features of the variant plant structural features of each variant nectar plant in the variant nectar plant distribution map compared with the normal plant structural features through the feature dislocation algorithm, and determining the dislocation amplitude of the structural features of each variant nectar plant according to the dislocation function; Performing matrix description processing on each variant nectar plant in the sensitive planting area based on the structural feature dislocation amplitude of each variant nectar plant to generate a structural feature dislocation matrix of the variant nectar plant; Obtaining a planting log of nectar plants in the ecological orchard and a preset planting management strategy implemented therein, and extracting a number of planting data outputted during the planting process of the nectar plants in accordance with the preset planting management strategy through the planting log; Constructing a Bayesian probability distribution network for the generation of variant nectar plants during the planting process of nectar plants implementing a preset planting management strategy, defining the variational lower bound of the Bayesian probability distribution network based on the constructed characteristic dislocation matrix, and establishing a variational distribution family based on a number of planting data; Preset a maximum lower boundary likelihood threshold, adjust the variational distribution family to maximize the variational lower boundary, and obtain a lower boundary likelihood value. If the lower boundary likelihood value is greater than or equal to the maximum lower boundary likelihood threshold, then output the adjusted current variational distribution family. Determine the probability of mutation of the nectar source plant during the planting process of the nectar source plant under the preset planting management strategy based on the current variational distribution family. If the mutation probability is greater than the preset mutation probability, the nectar plants in the ecological orchard are planted and managed with characteristic mutations and optimized; if the mutation probability is less than the preset mutation probability, the nectar plants in the ecological orchard are planted and managed with non-characteristic mutations and optimized to obtain the planting management strategy of the ecological orchard.

6. An ecological orchard planting and management system based on the symbiotic mechanism between natural enemy insects and nectar plants, characterized in that: The ecological orchard planting management system includes a memory and a processor. The memory stores an ecological orchard planting management method program based on the symbiotic mechanism of natural enemy insects and nectar plants. When the ecological orchard planting management method program is executed by the processor, the ecological orchard planting management method steps as described in any one of claims 1 to 5 are implemented.

Images