Intelligent monitoring and early warning device and early warning method for gibberellic disease environmental factors

Abstract

The invention relates to the technical field of agricultural environment monitoring, in particular to an intelligent monitoring and early warning device and method for gibberellic disease environmental factors. Dividing the real-time gibberellic disease distribution overview into a plurality of sub-distribution overviews, performing semi-variant spatial correlation hot spot interpolation on each real-time gibberellic disease sample among the sub-distribution overviews, and performing variational deduction on an average dynamic layout flow of the sub-distribution overviews to obtain a dynamic hot spot change map of real-time gibberellic disease distribution in a specified monitoring area; the method comprises the following steps of: testing the directed relation of the potential occurrence of the Fusarium head blight infection of the previous environmental factors, obtaining the existing occurrence rule of the Fusarium head blight affected by the environmental factors, and updating a dynamic hotspot transition map based on the hazard rating thermodynamic dynamic balance of the existing occurrence rule, so as to determine a prevention and control area and enable the intelligent monitoring and early warning device to make an early warning response. According to the invention, gibberellic disease environmental factors in the wheat planting process can be monitored in real time, prevention and control early warning response can be carried out on an unreasonable planting environment, and the planting environment and quality of wheat are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural environmental monitoring, and in particular to an intelligent monitoring and early warning device and early warning method for scab environmental factors. Background Art

[0002] Fusarium head blight is a major disease of wheat and other grain crops. Caused by the fungus Fusarium, it not only severely impacts crop yields but also produces toxins that contaminate food, threatening human and animal health. The occurrence of fusarium head blight is closely linked to a variety of environmental factors, such as temperature, humidity, rainfall, soil moisture, air humidity, and the crop growth period. Under specific climatic conditions, the disease is highly susceptible to outbreaks and spread. Therefore, real-time monitoring and intelligent early warning of the environmental conditions conducive to fusarium head blight are crucial for early disease prevention and control. With the development of the Internet of Things and artificial intelligence (AI), agricultural disease monitoring and early warning are gradually shifting from traditional empirical judgment to data-driven and model-based approaches. Fusarium head blight monitoring and early warning systems based on environmental factors are becoming a crucial component of smart agriculture.

[0003] However, some of the environmental factor early warning devices currently available on the market for ergot mostly use temperature and humidity as core parameters, lacking the comprehensive collection and analysis of multi-source environmental factors, making it difficult to accurately determine the multi-dimensional conditions for the occurrence of the disease. Some devices rely on static threshold judgments or empirical formulas, lack the ability to dynamically learn and adapt to changing trends of environmental factors, and have low early warning accuracy. At the same time, traditional ergot environmental factor early warning devices have difficulty accurately capturing the temporal dynamic changes in microscopic ergot bacteria, resulting in a significant reduction in the accuracy of the assessment of the suitability of real-time environmental factors in a specified area for the occurrence of ergot, making it impossible to accurately determine the rationality of environmental factors, and prone to false alarms or missed reports. Summary of the Invention

[0004] The present invention overcomes the deficiencies of the prior art and provides an intelligent monitoring and early warning device and method for environmental factors of scab.

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

[0006] A first aspect of the present invention provides an early warning method of an intelligent monitoring and early warning device for scab environmental factors, comprising the following steps:

[0007] S102: Performing time series singular decomposition on the past environmental factor data and the past scab infection data of the specified monitoring area within a preset historical time period using the tensor dimension of the established monitoring requirements, thereby obtaining a core variable tensor structure of the past environmental factor data and the past scab infection data aligned with the preset time period;

[0008] S104: Based on the past spatiotemporal node graph of the core variable tensor structure environmental factor-scab infection, convolutionally extract the temporal convolution blocks and spatial convolution blocks contained in the past spatiotemporal node graph and alternately superimpose them to obtain a real-time overview of scab distribution in the specified monitoring area when real-time environmental factor data is generated within the advanced monitoring time segment;

[0009] S106: Divide the real-time scab distribution overview into multiple sub-distribution profiles, interpolate each real-time scab sample between the sub-distribution profiles using semi-variation spatial correlation hotspots, and perform variational deduction on the average dynamic layout flow of the sub-distribution profiles to obtain a dynamic hotspot change map of the real-time scab distribution within the specified monitoring area;

[0010] S108: Test the directed connection between past environmental factors and potential past ergot infections, obtain the existing occurrence patterns of ergot affected by environmental factors, and update the dynamic hotspot change map based on the hazard rating thermodynamic balance of the existing occurrence patterns, so as to determine the prevention and control areas and enable the intelligent monitoring and early warning device to respond with early warnings.

[0011] More specifically, the step S102 includes the following steps:

[0012] Obtain the specified monitoring area and monitoring files for wheat planting, and extract the past environmental factor data and past scab infection data of the intelligent monitoring and early warning device for the specified monitoring area within a preset historical time period through the monitoring files;

[0013] Obtain established monitoring requirements when the intelligent monitoring and early warning device senses past environmental factor data and past scab infection data in a preset historical time period, and preset tensor dimensions and target monitoring ranks for each tensor dimension according to the established monitoring requirements;

[0014] A singular decomposition algorithm is introduced. At this time, the past environmental factor data and the past scab infection data are matrix-expanded in each tensor dimension to obtain a tensor flattened matrix for environmental factor monitoring. The tensor flattened matrix is ​​subjected to a time series singular decomposition in the singular decomposition algorithm based on the target monitoring rank.

[0015] Through time series singular decomposition, the time series tensor singular values ​​of each tensor dimension are obtained. According to the time series tensor singular values ​​of each tensor dimension, the core variable tensor structure of past environmental factor data and past ergot infection data is established with the preset time period as the alignment benchmark.

[0016] More specifically, the step S104 includes the following steps:

[0017] Based on the core variable tensor structure topology, a past spatiotemporal node graph of environmental factors and scab infection is constructed. A one-dimensional dilated causal convolution layer is introduced to convolute in the time dimension to extract the temporal changes of each past spatiotemporal node and generate several temporal convolution blocks.

[0018] An adaptive graph convolution layer is simultaneously introduced to calculate the spatial relationship between each past spatiotemporal node to obtain an adaptive adjacency matrix. Based on the adaptive adjacency matrix convolution, the dependency relationship between each past spatiotemporal node in space is captured to generate several spatial convolution blocks.

[0019] Obtaining wheat planting requirements, obtaining the advanced monitoring time segment of fusarium head blight infection in the specified monitoring area by the intelligent monitoring and early warning device based on the planting requirements, and obtaining real-time environmental factor data of the specified monitoring area in the advanced monitoring time segment through the intelligent monitoring and early warning device;

[0020] A convolution stacking order of real-time environmental factor data is constructed according to the advance monitoring time segment. Based on the convolution stacking order, several temporal convolution blocks and several spatial convolution blocks are alternately stacked to form a deep spatiotemporal prediction structure. According to the deep spatiotemporal prediction structure, an overview of the real-time distribution of ergot disease in the specified monitoring area when real-time environmental factor data is generated within the advance monitoring time segment is determined.

[0021] More specifically, the step S106 includes the following steps:

[0022] Obtaining a preset monitoring strategy of the intelligent monitoring and early warning device for the environmental factors of scab, and extracting the instantaneous monitoring time step of the intelligent monitoring and early warning device according to the preset monitoring strategy;

[0023] Based on the instantaneous monitoring time step, the advanced monitoring time segment is divided into several evenly distributed subsidiary time segments, and the real-time scab distribution corresponding to each subsidiary time segment is separated from the real-time scab distribution overview and marked as a sub-distribution profile;

[0024] The Gaussian model was introduced to calculate the Mahalanobis distance between each real-time scab sample in each sub-distribution profile. Based on the Mahalanobis distance, the spatial correlation between each real-time scab sample was fitted in the Gaussian model to obtain the semivariogram distribution function of the real-time scab sample.

[0025] A semivariogram interpolation model was constructed using the Kriging interpolation algorithm. Each half of the variogram distribution function was interpolated in the semivariogram interpolation model to obtain the Kriging hotspot interpolation equation for the real-time scab samples. The Kriging hotspot interpolation equation was solved to obtain the relative static proposed error matrix for each real-time scab sample in the corresponding subordinate time segment between each sub-distribution profile.

[0026] Based on the dynamic potential energy index constrained by the relative static proposed error matrix, for each real-time scab sample, the average dynamic layout flow of the neighborhood samples in each sub-distribution profile is deduced by minimizing the dynamic energy index. If the dynamic potential energy index is 0, the variational deduction operation is stopped and the layout trend clue field is output.

[0027] Obtain a geographic diagram of the specified monitoring area, construct a multidimensional space field of the specified monitoring area based on the geographic diagram, splice all sub-distribution profiles in the multidimensional space field according to the attached time segments based on the layout trend clue field, and generate a dynamic hotspot change map of the real-time ergot distribution in the specified monitoring area.

[0028] More specifically, the step S108 includes the following steps:

[0029] The potential relationship between the past environmental factor data and the past scab infection data is tested by using the timeline of past scab infection within the preset historical time period and the historical factor parameter interval generated by the corresponding past environmental factor data, and the existing occurrence pattern of scab affected by environmental factors is obtained;

[0030] Obtain a random sampling strategy for scab environmental factors from the intelligent monitoring and early warning device. Based on this random sampling strategy, the multidimensional space of the specified monitoring area is divided into N subspaces. At the same time, a thermal color gamut rating system suitable for scab risk assessment is obtained based on big data.

[0031] Constructing a binary color gamut search tree with existing occurrence rules based on the thermal color gamut rating system, obtaining a root stem basis and a current root stem value of the binary color gamut search tree, and traversing the hotspot chromaticity coefficient of each dynamic hotspot in the dynamic hotspot transition graph starting from the root stem basis;

[0032] If the hotspot chromaticity coefficient is greater than the current root value, the dynamic hotspot is iteratively inserted into the left trunk of the binary color gamut search tree to generate the left subtree; if the hotspot chromaticity coefficient is less than the current root value, the dynamic hotspot is iteratively inserted into the right trunk of the binary color gamut search tree to generate the right subtree;

[0033] Obtain the heights of the left subtree and the right subtree respectively, obtain the left tree height value and the right tree height value, calculate the balance deviation of each native trunk node on the binary search tree architecture based on the left tree height value and the right tree height value, and obtain several balance factors;

[0034] If the balance factor exceeds the dynamic balance factor threshold of the dynamic hotspot, the original trunk node where the balance factor is located is extracted and marked as the deleted trunk node, the leaf growth path of the deleted trunk node is obtained, and the height of the binary color gamut search tree is continuously updated from the leaf growth path upward until the balance factor does not exceed the dynamic balance factor threshold. The suitable thermal index for the proliferation of scab in each subspace area is output;

[0035] A thermal control query table is established based on planting needs. If a suitable thermal index can be found in the thermal control query table, the subspace area is marked as a control area and uploaded to the intelligent monitoring and early warning device to respond with an early warning signal.

[0036] More specifically, the method utilizes the timeline of past scab infection within a preset historical time period and the directed edges of historical factor parameter intervals correspondingly generated by past environmental factor data to test the potential occurrence relationship between past environmental factor data and past scab infection data, and obtains the existing occurrence pattern of scab affected by environmental factors, which specifically includes the following steps:

[0037] Extract the timeline of each past scab infection data that occurred in the monitoring area within the preset historical time period through the monitoring archive, and simultaneously obtain the historical factor parameter interval corresponding to each past environmental factor data within the preset historical time period;

[0038] Define past environmental factor data as Class I mileage points, define past scab infection data as Class II mileage points, preset anti-disturbance independent bins based on the occurrence timeline, and construct independent recursive condition sets for each Class I mileage point based on the historical factor parameter interval;

[0039] Discretely combine each Class I mileage point one by one to obtain several groups of discrete pairs of Class I mileage points. Test and calculate the degree of disturbance of Class II mileage points under the independent recursive condition set triggered by each group of discrete pairs of Class I mileage points, and obtain the disturbance frequency of each group of discrete pairs of Class I mileage points.

[0040] If the disturbance frequency is within the anti-disturbance independent bin, the discrete pairs of the first-class mileage points are deleted, and the independent recursive condition set triggered by the deleted discrete pairs of the first-class mileage points is marked as a separated set. If the disturbance frequency is not within the anti-disturbance independent bin, the discrete pairs of the first-class mileage points are skipped and the next set of judgment analysis is performed. Finally, the potential occurrence undirected edges of each first-class mileage point leading to each second-class mileage point are obtained.

[0041] Obtain the historical factor parameter intervals of each past scab infection data at each occurrence node on the occurrence timeline, define them as the hazard factor parameter intervals, and construct the sovereignty hazard conditions for each Class II mileage point based on the hazard factor parameter intervals;

[0042] If there is at least one or more sovereignty hazard conditions in the separation set, then the interval capacity based on the sovereignty hazard conditions is applied to the potential undirected edges corresponding to the second type of mileage points to add direction and strength, and obtain several potential occurrence directed edges;

[0043] A potential occurrence directed acyclic graph is constructed through several potential directed edges, and the existing occurrence rules of fusarium scabiei under the influence of environmental factors are determined based on the topological structure of the potential occurrence directed acyclic graph.

[0044] A second aspect of the present invention provides an intelligent monitoring and early warning device for environmental factors of scab, which is applied to an early warning method of any of the intelligent monitoring and early warning devices for environmental factors of scab, specifically comprising:

[0045] A Gibberellin environmental factor monitoring module, which is responsible for real-time monitoring of environmental factors around wheat planting in the field;

[0046] A monitoring and early warning module, which is used to monitor the control area of ​​unreasonable and unsuitable thermal index and respond with early warning signals;

[0047] A spatial modeling module, which is responsible for obtaining a geographical schematic diagram of a specified monitoring area and constructing a multidimensional spatial domain of the specified monitoring area based on the geographical schematic diagram;

[0048] A data storage and extraction module, the data storage and extraction module is used to store and extract the past environmental factor data and past scab infection data of the intelligent monitoring and early warning device for the specified monitoring area within a preset historical time period;

[0049] The data query module is used to query whether there is a suitable thermal index in the prevention and control thermal query table.

[0050] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are:

[0051] Through the tensor dimension of the established monitoring requirements, the past environmental factor data and the past scab infection data of the specified monitoring area within the preset historical time period are subjected to time series singular decomposition, and the core variable tensor structure of the past environmental factor data and the past scab infection data is obtained with the preset time period as the alignment benchmark; according to the core variable tensor structure of the environmental factor-scab infection past time and space node graph, the time series convolution blocks and spatial convolution blocks contained in the past time and space node graph are convolutionally extracted and alternately superimposed to obtain the real-time scab distribution overview when the specified monitoring area generates real-time environmental factor data within the advanced monitoring time segment ; Divide the real-time fusarium head blight distribution overview into multiple sub-distribution profiles, use semi-variation spatial correlation hotspots to interpolate each real-time fusarium head blight sample between each sub-distribution profile, and perform variational deduction of the average dynamic layout flow of the sub-distribution profiles to obtain a dynamic hotspot change map of the real-time fusarium head blight distribution within the specified monitoring area; test the directed connection between past environmental factors and past fusarium head blight infections, obtain the existing occurrence patterns of fusarium head blight under the influence of environmental factors, and update the dynamic hotspot change map based on the hazard rating thermodynamic balance of the existing occurrence patterns, thereby determining the prevention and control area and enabling the intelligent monitoring and early warning device to respond with an early warning. The present invention can perform real-time monitoring of fusarium head blight environmental factors during wheat planting and calculate the appropriateness and rationality of environmental factors to provide fusarium head blight growth. Based on the appropriateness and rationality, the intelligent monitoring and early warning device can make a timely prevention and control early warning response, thereby achieving scientific prevention and control of wheat planting and preventing fusarium head blight infection from reducing wheat planting quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] 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.

[0053] Figure 1 A flowchart of a first method of an early warning method of an intelligent monitoring and early warning device for environmental factors of scab is shown;

[0054] Figure 2 A second method flow chart of an early warning method of an intelligent monitoring and early warning device for scab environmental factors is shown;

[0055] Figure 3 The overall structure diagram of an intelligent monitoring and early warning device for environmental factors of ergot disease is shown. DETAILED DESCRIPTION

[0056] 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.

[0057] 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.

[0058] The first aspect of the present invention provides an early warning method of an intelligent monitoring and early warning device for scab environmental factors, such as Figure 1 As shown, the following steps are included:

[0059] S102: Performing time series singular decomposition on the past environmental factor data and the past scab infection data of the specified monitoring area within a preset historical time period using the tensor dimension of the established monitoring requirements, thereby obtaining a core variable tensor structure of the past environmental factor data and the past scab infection data aligned with the preset time period;

[0060] S104: Based on the past spatiotemporal node graph of the core variable tensor structure environmental factor-scab infection, convolutionally extract the temporal convolution blocks and spatial convolution blocks contained in the past spatiotemporal node graph and alternately superimpose them to obtain a real-time overview of scab distribution in the specified monitoring area when real-time environmental factor data is generated within the advanced monitoring time segment;

[0061] S106: Divide the real-time scab distribution overview into multiple sub-distribution profiles, interpolate each real-time scab sample between the sub-distribution profiles using semi-variation spatial correlation hotspots, and perform variational deduction on the average dynamic layout flow of the sub-distribution profiles to obtain a dynamic hotspot change map of the real-time scab distribution within the specified monitoring area;

[0062] S108: Test the directed connection between past environmental factors and potential past ergot infections, obtain the existing occurrence patterns of ergot affected by environmental factors, and update the dynamic hotspot change map based on the hazard rating thermodynamic balance of the existing occurrence patterns, so as to determine the prevention and control areas and enable the intelligent monitoring and early warning device to respond with early warnings.

[0063] More specifically, the step S102 includes the following steps:

[0064] Obtain the specified monitoring area and monitoring files for wheat planting, and extract the past environmental factor data and past scab infection data of the intelligent monitoring and early warning device for the specified monitoring area within a preset historical time period through the monitoring files;

[0065] Obtain established monitoring requirements when the intelligent monitoring and early warning device senses past environmental factor data and past scab infection data in a preset historical time period, and preset tensor dimensions and target monitoring ranks for each tensor dimension according to the established monitoring requirements;

[0066] A singular decomposition algorithm is introduced. At this time, the past environmental factor data and the past scab infection data are matrix-expanded in each tensor dimension to obtain a tensor flattened matrix for environmental factor monitoring. The tensor flattened matrix is ​​subjected to a time series singular decomposition in the singular decomposition algorithm based on the target monitoring rank.

[0067] Through time series singular decomposition, the time series tensor singular values ​​of each tensor dimension are obtained. According to the time series tensor singular values ​​of each tensor dimension, the core variable tensor structure of past environmental factor data and past ergot infection data is established with the preset time period as the alignment benchmark.

[0068] It should be noted that certain environmental conditions generated by environmental factors during wheat planting can make Gibberella spread and grow more comfortably. Therefore, for the prevention and control of Gibberella environmental factors, it is usually necessary to use predictive means to analyze whether the gain level of real-time environmental factors in helping Gibberella spread is reasonable. The prediction of Gibberella growth status under real-time environmental factor conditions mainly relies on the information support of past historical data. However, there is a situation where the tensor dimensions between the extracted past environmental factor data and the past Gibberella infection data cannot be aligned in time series. For example, in the real-time prediction process, a past Gibberella infection data at a certain historical time node relies on a spatial tensor, while a past environmental factor data that creates the spatial tensor is not for the historical time node, resulting in a time series matching error in the data information prediction, which makes the real-time prediction have a large information asymmetry phenomenon, and produces misplaced patterns, vector and matrix calculation errors for the Gibberella distribution under the real-time changes of historical environmental factors, reducing the credibility of the real-time distribution prediction of Gibberella. To this end, this method presets the tensor dimension and the corresponding target monitoring rank based on the established monitoring requirements. The tensor dimension can be any dimension, depending on the monitoring indicators of environmental factors and ergot infection in the established monitoring requirements. For example, if the monitoring of environmental factor data by the intelligent monitoring and early warning device in the historical time period is based on the spatial and environmental variable levels, then the tensor dimension is a three-dimensional tensor of longitude × latitude × environmental variables; and the target monitoring rank specifies the main feature dimension to be retained on each tensor dimension, which is used to control the component composition of the tensor decomposition. Subsequently, the past environmental factor data and past ergot infection data are matrix-expanded on the tensor dimension, thereby spreading the tensor into a matrix along each tensor dimension. The tensor symmetry between the past environmental factor data and the past ergot infection data can be clearly displayed in matrix form, thereby improving the accuracy of tensor unification between different data variables.

[0069] It should be noted that the past environmental factor data includes temperature, humidity, light intensity, soil texture, and soil pH. The tensor flattening matrix spread out by the singular decomposition algorithm is then used to extract the most important principal components (orthogonal directions) under each tensor dimension. This effectively removes redundant or low-energy dimensions and avoids interference from unnecessary tensor component noise. This allows the tensor components generated by the past environmental factor data based on a unified time series to be more correctly aligned with the past ergot infection data. This ensures the core tensor alignment of the past environmental factor data and the past ergot infection data under the constraints of a preset time period. This provides a reliable time series data structure foundation for subsequent real-time predictions of ergot distribution based on environmental factors, avoids prediction disorder, and improves the prediction accuracy of ergot distribution by intelligent monitoring and early warning devices.

[0070] More specifically, the step S104 is as follows: Figure 2 As shown, the specific steps include:

[0071] S202: Based on the core variable tensor structure topology, a past spatiotemporal node graph of environmental factors and scab infection is constructed. A one-dimensional dilated causal convolution layer is introduced to convolute in the time dimension to extract the temporal changes of each past spatiotemporal node, generating several temporal convolution blocks.

[0072] S204: Synchronously introduce an adaptive graph convolution layer to calculate the spatial relationship between each past spatiotemporal node to obtain an adaptive adjacency matrix. Based on the adaptive adjacency matrix convolution, the dependency relationship between each past spatiotemporal node in space is captured to generate several spatial convolution blocks.

[0073] S206: Obtaining wheat planting requirements, obtaining an advanced monitoring time segment for fusarium head blight infection in a specified monitoring area by an intelligent monitoring and early warning device based on the planting requirements, and obtaining real-time environmental factor data for the specified monitoring area in the advanced monitoring time segment by the intelligent monitoring and early warning device;

[0074] S208: Construct a convolution stacking order of real-time environmental factor data according to the advance monitoring time segment, alternately stack a number of temporal convolution blocks and a number of spatial convolution blocks based on the convolution stacking order to form a deep spatiotemporal prediction structure, and determine a real-time overview of the distribution of ergot disease in the specified monitoring area when the real-time environmental factor data is generated within the advance monitoring time segment according to the deep spatiotemporal prediction structure.

[0075] It should be noted that because both environmental factors and scab have temporal and spatial characteristics, traditional methods for predicting the relationship between environmental factors and scab infection typically require extensive training and validation using historical data. These methods rely on the data satisfying certain statistical properties, such as linear spatiotemporal relationships and spatiotemporal normal distributions. This significantly reduces the performance of traditional models in predicting the spatiotemporal distribution of scab. Furthermore, these models struggle to capture nonlinear relationships between spatiotemporal data and are less capable of recognizing the spatiotemporal patterns of complex environmental factors affecting scab. Consequently, they struggle to cope with dynamic changes in spatiotemporal data, such as structural mutations and seasonal variations, and thus fail to ensure high accuracy in predicting the real-time distribution of scab. To address this, our method first constructs a historical spatiotemporal node graph of environmental factors and scab infection based on the core variable tensor structure topology. This graph depicts the temporal characteristics and dependencies of past scab transmission and infection under the influence of environmental factors, providing a key basis for analyzing the trends in historical data. Then, a one-dimensional dilated causal convolutional layer is used to convolve in the time dimension to extract the temporal features of each past spatiotemporal node. This is done by stacking multiple layers to model and capture the long-term dependency of past environmental factors on past ergot infection, from short-term to long-term temporal changes, while only looking at the past to maintain predictive causality. Simultaneously, an adaptive graph convolutional layer is used to calculate the spatial relationships of each past spatiotemporal node to obtain an adaptive adjacency matrix, thereby enabling convolution to capture the interactive dependency relationships between each past spatiotemporal node in space. This is more flexible than traditional prediction models. On the one hand, it can automatically learn hidden spatial dependencies, enhance generalization capabilities, and combine prior knowledge and learning structures to improve prediction accuracy at the spatial level. On the other hand, it reduces unnecessary computational steps and improves prediction efficiency. The adaptive adjacency matrix can be an adaptive learning graph, which makes the prediction of ergot distribution independent of a fixed graph structure and has the ability to automatically discover potential node relationships.

[0076] It should be noted that the convolution of time and space will generate several time series convolution blocks and several space convolution blocks. These convolution blocks reveal the context of the historical development trend of the spread and infection of past ergot disease under the influence of past environmental factors, enhance the prediction expression ability, and prevent the gradient from disappearing. Therefore, it is only necessary to alternately stack these convolution blocks according to the real-time environmental factor data and the time series of the real-time data monitored to generate the real-time distribution status of ergot disease with the spatiotemporal trend of the real-time environmental factor as the main context. This method can be used to analyze the historical development trend of historical environmental factor data and historical ergot disease infection data based on the spatiotemporal convolution of the core variable tensor structure, so that the prediction of the real-time environmental factor monitoring data is more in line with the known spatiotemporal characteristics of the ergot disease distribution, thereby improving the accuracy of the real-time ergot disease distribution prediction.

[0077] More specifically, the step S106 includes the following steps:

[0078] Obtaining a preset monitoring strategy of the intelligent monitoring and early warning device for the environmental factors of scab, and extracting the instantaneous monitoring time step of the intelligent monitoring and early warning device according to the preset monitoring strategy;

[0079] Based on the instantaneous monitoring time step, the advanced monitoring time segment is divided into several evenly distributed subsidiary time segments, and the real-time scab distribution corresponding to each subsidiary time segment is separated from the real-time scab distribution overview and marked as a sub-distribution profile;

[0080] The Gaussian model was introduced to calculate the Mahalanobis distance between each real-time scab sample in each sub-distribution profile. Based on the Mahalanobis distance, the spatial correlation between each real-time scab sample was fitted in the Gaussian model to obtain the semivariogram distribution function of the real-time scab sample.

[0081] A semivariogram interpolation model was constructed using the Kriging interpolation algorithm. Each half of the variogram distribution function was interpolated in the semivariogram interpolation model to obtain the Kriging hotspot interpolation equation for the real-time scab samples. The Kriging hotspot interpolation equation was solved to obtain the relative static proposed error matrix for each real-time scab sample in the corresponding subordinate time segment between each sub-distribution profile.

[0082] Based on the dynamic potential energy index constrained by the relative static proposed error matrix, for each real-time scab sample, the average dynamic layout flow of the neighborhood samples in each sub-distribution profile is deduced by minimizing the dynamic energy index. If the dynamic potential energy index is 0, the variational deduction operation is stopped and the layout trend clue field is output.

[0083] Obtain a geographic diagram of the specified monitoring area, construct a multidimensional space field of the specified monitoring area based on the geographic diagram, splice all sub-distribution profiles in the multidimensional space field according to the attached time segments based on the layout trend clue field, and generate a dynamic hotspot change map of the real-time ergot distribution in the specified monitoring area.

[0084] It should be noted that the main body of fusarium spores is wheat pathogen conidia, which have a certain point distribution and infection trajectory when spreading in real time. However, due to its small size, it is difficult to observe with the naked eye. Therefore, in order to more accurately capture and clarify the infection trend and distribution pattern of real-time fusarium spores, this method obtains sub-distribution profiles at different times by evenly fragmenting the acquired real-time fusarium spore distribution overview according to the instantaneous monitoring time step of the intelligent monitoring and early warning device, and performs semi-variation distribution analysis and hotspot interpolation on each real-time fusarium spore sample contained in each sub-distribution profile, so as to generate the hotspot distribution of real-time fusarium spores at specific time nodes in each sub-distribution profile, thereby visualizing the temporal static activity of fusarium spores at different times in the advanced monitoring time segment. Next, the relative static proposed error matrix of each real-time fusarium sample in the corresponding subordinate time segment between each sub-distribution profile is calculated, and the average dynamic layout flow of each sub-distribution profile in the advance monitoring time segment is deduced based on the relative static proposed error matrix. This is a temporal dynamic clue that connects the real-time fusarium activity in all sub-distribution profiles, thereby establishing a dynamic connection between the real-time fusarium activity distribution at different times, and finally forming a hotspot activity change map that expresses the dynamic distribution of real-time fusarium in the specified monitoring area according to the advance monitoring time segment. The dynamic hotspot change map constructed by this method can clearly describe the real-time infection direction and distribution scale of fusarium in a specific time period, and can visualize the distribution status of real-time fusarium under the influence of real-time environmental factors, which can effectively improve the accuracy of the analysis of the suitability of environmental factors for fusarium growth and avoid false alarms and omissions in device warnings.

[0085] It should be noted that the semivariogram distribution function can quantitatively describe the spatial autocorrelation of all real-time ergot samples in a sub-distribution profile, that is, how the sample attribute value changes with increasing distance. This function is the core parameter of hotspot interpolation, and the assignment quantifies the hotspot location of real-time ergot samples. The Kriging hotspot interpolation equation established based on the semivariogram distribution function and the spatial location of the samples can ensure that the hotspot interpolation result is a linear unbiased estimate, improving the accuracy of hotspot positioning while obtaining a static confidence assessment between each hotspot sample, namely the relative static proposed error matrix. This relative static proposed error matrix expresses the dynamic smoothness of the real-time ergot samples between each sub-distribution profile. When the smoothness is high, this method uses variational inference to minimize the dynamic potential energy index to track the dynamic layout trajectory of hotspot samples as the time series activities of the advance monitoring time segment, so that the sub-distribution profiles at different times have a reasonable and smooth time series connection basis, making the dynamic hotspot changes in the advance monitoring time segment more accurate and reliable.

[0086] More specifically, the step S108 includes the following steps:

[0087] The potential relationship between the past environmental factor data and the past scab infection data is tested by using the timeline of past scab infection within the preset historical time period and the historical factor parameter interval generated by the corresponding past environmental factor data, and the existing occurrence pattern of scab affected by environmental factors is obtained;

[0088] Obtain a random sampling strategy for scab environmental factors from the intelligent monitoring and early warning device. Based on this random sampling strategy, the multidimensional space of the specified monitoring area is divided into N subspaces. At the same time, a thermal color gamut rating system suitable for scab risk assessment is obtained based on big data.

[0089] Constructing a binary color gamut search tree with existing occurrence rules based on the thermal color gamut rating system, obtaining a root stem basis and a current root stem value of the binary color gamut search tree, and traversing the hotspot chromaticity coefficient of each dynamic hotspot in the dynamic hotspot transition graph starting from the root stem basis;

[0090] If the hotspot chromaticity coefficient is greater than the current root value, the dynamic hotspot is iteratively inserted into the left trunk of the binary color gamut search tree to generate the left subtree; if the hotspot chromaticity coefficient is less than the current root value, the dynamic hotspot is iteratively inserted into the right trunk of the binary color gamut search tree to generate the right subtree;

[0091] Obtain the heights of the left subtree and the right subtree respectively, obtain the left tree height value and the right tree height value, calculate the balance deviation of each native trunk node on the binary search tree architecture based on the left tree height value and the right tree height value, and obtain several balance factors;

[0092] If the balance factor exceeds the dynamic balance factor threshold of the dynamic hotspot, the original trunk node where the balance factor is located is extracted and marked as the deleted trunk node, the leaf growth path of the deleted trunk node is obtained, and the height of the binary color gamut search tree is continuously updated from the leaf growth path upward until the balance factor does not exceed the dynamic balance factor threshold. The suitable thermal index for the proliferation of scab in each subspace area is output;

[0093] A thermal control query table is established based on planting needs. If a suitable thermal index can be found in the thermal control query table, the subspace area is marked as a control area and uploaded to the intelligent monitoring and early warning device to respond with an early warning signal.

[0094] It should be noted that when the dynamic hotspot change map shows that the frequency, number and range of hotspot activities in a certain area are large, it means that the environmental factors appearing here are more suitable for the growth and spread of ergot fungi. However, traditional early warning methods usually use the known real-time distribution of ergot fungi to directly assess the hazard level. This method makes it difficult to take into account the suitability and rationality of environmental factors for ergot infection, and thus cannot make correct early warning judgments for the real-time monitored environmental factor data. To this end, this method first uses the timeline of past ergot infection within a preset historical time period and the historical factor parameter intervals generated by the past environmental factor data to perform a directed edge test on the past environmental factor data and the past ergot infection data for potential occurrence connections, thereby further accurately capturing the existing occurrence patterns of ergot affected by environmental factors. Next, a thermal color gamut rating system that is consistent with the risk assessment of fusarium spp. is obtained. The thermal color gamut rating system is a reference to logical rules that accurately define the risk level of fusarium spp. through thermal colors. For example, thermal red represents a high risk level of fusarium spp. Based on the thermal color gamut rating system, a binary color gamut search tree with existing occurrence rules is constructed, which can assign corresponding thermal colors to the trends of environmental factors that lead to different fusarium spp. infections. This allows the real-time fusarium spp. distribution expressed using a dynamic hotspot transition map to accurately index the thermal color rating of the corresponding hazard risk in the binary color gamut search tree. Among them, if the hotspot chromaticity coefficient is greater than the current root value, it means that the RPG color value of the dynamic hotspot in the color gamut is low, in other words, it is in the light-colored area. Therefore, the dynamic hotspot is iteratively inserted into the left trunk end of the binary color gamut search tree, which is mainly responsible for recording light colors, thereby generating a left subtree belonging to the lighter-colored hotspot. If not, the dynamic hotspot has a high RPG color value within the color gamut—in other words, it's in the dark region. Therefore, the dynamic hotspot is iteratively inserted into the right trunk, which primarily records dark colors, to generate a right subtree belonging to the darker hotspot. The tree structure formed by the left and right subtrees now represents the global color gamut description of the dynamic hotspot. This method accurately assigns color values ​​to dynamic hotspot samples in real-time ergot distribution, enabling more precise rating of the risk of damage from corresponding environmental factors.

[0095] It should be noted that the balance factor is used to measure the balance deviation of the color gamut change relative to the next time series when each dynamic hotspot changes in time series, reflecting the reliability of the color gamut update of the intelligent monitoring and early warning device for real-time dynamic monitoring of ergot disease. If the balance factor exceeds the dynamic balance factor threshold of the dynamic hotspot, it means that the color gamut update error of the dynamic hotspot is in the given time series change, and the RPG color value of the dynamic hotspot should be re-updated. This method deletes the native trunk node where the balance factor is located, and continuously updates the height of the binary color gamut search tree upward according to the original leaf growth path, thereby re-assigning the RPG color value of the next time series node, thereby achieving the dynamic hazard rating update effect of the environmental factors under the real-time distribution of ergot disease. Among them, if the suitable thermal index can be queried in the prevention and control thermal query table, it means that the environmental factors in the area are more conducive to the growth, reproduction, spread and other behaviors of ergot pathogens. Therefore, the area is regarded as a high pathogen hazard area, and the intelligent monitoring and early warning device needs to be controlled to issue corresponding early warning prompts to the area. This method can assign colors to real-time environmental factors using the real-time ergot distribution hazard rating method in the thermal color range, thereby accurately evaluating whether the environmental factors in the region are suitable for the growth of ergot, making prevention and control warnings more accurate, avoiding false alarms or missed alarms, and improving warning performance.

[0096] More specifically, the method utilizes the timeline of past scab infection within a preset historical time period and the directed edges of historical factor parameter intervals correspondingly generated by past environmental factor data to test the potential occurrence relationship between past environmental factor data and past scab infection data, and obtains the existing occurrence pattern of scab affected by environmental factors, which specifically includes the following steps:

[0097] Extract the timeline of each past scab infection data that occurred in the monitoring area within the preset historical time period through the monitoring archive, and simultaneously obtain the historical factor parameter interval corresponding to each past environmental factor data within the preset historical time period;

[0098] Define past environmental factor data as Class I mileage points, define past scab infection data as Class II mileage points, preset anti-disturbance independent bins based on the occurrence timeline, and construct independent recursive condition sets for each Class I mileage point based on the historical factor parameter interval;

[0099] Discretely combine each Class I mileage point one by one to obtain several groups of discrete pairs of Class I mileage points. Test and calculate the degree of disturbance of Class II mileage points under the independent recursive condition set triggered by each group of discrete pairs of Class I mileage points, and obtain the disturbance frequency of each group of discrete pairs of Class I mileage points.

[0100] If the disturbance frequency is within the anti-disturbance independent bin, the discrete pairs of the first-class mileage points are deleted, and the independent recursive condition set triggered by the deleted discrete pairs of the first-class mileage points is marked as a separated set. If the disturbance frequency is not within the anti-disturbance independent bin, the discrete pairs of the first-class mileage points are skipped and the next set of judgment analysis is performed. Finally, the potential occurrence undirected edges of each first-class mileage point leading to each second-class mileage point are obtained.

[0101] Obtain the historical factor parameter intervals of each past scab infection data at each occurrence node on the occurrence timeline, define them as the hazard factor parameter intervals, and construct the sovereignty hazard conditions for each Class II mileage point based on the hazard factor parameter intervals;

[0102] If there is at least one or more sovereignty hazard conditions in the separation set, then the interval capacity based on the sovereignty hazard conditions is applied to the potential undirected edges corresponding to the second type of mileage points to add direction and strength, and obtain several potential occurrence directed edges;

[0103] A potential occurrence directed acyclic graph is constructed through several potential directed edges, and the existing occurrence rules of fusarium scabiei under the influence of environmental factors are determined based on the topological structure of the potential occurrence directed acyclic graph.

[0104] It should be noted that the step of testing the potential occurrence connection directed edges of past environmental factor data and past ergot infection data is that this method obtains the timeline of the occurrence of each past ergot infection data in the preset historical time period in the specified monitoring area, and the historical factor parameter interval generated by each past environmental factor data in the preset historical time period. Because some past environmental factors jointly generate historical factor parameter intervals and do not lead to the occurrence of past ergot infection, this method first tests and calculates the disturbance degree of each group of discrete pairs of first-class mileage points on the second-class mileage points under the triggering independent recursive condition set. If the disturbance frequency is within the anti-disturbance independent bin, it means that the historical parameters imposed by this group of past environmental factors are on the timeline of past ergot infection, that is, this group of past environmental factors caused the occurrence of past ergot infection. Therefore, this group of past environmental factors does not have the independence of changing ergot infection. Therefore, this group of discrete pairs of first-class mileage points is deleted, and the imposed historical factor parameter interval is regarded as a separate set. If it is not located, it means that this group of past environmental factors does not affect the infection of ergot disease when producing specific historical factor parameters, and is independent. Therefore, the past environmental factor data combination of this group of mileage points is ignored, so that the potential irregular connection between past environmental factors and the occurrence of ergot disease infection can be preliminarily clarified. The potential occurrence of undirected edges lays the foundation for the subsequent exploration of rules.

[0105] It should be noted that after clarifying the fuzzy potential regular connections, it is necessary to use the historical factor parameter intervals of past ergot occurrences to further explore and reveal the specific direction and strength of the potential undirected connection. To this end, this method constructs the sovereign hazard conditions for each Class II mileage point based on the hazard factor parameter intervals. If the separated set contains at least one or more sovereign hazard conditions, it means that the historical factor parameters that led to the past ergot occurrence are highly consistent with the historical factor parameters imposed by the past environmental factor data with potential undirected connections. Therefore, it can be basically determined that the past environmental factors at the time of occurrence caused the past ergot infection. Therefore, based on the interval capacity of the sovereign hazard conditions, the direction and strength of the potential undirected edge corresponding to the Class II mileage point are further imposed, thereby clarifying the existing occurrence regularity of ergot under the influence of environmental factors. This method can explore and analyze the potential occurrence regularity between historical environmental factor data and historical ergot infection data, providing a reliable and credible hazard rating logical clue for the growth suitability of real-time environmental factors for real-time ergot, making the hazard rating more accurate and correct.

[0106] The second aspect of the present invention provides an intelligent monitoring and early warning device for environmental factors of scab, which is applied to the early warning method of the intelligent monitoring and early warning device for environmental factors of scab, such as Figure 3 As shown, specifically including:

[0107] Gibberellic acid environmental factor monitoring module 1011, which is responsible for real-time monitoring of environmental factors around wheat planting in the field;

[0108] A monitoring and early warning module 1012 is used to monitor the control area of ​​unreasonable and unsuitable thermal index and respond with an early warning signal;

[0109] A spatial modeling module 1013 is responsible for obtaining a geographic diagram of a specified monitoring area and constructing a multi-dimensional spatial domain of the specified monitoring area based on the geographic diagram;

[0110] A data storage and extraction module 1014 is used to store and extract past environmental factor data and past scab infection data of the intelligent monitoring and early warning device for a specified monitoring area within a preset historical time period;

[0111] The data query module is used to query whether there is a suitable thermal index in the prevention and control thermal query table.

[0112] 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 early warning method for an intelligent monitoring and early warning device for scab environmental factors, characterized in that: The following steps are involved: S102: Performing time series singular decomposition on the past environmental factor data and the past scab infection data of the specified monitoring area within a preset historical time period using the tensor dimension of the established monitoring requirements, thereby obtaining a core variable tensor structure of the past environmental factor data and the past scab infection data aligned with the preset time period; S104: Based on the past spatiotemporal node graph of the core variable tensor structure environmental factor-scab infection, convolutionally extract the temporal convolution blocks and spatial convolution blocks contained in the past spatiotemporal node graph and alternately superimpose them to obtain a real-time overview of scab distribution in the specified monitoring area when real-time environmental factor data is generated within the advanced monitoring time segment; S106: Divide the real-time scab distribution overview into multiple sub-distribution profiles, interpolate each real-time scab sample between the sub-distribution profiles using semi-variation spatial correlation hotspots, and perform variational deduction on the average dynamic layout flow of the sub-distribution profiles to obtain a dynamic hotspot change map of the real-time scab distribution within the specified monitoring area; S108: Test the directed connection between past environmental factors and potential past ergot infections, obtain the existing occurrence patterns of ergot affected by environmental factors, and update the dynamic hotspot change map based on the hazard rating thermodynamic balance of the existing occurrence patterns, so as to determine the prevention and control areas and enable the intelligent monitoring and early warning device to respond with early warnings.

2. The early warning method of the intelligent monitoring and early warning device for scab environmental factors according to claim 1, characterized in that: The step S102 specifically includes the following steps: Obtain the specified monitoring area and monitoring files for wheat planting, and extract the past environmental factor data and past scab infection data of the intelligent monitoring and early warning device for the specified monitoring area within a preset historical time period through the monitoring files; Obtain established monitoring requirements when the intelligent monitoring and early warning device senses past environmental factor data and past scab infection data in a preset historical time period, and preset tensor dimensions and target monitoring ranks for each tensor dimension according to the established monitoring requirements; A singular decomposition algorithm is introduced. At this time, the past environmental factor data and the past scab infection data are matrix-expanded in each tensor dimension to obtain a tensor flattened matrix for environmental factor monitoring. The tensor flattened matrix is ​​subjected to a time series singular decomposition in the singular decomposition algorithm based on the target monitoring rank. Through time series singular decomposition, the time series tensor singular values ​​of each tensor dimension are obtained. According to the time series tensor singular values ​​of each tensor dimension, the core variable tensor structure of past environmental factor data and past ergot infection data is established with the preset time period as the alignment benchmark.

3. The early warning method of the intelligent monitoring and early warning device for scab environmental factors according to claim 1, characterized in that: The step S104 specifically includes the following steps: Based on the core variable tensor structure topology, a past spatiotemporal node graph of environmental factors and scab infection is constructed. A one-dimensional dilated causal convolution layer is introduced to convolute in the time dimension to extract the temporal changes of each past spatiotemporal node and generate several temporal convolution blocks. An adaptive graph convolution layer is simultaneously introduced to calculate the spatial relationship between each past spatiotemporal node to obtain an adaptive adjacency matrix. Based on the adaptive adjacency matrix convolution, the dependency relationship between each past spatiotemporal node in space is captured to generate several spatial convolution blocks. Obtaining wheat planting requirements, obtaining the advanced monitoring time segment of fusarium head blight infection in the specified monitoring area by the intelligent monitoring and early warning device based on the planting requirements, and obtaining real-time environmental factor data of the specified monitoring area in the advanced monitoring time segment through the intelligent monitoring and early warning device; A convolution stacking order of real-time environmental factor data is constructed according to the advance monitoring time segment. Based on the convolution stacking order, several temporal convolution blocks and several spatial convolution blocks are alternately stacked to form a deep spatiotemporal prediction structure. According to the deep spatiotemporal prediction structure, an overview of the real-time distribution of ergot disease in the specified monitoring area when real-time environmental factor data is generated within the advance monitoring time segment is determined.

4. The early warning method of the intelligent monitoring and early warning device for scab environmental factors according to claim 1, characterized in that: The step S106 specifically includes the following steps: Obtaining a preset monitoring strategy of the intelligent monitoring and early warning device for the environmental factors of scab, and extracting the instantaneous monitoring time step of the intelligent monitoring and early warning device according to the preset monitoring strategy; Based on the instantaneous monitoring time step, the advanced monitoring time segment is divided into several evenly distributed subsidiary time segments, and the real-time scab distribution corresponding to each subsidiary time segment is separated from the real-time scab distribution overview and marked as a sub-distribution profile; The Gaussian model was introduced to calculate the Mahalanobis distance between each real-time scab sample in each sub-distribution profile. Based on the Mahalanobis distance, the spatial correlation between each real-time scab sample was fitted in the Gaussian model to obtain the semivariogram distribution function of the real-time scab sample. A semivariogram interpolation model was constructed using the Kriging interpolation algorithm. Each half of the variogram distribution function was interpolated in the semivariogram interpolation model to obtain the Kriging hotspot interpolation equation for the real-time scab samples. The Kriging hotspot interpolation equation was solved to obtain the relative static proposed error matrix for each real-time scab sample in the corresponding subordinate time segment between each sub-distribution profile. Based on the dynamic potential energy index constrained by the relative static proposed error matrix, for each real-time scab sample, the average dynamic layout flow of the neighborhood samples in each sub-distribution profile is deduced by minimizing the dynamic energy index. If the dynamic potential energy index is 0, the variational deduction operation is stopped and the layout trend clue field is output. Obtain a geographic diagram of the specified monitoring area, construct a multidimensional space field of the specified monitoring area based on the geographic diagram, splice all sub-distribution profiles in the multidimensional space field according to the attached time segments based on the layout trend clue field, and generate a dynamic hotspot change map of the real-time ergot distribution in the specified monitoring area.

5. The early warning method of the intelligent monitoring and early warning device for scab environmental factors according to claim 1, characterized in that: The step S108 specifically includes the following steps: The potential relationship between the past environmental factor data and the past scab infection data is tested by using the timeline of past scab infection within the preset historical time period and the historical factor parameter interval generated by the corresponding past environmental factor data, and the existing occurrence pattern of scab affected by environmental factors is obtained; Obtain a random sampling strategy for scab environmental factors from the intelligent monitoring and early warning device. Based on this random sampling strategy, the multidimensional space of the specified monitoring area is divided into N subspaces. At the same time, a thermal color gamut rating system suitable for scab risk assessment is obtained based on big data. Constructing a binary color gamut search tree with existing occurrence rules based on the thermal color gamut rating system, obtaining a root stem basis and a current root stem value of the binary color gamut search tree, and traversing the hotspot chromaticity coefficient of each dynamic hotspot in the dynamic hotspot transition graph starting from the root stem basis; If the hotspot chromaticity coefficient is greater than the current root value, the dynamic hotspot is iteratively inserted into the left trunk of the binary color gamut search tree to generate the left subtree; if the hotspot chromaticity coefficient is less than the current root value, the dynamic hotspot is iteratively inserted into the right trunk of the binary color gamut search tree to generate the right subtree; Obtain the heights of the left subtree and the right subtree respectively, obtain the left tree height value and the right tree height value, calculate the balance deviation of each native trunk node on the binary search tree architecture based on the left tree height value and the right tree height value, and obtain several balance factors; If the balance factor exceeds the dynamic balance factor threshold of the dynamic hotspot, the original trunk node where the balance factor is located is extracted and marked as the deleted trunk node, the leaf growth path of the deleted trunk node is obtained, and the height of the binary color gamut search tree is continuously updated from the leaf growth path upward until the balance factor does not exceed the dynamic balance factor threshold. The suitable thermal index for the proliferation of scab in each subspace area is output; A thermal control query table is established based on planting needs. If a suitable thermal index can be found in the thermal control query table, the subspace area is marked as a control area and uploaded to the intelligent monitoring and early warning device to respond with an early warning signal.

6. The early warning method of the intelligent monitoring and early warning device for scab environmental factors according to claim 5, characterized in that: The method uses the timeline of past scab infection within a preset historical time period and the directed edges of the historical factor parameter intervals correspondingly generated by the past environmental factor data to test the potential occurrence relationship between the past environmental factor data and the past scab infection data, and obtains the existing occurrence pattern of scab affected by environmental factors, which specifically includes the following steps: Extract the timeline of each past scab infection data that occurred in the monitoring area within the preset historical time period through the monitoring archive, and simultaneously obtain the historical factor parameter interval corresponding to each past environmental factor data within the preset historical time period; Define past environmental factor data as Class I mileage points, define past scab infection data as Class II mileage points, preset anti-disturbance independent bins based on the occurrence timeline, and construct independent recursive condition sets for each Class I mileage point based on the historical factor parameter interval; Discretely combine each Class I mileage point one by one to obtain several groups of discrete pairs of Class I mileage points. Test and calculate the degree of disturbance of Class II mileage points under the independent recursive condition set triggered by each group of discrete pairs of Class I mileage points, and obtain the disturbance frequency of each group of discrete pairs of Class I mileage points. If the disturbance frequency is within the anti-disturbance independent bin, the discrete pairs of the first-class mileage points are deleted, and the independent recursive condition set triggered by the deleted discrete pairs of the first-class mileage points is marked as a separated set. If the disturbance frequency is not within the anti-disturbance independent bin, the discrete pairs of the first-class mileage points are skipped to perform the next set of judgment analysis, and finally the potential occurrence undirected edges of each first-class mileage point leading to each second-class mileage point are obtained. Obtain the historical factor parameter intervals of each past scab infection data at each occurrence node on the occurrence timeline, define them as the hazard factor parameter intervals, and construct the sovereignty hazard conditions for each Class II mileage point based on the hazard factor parameter intervals; If there is at least one or more sovereignty hazard conditions in the separation set, then the interval capacity based on the sovereignty hazard conditions is applied to the potential undirected edges corresponding to the second type of mileage points to add direction and strength, and obtain several potential occurrence directed edges; A potential occurrence directed acyclic graph is constructed through several potential directed edges, and the existing occurrence rules of fusarium scabiei under the influence of environmental factors are determined based on the topological structure of the potential occurrence directed acyclic graph.

7. An intelligent monitoring and early warning device for environmental factors of scab, applied to the early warning method of an intelligent monitoring and early warning device for environmental factors of scab according to any one of claims 1 to 6, characterized in that: Specifically include: A Gibberellin environmental factor monitoring module, which is responsible for real-time monitoring of environmental factors around wheat planting in the field; A monitoring and early warning module, which is used to monitor the control area of ​​unreasonable and unsuitable thermal index and respond with early warning signals; A spatial modeling module, which is responsible for obtaining a geographical schematic diagram of a specified monitoring area and constructing a multidimensional spatial domain of the specified monitoring area based on the geographical schematic diagram; A data storage and extraction module, the data storage and extraction module is used to store and extract the past environmental factor data and past scab infection data of the intelligent monitoring and early warning device for the specified monitoring area within a preset historical time period; The data query module is used to query whether there is a suitable thermal index in the prevention and control thermal query table.

Images