Monitoring optimization system for garden plant diseases and insect pests

Abstract

The invention relates to the field of pest monitoring, in particular to a monitoring optimization system for garden plant diseases and insect pests, and the system comprises a data analysis unit which is used for determining a hotspot distribution state according to a historical hotspot aggregation degree and a historical hotspot number, and determining a region division mode according to the hotspot distribution state; the first division unit is used for executing hotspot reference division to obtain a first division area and a second division area; the second division unit is used for executing path reference division to determine a multi-flow path and a region range area corresponding to the multi-flow path; the data storage unit is used for responding to different storage conditions so as to select the storage modules corresponding to different insect pest related data; the insect pest monitoring effect is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of pest monitoring, and more particularly to an optimized monitoring system for diseases and pests of garden plants. Background Technology

[0002] The invasion of pests and diseases is one of the key factors restricting the sustainable development of forestry. The reasonable division of monitoring areas is a crucial link in pest and disease control. The division of monitoring areas can make the subsequent analysis of pest data in the divided areas more accurate, such as the determination of the severity of pests in the divided areas. However, existing technologies usually use a single indicator, such as tree species distribution or terrain conditions, to make rough divisions. Rough divisions can easily lead to the inability of pest data analysis accuracy to meet the actual needs, resulting in the omission of some pest data in the analysis. Therefore, improving the accuracy and scientific nature of regional division is an urgent prerequisite for achieving precise control of forest pests and diseases.

[0003] Chinese Patent Publication No. CN119445359A discloses a method for determining the distribution area of ​​forest pests and diseases, including: acquiring remote sensing images of forest-covered areas using UAV remote sensing technology; assessing whether the pest and disease status of the forest-covered areas meets requirements based on the remote sensing images; if not, sequentially dividing the forest-covered areas using grid division, ring division, and random division methods to obtain grid-divided areas, grid division values, ring-divided areas, ring division values, random division areas, and random division values. This invention employs multiple division methods to process the forest-covered areas, ultimately selecting the division area with the highest coverage rate of pest and disease areas, thus improving the accuracy of forest pest and disease area division. However, while the above technical solution discloses different division methods for forest-covered areas where pest and disease status does not meet requirements, they all use pest coefficients to divide areas into different shapes. When continuous pest monitoring is required and the pest coefficient is below a preset pest threshold, the inability to select a division method according to the actual scene characteristics leads to poor area division results, which in turn easily results in poor pest monitoring. Summary of the Invention

[0004] Therefore, the present invention provides a monitoring and optimization system for garden plant diseases and pests, which overcomes the problem in the prior art that when continuous pest monitoring is required and the pest coefficient is lower than the preset pest threshold, the system cannot select the division method according to the actual scene characteristics, resulting in poor area division effect and thus poor pest monitoring effect.

[0005] To achieve the above objectives, the present invention provides a monitoring and optimization system for diseases and pests of garden plants, comprising: The data analysis unit is used to determine the distribution status of hotspots based on the historical hotspot concentration and the number of historical hotspots, and to determine the regional division method based on the hotspot distribution status, whether it is hotspot-based division, uniform division, or path-based division. The first partitioning unit, which is connected to the data analysis unit, is used to perform hotspot baseline partitioning to determine the first partitioning region and the second partitioning region based on the number of hotspots in the region to be analyzed corresponding to the hotspots. The second division unit, which is connected to the data analysis unit, is used to perform path benchmark division to determine multi-flow paths, and to determine the area range based on the associated influence distance of the multi-flow paths, the associated flow ratio, or the path flow reference value. The data storage unit is connected to the first partitioning unit and the second partitioning unit respectively. The data storage unit includes a first storage module and a second storage module, which are used to select the storage module corresponding to different pest-related data in response to different storage conditions.

[0006] Furthermore, for target scene areas where the historical hotspot concentration is greater than the preset historical hotspot concentration and the number of historical hotspots is greater than the preset number of historical hotspots, the data analysis unit determines the area division method as hotspot baseline division.

[0007] Furthermore, the first partitioning unit responds to the preset hotspot partitioning conditions to perform hotspot baseline partitioning, constructs several first partitioning regions, and records the part outside the first partitioning regions in the target scene region as the second partitioning region; For each hotspot, the surrounding area is determined to obtain the number of other hotspots in the area to be analyzed corresponding to each hotspot, and the effective area with a number of hotspots greater than the preset number is recorded as the first division area; The preset hotspot division condition is that the data analysis unit determines the region division method as a hotspot baseline division.

[0008] Furthermore, for target scene areas where the number of historical hotspots is less than or equal to the preset number of historical hotspots, the data analysis unit determines the area division method as path-based division.

[0009] Furthermore, the second partitioning unit responds to the first preset path partitioning condition and determines to construct a region based on multiple traffic paths; The first preset path division condition is that the data analysis unit determines the region division method as path benchmark division and the flow balance degree is less than the preset flow balance degree.

[0010] Furthermore, in response to the first execution condition, the second division unit performs correlation impact analysis on each multi-traffic path in a preset order. In the correlation impact analysis of a single multi-traffic path, the correlation impact distance corresponding to that multi-traffic path is detected. If the distance of the associated influence is less than the preset distance of the associated influence, the area of ​​the region is determined according to the ratio of the associated flow rates. If the distance of the associated influence is greater than or equal to the preset distance of the associated influence, the area range is determined according to the path flow reference value corresponding to the multiple flow paths. The first execution condition is to determine whether to construct a region based on multiple traffic paths.

[0011] Furthermore, the second partitioning unit responds to the second preset path partitioning condition and determines the region construction based on the complex reference values ​​of the elements; The second preset path division condition is that the data analysis unit determines the area division method as path benchmark division and the flow balance degree is greater than or equal to the preset flow balance degree.

[0012] Furthermore, in response to the second execution condition, the second partitioning unit performs element analysis on each multi-traffic path in a preset order. When performing element analysis on a single multi-traffic path, it detects the complex reference value of the element corresponding to that multi-traffic path. If the element complexity reference value is greater than the preset element complexity reference value, then the area corresponding to the multi-traffic path is determined based on the element complexity reference value. If the element complexity reference value is less than or equal to the preset element complexity reference value, then the area of ​​the region range is the base area of ​​the region range. The second execution condition is to determine whether to construct a region based on the complex reference values ​​of the elements.

[0013] Furthermore, for target scene areas where the hotspot distribution status is that the historical hotspot aggregation degree is less than or equal to the preset historical hotspot aggregation degree and the number of historical hotspots is greater than the preset number of historical hotspots, the data analysis unit determines the region division method as uniform division.

[0014] Furthermore, in response to the first storage condition, the data storage unit stores the pest-related data of the first divided area into the first storage module and the pest-related data of the second divided area into the second storage module. In response to the second storage condition, the data storage unit stores the pest-related data corresponding to the multi-traffic paths into the first storage module and the pest-related data corresponding to the ordinary traffic paths into the second storage module. The first storage condition is that the first partitioned region and the second partitioned region have been obtained; The second storage condition is that multiple traffic paths and their corresponding area ranges, as well as ordinary traffic paths and their corresponding area ranges, have been obtained.

[0015] Compared with the prior art, the beneficial effects of the present invention are that, in the technical solution of the present invention, the data analysis unit determines the region division method as hotspot benchmark division, uniform division or path benchmark division based on the number of historical hotspots and the degree of historical hotspot aggregation. The number of historical hotspots and the degree of historical hotspot aggregation reflect the actual distribution of pests and diseases in the target scene area in history and correspond to different region division methods, so that the selection of region division method is more in line with the actual working scenario, thereby improving the accuracy of region division and further improving the effectiveness of the finally obtained pest-related data.

[0016] Furthermore, based on the path-based division method, the region construction is determined by comparing the flow balance degree with the preset flow balance degree. The region construction is based on multiple flow paths or complex element reference values. The flow balance degree reflects the balance of personnel flow among multiple flow paths in the target scene area, and different region construction methods are determined accordingly. The impact of personnel flow on garden pests is taken into account, so that when the severity of historical pests is low, the region division and construction can be effectively carried out, thereby improving the subsequent pest data collection and analysis effect.

[0017] Furthermore, under the condition of constructing regions based on multiple traffic paths, the correlation impact analysis is performed on each multiple traffic path in a preset order. Based on the comparison results of the correlation impact distance and the preset correlation impact distance, different methods for confirming the area of ​​the region are determined. By performing the correlation impact analysis in a preset order, the orderly nature of the correlation impact analysis is effectively guaranteed, avoiding data redundancy caused by repeated analysis. The correlation impact distance reflects the degree of mutual influence between multiple traffic paths, making the determination of the area of ​​the region more reasonable and avoiding the problem of poor effect of region construction based on multiple traffic paths caused by the fixed setting of the area of ​​the region. Attached Figure Description

[0018] Figure 1 This is a unit connection diagram of the monitoring and optimization system for garden plant diseases and pests of the present invention; Figure 2 This is a flowchart illustrating how the region division method is determined based on the hotspot distribution status in this invention. Figure 3 This is a flowchart illustrating the correlation impact analysis of a single multi-traffic path according to the present invention. Detailed Implementation

[0019] To make the objectives and advantages of this invention clearer, the invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0021] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0022] Please see Figures 1 to 3 As shown, the present invention provides a monitoring and optimization system for diseases and pests of garden plants, comprising: The data analysis unit is used to determine the distribution status of hotspots based on the historical hotspot concentration and the number of historical hotspots, and to determine the regional division method based on the hotspot distribution status, whether it is hotspot-based division, uniform division, or path-based division. The first partitioning unit, which is connected to the data analysis unit, is used to perform hotspot baseline partitioning to determine the first partitioning region and the second partitioning region based on the number of hotspots in the region to be analyzed corresponding to the hotspots. The second division unit, which is connected to the data analysis unit, is used to perform path benchmark division to determine multi-flow paths, and to determine the area range based on the associated influence distance of the multi-flow paths, the associated flow ratio, or the path flow reference value. The data storage unit is connected to the first partitioning unit and the second partitioning unit respectively. The data storage unit includes a first storage module and a second storage module, which are used to select the storage module corresponding to different pest-related data in response to different storage conditions.

[0023] This invention is applied to the monitoring of pests and diseases in garden plants, particularly the division and processing of monitoring areas for pest and disease data collection. The target scene area is the garden scenic area that needs to be monitored for pests and diseases. In this invention, the user uses several image acquisition devices to collect and monitor garden monitoring images. The garden monitoring images are images of the monitoring area captured by the image acquisition devices. For a single image acquisition device, the number of garden monitoring images collected in each monitoring cycle is the same. In specific implementation, the number of garden monitoring images collected in a single monitoring cycle is set by the user. The greater the user's requirement for pest monitoring accuracy, the greater the number of garden monitoring images collected in a single monitoring cycle. The monitoring area is a portion of the target scene area captured by the image acquisition device. The setting position of each image acquisition device is set by the user.

[0024] This invention also utilizes several historical records. Each historical record includes at least the pest characteristic value, number of hotspots, radius length, path flow reference value, flow balance, historical hotspot aggregation degree, number of historical hotspots, reference distance, associated influence distance, associated flow ratio, element complexity reference value, weight, and area of ​​the region during a single historical process. Each historical record is also recorded with a corresponding qualification mark, which indicates whether the historical record meets the user's needs. The use of self-defined indicators (e.g., the area of ​​areas with missed pests and diseases) to determine whether the historical record meets the user's needs is a concept already known to those skilled in the art and will not be elaborated upon here.

[0025] This invention employs a continuous cyclical monitoring cycle. At the end of each monitoring cycle, the data analysis unit determines the regional division method. The duration of a single monitoring cycle is set by the user. The greater the user's demand for monitoring frequency of plant diseases and pests, the shorter the duration of a single monitoring cycle. One possible value is provided: the duration of a single monitoring cycle is 1 month.

[0026] Specifically, for target scene areas where the historical hotspot concentration is greater than the preset historical hotspot concentration and the number of historical hotspots is greater than the preset number of historical hotspots, the data analysis unit determines the area division method as hotspot baseline division.

[0027] If the historical hotspot concentration of the target scene area is greater than the preset historical hotspot concentration and the number of historical hotspots is greater than the preset number of historical hotspots, then the hotspot distribution state of the target scene area is the first preset hotspot distribution state.

[0028] The hotspot identification method involves identifying the garden monitoring images corresponding to each image acquisition device to determine whether there are pest-infested areas within the images. Images containing pest-infested areas are designated as pest images. For a single image acquisition device, the ratio of pest images in the most recent monitoring cycle to the total number of acquired garden monitoring images is recorded as the pest characterization value for that device: pest characterization value = number of pest images / total number of garden monitoring images. If the pest characterization value is greater than a preset value, the image acquisition device is considered a target device, and the center point of the monitoring area corresponding to the target device is designated as a hotspot. In this invention, the acquired garden monitoring images undergo preprocessing operations such as image enhancement, denoising, and size normalization. A target detection model is used to extract features layer by layer from the images via a convolutional neural network. The extracted features are then used to determine whether pests exist within the image and to accurately locate their positions, thus confirming whether the image is a pest image. The target detection model is the YOLO model, but it is not limited to this model.

[0029] A training method for the target detection model used in this invention is provided, comprising: First, plant images are collected for model training. These images are taken under different lighting conditions, angles, backgrounds, and growth stages to ensure the robustness of the model. Each plant image is labeled (e.g., healthy, powdery_mildew) and the outline of each lesion or pest is marked using a tool. Then, the plant images are preprocessed, including but not limited to image enhancement, geometric transformation, denoising, and color transformation. Secondly, define an objective, such as "cross-entropy loss," to measure the gap between the model's current predictions and the true labels. Select an Adam algorithm to guide the model in adjusting its internal parameters (such as weights and biases) to minimize the loss, and perform iterative training. Finally, the model was tested using a test set, and the model whose final accuracy met the user's requirements was selected as the object detection model.

[0030] The preset pest characterization value can be set by the user according to the actual application needs. It can be understood that the higher the user’s requirement for the recognition accuracy of pest areas in the target scene area, the smaller the preset pest characterization value will be. A value selection method is provided to extract the pest characterization value corresponding to the historical records that meet the user’s needs, remove outliers from the pest characterization value, and record the average value of the pest characterization value after removing outliers as the preset pest characterization value.

[0031] The number of historical hotspots refers to the number of hotspots within the most recent monitoring period.

[0032] The historical hotspot clustering degree is determined by randomly selecting the area to be analyzed for each hotspot. When analyzing the area of ​​a single hotspot, a circular region with the hotspot's location as the center and a radius of a preset length is established, and this circular region is designated as the region to be analyzed. If the number of hotspots within the region to be analyzed is greater than a preset number, then the region to be analyzed is designated as a valid region. The number of hotspots within the region to be analyzed does not include the hotspot at the center. If the number of hotspots within the region to be analyzed is less than or equal to the preset number, then the region to be analyzed is designated as an invalid region. Historical hotspot clustering degree = number of valid regions / number of regions to be analyzed. It is worth noting that if a hotspot cannot establish a circular region that is completely within the target scene area, then there is no need to perform area analysis for that hotspot.

[0033] Users can adaptively set the preset quantity and preset length values ​​according to actual application needs. It is understood that the more effective areas there are, the greater the concentration of hotspots. Therefore, the higher the user's requirement for the accuracy of subsequent hotspot baseline division, the larger the preset quantity and preset length values ​​will be. One method is provided to extract the number of hotspots and radius length corresponding to the historical records that meet the user's needs, remove outliers from the number of hotspots and radius length, and record the average value of the number of hotspots and radius length after removing outliers as the preset quantity and preset length values, respectively. The methods for removing outliers include, but are not limited to, the 3σ criterion method or the IQR method.

[0034] Specifically, the first partitioning unit responds to the preset hotspot partitioning conditions to perform hotspot baseline partitioning, constructs several first partitioning regions, and records the part outside the first partitioning region in the target scene region as the second partitioning region; For each hotspot, the surrounding area is determined to obtain the number of other hotspots in the area to be analyzed corresponding to each hotspot, and the effective area with a number of hotspots greater than the preset number is recorded as the first division area; The preset hotspot division condition is that the data analysis unit determines the region division method as a hotspot baseline division.

[0035] The effective area is the first partitioned area.

[0036] Specifically, for target scene areas where the number of historical hotspots is less than or equal to the preset number of historical hotspots, the data analysis unit determines the area division method as path-based division.

[0037] If the number of historical hotspots in the target scene area is less than or equal to the preset number of historical hotspots, then the hotspot distribution state of the target scene area is recorded as the second preset hotspot distribution state.

[0038] Specifically, the second partitioning unit responds to the first preset path partitioning condition and determines to construct a region based on multiple traffic paths; The first preset path division condition is that the data analysis unit determines the region division method as path benchmark division and the flow balance degree is less than the preset flow balance degree.

[0039] In this invention, several regional paths are set within the target scene area. These regional paths are the paths that pedestrians can move within the garden scenic area. The regional paths are set by the user. It is worth noting that different regional paths are allowed to partially intersect or overlap. For a single regional path, if its path flow reference value is greater than the preset path flow reference value, then the regional path is a multi-flow path. If its path flow reference value is less than or equal to the preset path flow reference value, then the regional path is a normal flow path. The range corresponding to the normal flow path is the area of ​​the baseline region.

[0040] The path flow reference value is determined by obtaining the reference pedestrian flow corresponding to the image acquisition device within the reference range of the area path, and recording the average value of the reference pedestrian flow as the path flow reference value. For a single area path, its corresponding reference range is the area corresponding to the smallest rectangle that can completely select the area path.

[0041] The method for determining the reference pedestrian flow corresponding to the image acquisition device within the reference range of the area path is as follows: extract several garden monitoring images acquired by each image acquisition device within the reference range of the area path in the most recent monitoring period; for a single image acquisition device, obtain the number of people in each garden monitoring image through a multi-target tracking algorithm (such as DeepSORT); and record the average number of people as the pedestrian flow corresponding to that image acquisition device. The average pedestrian flow of each image acquisition device is recorded as the reference pedestrian flow corresponding to the image acquisition device within the reference range of the area path.

[0042] Flow balance ;in, For the first Reference values ​​for path traffic in the region. This represents the average value of the reference values ​​for path traffic within the region. This represents the number of regional paths within the target scene area.

[0043] Users can adaptively set the preset path flow reference value and preset flow balance value according to actual application needs. It is understandable that in target scenario areas where the number of historical hotspots is less than or equal to the preset number of historical hotspots, considering the small number of historical hotspots, it is not possible to divide the area based on the number of historical hotspots. Considering the impact of personnel flow in gardens and scenic areas on pests and diseases, the area is constructed based on the personnel flow within the area path range. Therefore, the greater the accuracy of the area construction based on multiple flow paths, the smaller the preset path flow reference value and the larger the preset flow balance value. A value selection method is provided to extract the path flow reference value and flow balance value corresponding to the historical records that meet the user's needs, remove outliers from the path flow reference value and flow balance value respectively, and record the average value of the path flow reference value and flow balance value after removing outliers as the preset path flow reference value and preset flow balance value respectively.

[0044] Specifically, the second division unit responds to the first execution condition and performs correlation impact analysis on each multi-traffic path in a preset order. In the correlation impact analysis of a single multi-traffic path, the correlation impact distance corresponding to that multi-traffic path is detected. If the distance of the associated influence is less than the preset distance of the associated influence, the area of ​​the region is determined according to the ratio of the associated flow rates. If the distance of the associated influence is greater than or equal to the preset distance of the associated influence, the area range is determined according to the path flow reference value corresponding to the multiple flow paths. The first execution condition is to determine whether to construct a region based on multiple traffic paths.

[0045] In this invention, the preset order is the path flow reference value from largest to smallest. For a single multi-flow path, the method for confirming its corresponding associated influence distance is as follows: the single multi-flow path is recorded as the target multi-flow path. By uniformly selecting several nodes on the target multi-flow path, and ensuring that the path length between adjacent nodes is equal, the path distance between the target multi-flow path and other multi-flow paths is detected. The method for confirming the path distance is as follows: for any other multi-flow path, the minimum distance between each node on the target multi-flow path and the multi-flow path is detected, and the average of the minimum distances is recorded as the path distance between the target multi-flow path and the multi-flow path. The minimum path distance is recorded as the associated influence distance, and the other multi-flow path with the smallest path distance is recorded as the associated path of the target multi-flow path.

[0046] The preset correlation influence distance can be set by the user according to the actual application scenario. It can be understood that the correlation influence distance reflects the degree of influence that the target multi-traffic path may be affected by other multi-traffic paths. It can also be understood that the greater the correlation influence between multi-traffic paths, the smaller the correlation influence distance. Therefore, the greater the user's tolerance for the degree of correlation influence, the larger the preset correlation influence distance. A value setting method is provided, which extracts the correlation influence distance corresponding to the historical records that meet the user's needs, removes outliers from the correlation influence distance, and records the average correlation influence distance after removing outliers as the preset correlation influence distance.

[0047] The associated traffic ratio = the reference value of the path flow of the target multi-traffic path / the reference value of the path flow of the associated path; the area of ​​the region = the area of ​​the baseline region × (associated traffic ratio / preset associated traffic ratio). The preset associated traffic ratio can be set by the user according to the actual application needs. It can be understood that the greater the user's tolerance for the impact of the associated traffic ratio on the area of ​​the region, the larger the preset associated traffic ratio will be. One value is provided, the associated traffic ratio corresponding to the historical records that meet the user's needs is extracted, the outliers in the associated traffic ratio are removed, and the average value of the associated traffic ratio after removing the outliers is recorded as the preset associated traffic ratio value.

[0048] When determining the area range based on the path flow reference value corresponding to multiple flow paths, the area range = base area range × (path flow reference value / preset path flow reference value).

[0049] The value of the baseline area can be adaptively set by the user according to the actual application needs. It can be understood that the higher the user's requirement for the accuracy of the area division, the larger the value of the baseline area. One value is provided, the area corresponding to the historical records that meet the user's needs is extracted, outliers in the area are removed, and the average value of the area after removing outliers is recorded as the value of the baseline area.

[0050] Specifically, the second partitioning unit responds to the second preset path partitioning condition and determines to construct a region based on the complex reference value of the element; The second preset path division condition is that the data analysis unit determines the area division method as path benchmark division and the flow balance degree is greater than or equal to the preset flow balance degree.

[0051] Specifically, the second partitioning unit responds to the second execution condition and performs element analysis on each multi-traffic path in a preset order. When performing element analysis on a single multi-traffic path, it detects the complex reference value of the element corresponding to that multi-traffic path. If the element complexity reference value is greater than the preset element complexity reference value, then the area corresponding to the multi-traffic path is determined based on the element complexity reference value. If the element complexity reference value is less than or equal to the preset element complexity reference value, then the area of ​​the region range is the base area of ​​the region range. The second execution condition is to determine whether to construct a region based on the complex reference values ​​of the elements.

[0052] The method for confirming the element complexity reference value is as follows: Element complexity reference value = α1 × influencing element volume + α2 × influencing element aggregation degree; where α1 is the first weight coefficient, α2 is the second weight coefficient, and α1 + α2 = 1. The values ​​of α1 and α2 can be set directly by the user based on domain experience, or by using statistical methods, such as regression analysis or principal component analysis, to determine the contribution of influencing element volume and influencing element aggregation degree to the element complexity reference value, thereby determining the corresponding weight coefficient values. The greater the contribution, the greater the weight coefficient value. The weights can also be adjusted through historical training (such as machine learning). This invention provides one set of values, where α1 = 0.5 and α2 = 0.5.

[0053] Among them, modern algorithms based on deep learning identify the volume of influencing elements and the clustering degree of influencing elements within the reference range of multi-traffic paths in the real-world 3D model. Influencing elements include, but are not limited to, wooden benches, i.e., components within the target scene area that are prone to causing or aggravating pests and diseases. The clustering degree of influencing elements is determined by uniformly dividing a single multi-traffic path into n sub-segments, with each sub-segment corresponding to an equal path length, and counting the number of influencing elements within a single sub-segment. Affects the aggregation degree of elements ;in, .

[0054] The area of ​​the region = the area of ​​the baseline region × (the element complexity reference value / the preset element complexity reference value). The preset element complexity reference value can be set by the user according to the actual application needs. It can be understood that the greater the user's tolerance for the impact of the element complexity reference value on the area of ​​the region, the larger the preset element complexity reference value will be. A value selection method is provided, which extracts the element complexity reference value corresponding to the historical records that meet the user's needs, removes outliers from the element complexity reference value, and records the average value of the element complexity reference value after removing outliers as the preset element complexity reference value.

[0055] Specifically, for target scene areas where the hotspot distribution status is that the historical hotspot aggregation degree is less than or equal to the preset historical hotspot aggregation degree and the number of historical hotspots is greater than the preset number of historical hotspots, the data analysis unit determines the area division method as uniform division.

[0056] If the historical hotspot concentration of the target scene area is less than or equal to the preset historical hotspot concentration and the number of historical hotspots is greater than the preset number of historical hotspots, then the hotspot distribution state of the target scene area is recorded as the third preset hotspot distribution state.

[0057] For the target scene area in the third preset hotspot distribution state, the first division unit divides the target scene area into several square areas of equal area. Among them, other areas of the target scene area that cannot be divided into squares due to their own shape are merged into the nearest square area.

[0058] Users can adaptively set the preset number of historical hotspots and the preset historical hotspot concentration based on their actual application needs. It's understood that a larger number of historical hotspots indicates a greater degree of pest infestation, and a higher historical hotspot concentration indicates a greater concentration of infested areas. Therefore, for target scenario areas where the historical hotspot concentration is greater than the preset historical hotspot concentration and the number of historical hotspots is greater than the preset number of historical hotspots, the area division method is based on hotspot baseline division. Thus, the greater the precision of the hotspot baseline division method set by the user, the larger the values ​​of the preset historical hotspot concentration and the preset number of historical hotspots. A value setting method is provided, extracting the historical hotspot quantity and historical hotspot concentration corresponding to the historical records that meet the user's needs. Outliers in both the historical hotspot quantity and historical hotspot concentration are removed, and the average values ​​of the historical hotspot quantity and historical hotspot concentration after removing outliers are recorded as the preset values ​​for the preset number of historical hotspots and the preset historical hotspot concentration.

[0059] Specifically, in response to the first storage condition, the data storage unit stores the pest-related data of the first divided area into the first storage module and the pest-related data of the second divided area into the second storage module. In response to the second storage condition, the data storage unit stores the pest-related data corresponding to the multi-traffic paths into the first storage module and the pest-related data corresponding to the ordinary traffic paths into the second storage module. The first storage condition is that the first partitioned region and the second partitioned region have been obtained; The second storage condition is that multiple traffic paths and their corresponding area ranges, as well as ordinary traffic paths and their corresponding area ranges, have been obtained.

[0060] This invention improves the clustering performance of pest data by dividing the data into regions. Furthermore, it modularizes the pest-related data corresponding to different regions and multiple traffic paths for storage, allowing users to understand the pest-related data more intuitively. The pest-related data in the first storage module is marked as key data, while the pest-related data in the second storage module is marked as non-key data.

[0061] For a single zone (first zone or second zone), the pest-related data are the garden monitoring images collected by each image acquisition device in that zone during the most recent monitoring cycle. For a single path (multi-traffic path or ordinary traffic path), the pest-related data are the garden monitoring images collected by each image acquisition device in the most recent monitoring cycle within the defined area.

[0062] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A monitoring and optimization system for diseases and pests of garden plants, characterized in that, include: The data analysis unit is used to determine the distribution status of hotspots based on the historical hotspot concentration and the number of historical hotspots, and to determine the regional division method based on the hotspot distribution status: hotspot-based division, uniform division, or path-based division. The first partitioning unit, which is connected to the data analysis unit, is used to perform hotspot baseline partitioning to determine the first partitioning region and the second partitioning region based on the number of hotspots in the region to be analyzed corresponding to the hotspots. The second division unit, which is connected to the data analysis unit, is used to perform path benchmark division to determine multi-flow paths, and to determine the area range based on the associated influence distance of the multi-flow paths, the associated flow ratio, or the path flow reference value. The data storage unit is connected to the first partitioning unit and the second partitioning unit respectively. The data storage unit includes a first storage module and a second storage module, which are used to select the storage module corresponding to different pest-related data in response to different storage conditions.

2. The monitoring and optimization system for garden plant diseases and pests according to claim 1, characterized in that, For target scene areas where the historical hotspot concentration is greater than the preset historical hotspot concentration and the number of historical hotspots is greater than the preset number of historical hotspots, the data analysis unit determines the area division method as hotspot baseline division.

3. The monitoring and optimization system for garden plant diseases and pests according to claim 2, characterized in that, The first partitioning unit responds to the preset hotspot partitioning conditions to perform hotspot baseline partitioning, constructs several first partitioning regions, and records the part outside the first partitioning region in the target scene region as the second partitioning region; For each hotspot, the surrounding area is determined to obtain the number of other hotspots in the area to be analyzed corresponding to each hotspot, and the effective area with a number of hotspots greater than the preset number is recorded as the first division area; The preset hotspot division condition is that the data analysis unit determines the region division method as a hotspot baseline division.

4. The monitoring and optimization system for garden plant diseases and pests according to claim 1, characterized in that, For target scene areas where the number of historical hotspots is less than or equal to the preset number of historical hotspots, the data analysis unit determines the area division method as path-based division.

5. The monitoring and optimization system for garden plant diseases and pests according to claim 4, characterized in that, The second partitioning unit responds to the first preset path partitioning condition and determines to construct a region based on multiple traffic paths; The first preset path division condition is that the data analysis unit determines the region division method as path benchmark division and the flow balance degree is less than the preset flow balance degree.

6. The monitoring and optimization system for garden plant diseases and pests according to claim 5, characterized in that, The second division unit responds to the first execution condition and performs correlation impact analysis on each multi-traffic path in a preset order. In the correlation impact analysis of a single multi-traffic path, the correlation impact distance corresponding to that multi-traffic path is detected. If the distance of the associated influence is less than the preset distance of the associated influence, the area of ​​the region is determined according to the ratio of the associated flow rates. If the distance of the associated influence is greater than or equal to the preset distance of the associated influence, the area range is determined according to the path flow reference value corresponding to the multiple flow paths. The first execution condition is to determine whether to construct a region based on multiple traffic paths.

7. The monitoring and optimization system for garden plant diseases and pests according to claim 4, characterized in that, The second partitioning unit responds to the second preset path partitioning condition and determines the region construction based on the complex reference value of the element. The second preset path division condition is that the data analysis unit determines the area division method as path benchmark division and the flow balance degree is greater than or equal to the preset flow balance degree.

8. The monitoring and optimization system for garden plant diseases and pests according to claim 7, characterized in that, The second partitioning unit responds to the second execution condition and performs element analysis on each multi-traffic path in a preset order. When performing element analysis on a single multi-traffic path, it detects the complex reference value of the element corresponding to that multi-traffic path. If the element complexity reference value is greater than the preset element complexity reference value, then the area corresponding to the multi-traffic path is determined based on the element complexity reference value. If the element complexity reference value is less than or equal to the preset element complexity reference value, then the area of ​​the region range is the base area of ​​the region range. The second execution condition is to determine whether to construct a region based on the complex reference values ​​of the elements.

9. The monitoring and optimization system for garden plant diseases and pests according to claim 1, characterized in that, For target scene areas where the historical hotspot concentration is less than or equal to the preset historical hotspot concentration and the number of historical hotspots is greater than the preset number of historical hotspots, the data analysis unit determines the area division method as uniform division.

10. The monitoring and optimization system for garden plant diseases and pests according to claim 3, 6, or 8, characterized in that, In response to the first storage condition, the data storage unit stores the pest-related data of the first divided area into the first storage module and the pest-related data of the second divided area into the second storage module. In response to the second storage condition, the data storage unit stores the pest-related data corresponding to the multi-traffic paths into the first storage module and the pest-related data corresponding to the ordinary traffic paths into the second storage module. The first storage condition is that the first partitioned region and the second partitioned region have been obtained; The second storage condition is that multiple traffic paths and their corresponding area ranges, as well as ordinary traffic paths and their corresponding area ranges, have been obtained.

Images