Peanut growth data-based disease treatment data acquisition system

Abstract

The invention discloses a disease treatment data acquisition system based on peanut growth data. Comprising a plant multi-block division monitoring module, a block multi-dimensional growth data acquisition module, a plant pesticide application track data mapping module, a multi-block plant disease state intervention analysis module, a plant disease regulation index quantitative qualification judgment module and a plant data comprehensive management platform, and aims to pre-define each growth cycle of peanuts; meanwhile, the field is divided into a plurality of independent management blocks, growth data of peanuts planted in different time dimensions in different blocks are collected, multi-frequency collection is conducted on different blocks in combination with the disease occurrence rule of the same period in the history, multi-stage accurate early warning is conducted based on multi-block disease data, pesticide application data of different blocks are dynamically adjusted, and the management efficiency of the peanuts is improved. Serialized tracking monitoring is automatically started after pesticide is applied to different blocks, and quantitative evaluation is carried out on pesticide application work of different blocks in different peanut growth periods by comparing disease states of multiple time points before pesticide application and after pesticide application of the same block.

Description

Technical Field

[0001] This invention relates to the field of data acquisition, specifically a data acquisition system for disease control based on peanut growth data. Background Technology

[0002] Peanuts, as an important oilseed and cash crop, are crucial for ensuring food and oil security through stable production. However, peanuts are susceptible to various foliar and soil-borne diseases during their growth cycle. Agricultural technicians or farmers rely on periodic field inspections and personal experience to determine disease occurrence and the timing of pesticide application. This method is highly subjective, inefficient, has limited coverage, and struggles to make accurate judgments in the early stages of disease outbreaks when symptoms are not obvious, often missing the optimal window for control. Furthermore, current pesticide application methods, whether manual or mechanical based on simple prescription maps, mostly treat the entire field or large area as a homogeneous unit, spraying a uniform dosage. This ignores the spatial heterogeneity of actual disease distribution in the field, resulting in pesticide waste and environmental pollution in disease-free or mildly affected areas, while severely affected areas may receive insufficient pesticides, significantly reducing control effectiveness. This violates the core principle of precision agriculture, "on-demand application," and fails to quantitatively correlate the precise geographical scope of a single application with changes in disease status before and after application in the corresponding area, leaving the assessment of control effectiveness only at a qualitative level.

[0003] This application aims to predefine the various growth cycles of peanuts, divide the field into multiple independent management blocks, collect growth data of peanuts planted in different blocks at different time dimensions, combine the disease occurrence patterns of the same period in history to collect data on different blocks at multiple frequencies, conduct multi-level precise early warning based on disease data of multiple blocks, dynamically adjust the pesticide application data of different blocks, automatically start serialized tracking and monitoring after pesticide application in different blocks, and quantitatively evaluate the pesticide application operations of different blocks under different peanut growth cycles by comparing the disease status of the same block before and after pesticide application at multiple time points, thereby achieving precise and intelligent peanut disease prevention and control. Summary of the Invention

[0004] The purpose of this invention is to provide a data acquisition system for disease control based on peanut growth data, so as to solve the problems in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A data acquisition system for disease control based on peanut growth data, comprising a multi-block plant segmentation monitoring module, a multi-dimensional growth data acquisition module for each block, a plant application trajectory data mapping module, a multi-block plant disease status intervention and analysis module, a plant disease control index quantitative qualification judgment module, and a plant data comprehensive management platform. The multi-block division monitoring module collects spatial data, environmental data, area data, and basic soil data of peanut plant planting, and summarizes them into data information for the target peanut field. It dynamically divides the target peanut field into multiple blocks and establishes a multi-block peanut planting file. The multidimensional growth data acquisition module of the partitioned blocks collects growth data of peanut planting in different blocks at different time dimensions according to different partitions, performs preprocessing, identifies disease types and lesion areas in different blocks under different growth cycles, constructs a disease data early warning rule library, and provides intelligent early warning for plant disease data in different blocks; The plant application trajectory data mapping module summarizes the information on the areas requiring application, disease data, and early warning markers under different growth cycles, constructs multi-block application files under different cycles, and generates application prescriptions and application trajectories for multi-block applications under different cycles for effective application data monitoring. The multi-block plant disease status intervention and analysis module evaluates different blocks at multiple time periods after each application, sets up control image data before application, tracks and compares the disease status of plants in multiple blocks before and after application, and analyzes the treatment effect of application in each block. The quantitative qualification assessment module for plant disease control indicators summarizes and analyzes the number of effective and ineffective pesticide application markers in different blocks of the target peanut field under different growth stages, and makes a qualification assessment for the pesticide control data of each block.

[0006] Further configuration: The plant multi-block division monitoring module includes a basic spatial information acquisition and survey sub-module and a multi-factor calibration dynamic block division and archiving sub-module. The basic spatial information acquisition and survey sub-module includes a GPS positioning unit, a UAV data connection and acquisition unit, and a historical data import unit. The GPS positioning unit obtains the geographic boundary coordinates of the target peanut field. The UAV data connection and acquisition unit connects to an external UAV to collect images within the target peanut field. The historical data import unit imports historical planting data of the target peanut field, including historical environmental data, historical plant disease and pest data, historical yield maps, and historical soil sampling and monitoring data. This data is then summarized into target peanut field data information and sent to the multi-factor calibration dynamic block division and archiving sub-module.

[0007] Further settings: The multi-factor calibration dynamic block division and archiving submodule pre-calculates the area of ​​the target peanut field based on its geographical boundary coordinates, pre-divides it according to the same management area, and summarizes it into initial blocks. Based on images of peanut plants within each initial block collected by drones, it determines the row spacing and plant spacing of peanut plants within each initial block, and estimates the number of peanut plants within each initial block. If the number of peanut plants within an initial block exceeds a set threshold, it is further divided into equal areas to obtain different blocks of the target peanut field. Simultaneously, it collects historical planting data... If all areas corresponding to a given block are high-incidence areas for plant diseases and pests, they are marked. If only some areas corresponding to a given block are high-incidence areas for plant diseases and pests, the areas within the current block that are high-incidence areas for plant diseases and pests are divided into separate blocks, and each block is assigned a unique identifier. All block information within the current target peanut field is summarized, including block identifier, block location, block area, estimated number of plants in the block, and high-incidence marking for plant diseases and pests in the block. Based on the historical data import unit, environmental topographic data and soil data are matched for each block information.

[0008] Further setup: The multi-dimensional growth data acquisition module for different sections includes several multispectral cameras, environmental sensors, a multi-temporal dimension differentiated data acquisition sub-module, and a plant growth and disease data fusion and evaluation sub-module. The multi-temporal dimension differentiated data acquisition sub-module pre-constructs a data acquisition framework for each cycle according to the peanut growth cycle within the target peanut field. Administrators pre-set the basic data acquisition frequency for the multi-temporal dimension differentiated data acquisition sub-module through the plant data integrated management platform. Combined with the data connection acquisition unit from several multispectral cameras, environmental sensors, and drones, it acquires plant canopy image data and key environmental data for each section within the target peanut field according to different cycle periods. Simultaneously, it imports historical plant disease and pest data from different cycles through the historical data import unit to determine whether the current cycle period falls within a historically high-incidence period for plant diseases and pests. If so, the multi-temporal dimension differentiated data acquisition... The collection module automatically increases the data collection frequency of all divided blocks within the time period to a high-frequency mode until the marked historical high-incidence period of plant diseases and pests ends. If not, the administrator uses the plant data comprehensive management platform to pre-establish a disease high-incidence environmental condition threshold database, obtains the real-time environmental monitoring data for the current period, and compares the real-time environmental monitoring data with any environmental data in the disease high-incidence environmental condition threshold database. If any environmental data is greater than the corresponding disease high-incidence environmental data threshold, the environmental data for the current period is determined to be a disease high-incidence environment. The multi-time dimension differentiated data collection submodule automatically increases the data collection frequency of all divided blocks within the time period to a high-frequency mode until the real-time monitored environmental data are all below the disease high-incidence environmental condition threshold, and then returns to the basic collection frequency. The peanut growth data collected according to the data collection framework for each period is summarized and archived.

[0009] Further setup: The plant growth and disease data fusion and evaluation submodule acquires real-time plant canopy image data of peanuts at different growth stages within each cycle's data acquisition framework. It preprocesses the real-time acquired plant image data, filtering out unqualified images, and classifies and summarizes the acquired images according to different growth stages and block information. It extracts features from plant canopy image data of different blocks within the same growth stage, including peanut plant leaf color, texture, and morphology data. A pre-trained disease spot learning model is constructed, and the first acquired plant canopy image data for each block within each growth stage is pre-extracted. As a reference image for different peanut plant growth cycles, the reference images of different blocks under a certain growth cycle are input into a pre-trained disease spot learning model to detect plant disease spots. The disease spot images of the reference images under each block are identified, the disease type and lesion area of ​​the reference images of different blocks under the current growth cycle are identified, the disease type of different blocks under the current growth cycle is identified and disease is marked, and the lesion area of ​​the reference images under the blocks with disease markings under the current growth cycle is identified. The administrator can pre-set the threshold of the total lesion area of ​​different blocks under different growth cycles through the plant data integrated management platform. Pre-determine the growth cycle period to which the disease-marked blocks belong. If the proportion of lesion area in the baseline image of the disease-marked blocks in a certain growth cycle is greater than the pre-set threshold for lesion area in different blocks in the current growth cycle, mark the block with a severe warning and send a severe warning to the plant data management platform. If the percentage of lesion area in the baseline image of a disease-marked block during a certain growth cycle is less than a pre-set threshold for lesion area in different blocks during the current growth cycle, continuously collected plant canopy image data for the disease-marked block during the current growth cycle is input into a pre-trained disease spot learning model. This model monitors the percentage of lesion area within different continuously collected plant canopy images of that block. The percentage of lesion area within different continuously collected plant canopy images of that block during the current growth cycle is set as follows: Set a pre-defined threshold for lesion area in different regions during the current growth cycle. If the percentage of lesion area inside the canopy images of different plants continuously collected in this block during the current growth cycle... ,and The area is marked as a severe warning and a severe warning is sent to the plant data management platform; if the proportion of lesion area inside different plant canopy images continuously collected in this area during the current growth cycle is... ,and The area is marked as having a moderate warning level, and a moderate warning is sent to the plant data management platform; if the maximum percentage of lesion area within the canopy images of different plants continuously collected during the current growth cycle is... ,in A mild warning is marked for the block based on a pre-set adjustment coefficient, and a mild warning is sent to the plant data management platform. The warning signals of different blocks under different growth cycles are then collected and uploaded to the plant data management platform.

[0010] Further settings: The plant application trajectory data mapping module includes a sub-module for optimizing application trajectory data by section and a sub-module for recording and monitoring effective application data. The sub-module for optimizing application trajectory data by section acquires information on each section of the target peanut field at different growth stages, including disease data and warning markers. It summarizes the location, disease type, and warning signals of the sections requiring application at different growth stages, constructs application files for multiple sections at different stages, and generates application prescriptions and application trajectories for multiple sections at different stages. The application prescription includes the type and concentration of pesticide solution for each section at different stages, and the application trajectory includes the operating speed and number of applications for each section at different stages. The administrator needs to predefine sections with different warning markers through the plant data integrated management platform, and plan different threshold operating speeds and different number of applications for sections with severe, moderate, and mild warnings. For sections without warnings, the application equipment is automatically shut down, and the application prescription and application trajectory are sent to the plant data integrated management platform for manual review. The effective pesticide application data recording and monitoring submodule acquires the number of pesticide applications, pesticide application data, and pesticide application trajectory for different blocks in different periods of the target peanut field. The pesticide application data includes the type of pesticide solution, concentration of pesticide solution, start and end interval of pesticide application, and meteorological data of pesticide application, which are summarized into effective pesticide application data for different blocks under different plant growth cycles.

[0011] Further configuration: The multi-block plant disease status intervention analysis module includes a multi-block drug treatment intervention data association submodule and a plant status tracking and comparison submodule before and after intervention. The multi-block drug treatment intervention data association submodule obtains effective drug treatment data for different blocks under different cycles, records the number of drug treatment operations in different blocks, extracts the end time of each drug treatment operation in different blocks under different cycles, and marks it as a drug treatment intervention. At any given time, an initial short-term assessment and a secondary assessment are conducted for different blocks after each application operation. The initial short-term assessment period is set as follows: The second evaluation period is set as ,in, , The system is pre-configured by the administrator to obtain plant canopy image data collected by the multidimensional growth data acquisition module for different application blocks during the initial short-term evaluation period and the secondary evaluation period after each application. The plant canopy image data collected during the initial short-term evaluation period and the secondary evaluation period after each application in each block are correlated with the current application data and summarized into application intervention evaluation image data for different blocks under different growth cycles.

[0012] Further settings: The submodule for tracking and comparing plant status before and after intervention application acquires image data of each intervention application in multiple blocks at different growth stages, and simultaneously extracts the intervention data for each application in different blocks at different growth stages. The last canopy image data collected by the multidimensional growth data acquisition module of the previous time block is used to label the pesticide application control image data. The pesticide application intervention evaluation image data and pesticide application control image data of each pesticide application operation in different blocks under different growth cycles are correlated, and the pesticide application control image data of each pesticide application operation in different blocks are extracted separately. Time-lapse image data Real-time image data is input into a pre-trained lesion spot learning model. The model identifies lesion area from images extracted during the current application of pesticides in a given area. It also includes control images extracted before and after a specific application in a given area. Time-lapse image data The lesion areas identified from the time-lapse image data are respectively , , The formula is used to determine the disease status of the plants in the current pesticide application area: in, For pesticide application operations The image recognition system for lesion area reduction sets a threshold. When the image extracted from the current application of pesticides in a certain area shows that the plant disease area status meets the above formula, the application is marked as valid. When the image extracted from the current application of pesticides in a certain area shows that the plant disease area status does not meet the above formula, the application is marked as invalid.

[0013] Further configuration: The plant disease control indicator quantification qualification judgment module includes a plant control effect indicator quantification analysis submodule and a block qualification assessment submodule. The plant control effect indicator quantification analysis submodule summarizes the number of effective and ineffective pesticide application operation markers in different blocks of the target peanut field under different growth stages, and analyzes the proportion of the number of effective pesticide application operation markers in different blocks under different growth stages to the total number of pesticide applications. If the proportion of the number of effective pesticide application operation markers in a certain block under different peanut plant growth stages to the total number of pesticide applications is greater than or equal to the set quantity threshold, the block information is sent to the block qualification assessment submodule, which marks the block as qualified. If the proportion of the number of effective pesticide application markers in a certain block under different peanut plant growth stages to the total number of pesticide applications is less than the set quantity threshold, the block qualification assessment submodule marks the block as unqualified. The block qualification assessment submodule summarizes the block information of qualified and unqualified blocks in the target peanut field and sends it to the plant data integrated management platform for manual review.

[0014] Compared with existing technologies, the beneficial effects of this invention are: it aims to predefine the various growth cycles of peanuts, divide the field into multiple independent management blocks, collect growth data of peanuts planted in different blocks at different time dimensions, combine the disease occurrence patterns of the same period in history to collect data on different blocks at multiple frequencies, conduct multi-level precise early warning based on disease data of multiple blocks, dynamically adjust the pesticide application data of different blocks, automatically start serialized tracking and monitoring after pesticide application in different blocks, and quantitatively evaluate the pesticide application operations of different blocks under different peanut growth cycles by comparing the disease status of the same block before and after pesticide application at multiple time points, thereby achieving precise and intelligent peanut disease prevention and control. Attached Figure Description

[0015] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0016] Figure 1 This is a schematic diagram illustrating the implementation process of a data acquisition system for disease control based on peanut growth data according to the present invention. Figure 2 This invention provides a module for a data acquisition system for disease control based on peanut growth data. Figure 1 ; Figure 3 This invention provides a module for a data acquisition system for disease control based on peanut growth data. Figure 2 ; Figure 4 This invention provides a module for a data acquisition system for disease control based on peanut growth data. Figure 3 ; Figure 5This invention provides a module for a data acquisition system for disease control based on peanut growth data. Figure 4 ; Figure 6 This invention provides a module for a data acquisition system for disease control based on peanut growth data. Figure 5 . Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figures 1-6 In this embodiment of the invention, a disease control data acquisition system based on peanut growth data is provided. The system includes a plant multi-block division monitoring module, a multi-dimensional growth data acquisition module for each block, a plant pesticide application trajectory data mapping module, a multi-block plant disease status intervention analysis module, a plant disease control index quantitative qualification judgment module, and a plant data comprehensive management platform. The multi-block division monitoring module collects spatial data, environmental data, area data, and basic soil data of peanut plant planting, and summarizes them into data information for the target peanut field. It dynamically divides the target peanut field into multiple blocks and establishes a multi-block peanut planting file. It needs to be explained in detail that the plant multi-block division monitoring module includes a basic spatial information acquisition and survey sub-module and a multi-factor calibration dynamic block division and archiving sub-module. The basic spatial information acquisition and survey sub-module includes a GPS positioning unit, a UAV data connection and acquisition unit, and a historical data import unit. The GPS positioning unit obtains the geographical boundary coordinates of the target peanut field. The UAV data connection and acquisition unit connects to an external UAV to collect images within the target peanut field. The historical data import unit imports the historical planting data of the target peanut field, including historical environmental data, historical plant disease and pest data, historical yield maps, and historical soil sampling and monitoring data. These are then summarized into data information for the target peanut field and sent to the multi-factor calibration dynamic block division and archiving sub-module.

[0019] Further explanation is needed. The multi-factor calibration dynamic block division and archiving submodule pre-calculates the area of ​​the target peanut field based on the geographical boundary coordinates of the target peanut field, pre-divides it according to the same management area, and summarizes it into initial blocks. Based on the peanut plant images collected by the drone within each initial block, it determines the row spacing and plant spacing of the peanut plants within each initial block, and estimates the number of peanut plants within each initial block. If the number of peanut plants in a certain initial block is greater than a set threshold, it is divided a second time according to the same area to obtain different blocks of the target peanut field. At the same time, historical planting data is collected. If all the locations corresponding to the divided blocks are high-incidence areas for plant diseases and pests, they are marked. If only some of the locations corresponding to the divided blocks are high-incidence areas for plant diseases and pests, the high-incidence areas within the current block are divided into separate blocks, and each block is assigned a unique identifier. All block information within the current target peanut field is summarized, including block identifier, block location, block area, estimated number of plants in the block, and high-incidence marking for diseases and pests in the block. Based on the historical data import unit, environmental topographic data and soil data are matched for each block information.

[0020] The multidimensional growth data acquisition module of the partitioned blocks collects growth data of peanut planting in different blocks at different time dimensions according to different partitions, performs preprocessing, identifies disease types and lesion areas in different blocks under different growth cycles, constructs a disease data early warning rule library, and provides intelligent early warning for plant disease data in different blocks; To further explain, the multi-dimensional growth data acquisition module for different sections includes several multispectral cameras, environmental sensors, a multi-temporal dimension differentiated data acquisition sub-module, and a plant growth and disease data fusion and evaluation sub-module. The multi-temporal dimension differentiated data acquisition sub-module pre-constructs a data acquisition framework for each cycle according to the peanut growth cycle within the target peanut field. Administrators pre-set the basic data acquisition frequency for the multi-temporal dimension differentiated data acquisition sub-module through the plant data integrated management platform. Combined with several multispectral cameras, environmental sensors, and drone data connection acquisition units, it collects plant canopy image data and key environmental data for each section within the target peanut field according to different cycle periods. Simultaneously, it imports historical plant disease and pest data from different cycles of the peanut plants through the historical data import unit to determine whether the current cycle period falls within a historically high-incidence period for plant diseases and pests. If so, the multi-temporal dimension differentiated data acquisition... The collection module automatically increases the data collection frequency of all divided blocks within the time period to a high-frequency mode until the marked historical high-incidence period of plant diseases and pests ends. If not, the administrator uses the plant data comprehensive management platform to pre-establish a disease high-incidence environmental condition threshold database, obtains the real-time environmental monitoring data for the current period, and compares the real-time environmental monitoring data with any environmental data in the disease high-incidence environmental condition threshold database. If any environmental data is greater than the corresponding disease high-incidence environmental data threshold, the environmental data for the current period is determined to be a disease high-incidence environment. The multi-time dimension differentiated data collection submodule automatically increases the data collection frequency of all divided blocks within the time period to a high-frequency mode until the real-time monitored environmental data are all below the disease high-incidence environmental condition threshold, and then returns to the basic collection frequency. The peanut growth data collected according to the data collection framework for each period is summarized and archived.

[0021] Further explanation is needed: the plant growth and disease data fusion and evaluation submodule acquires real-time plant canopy image data of peanuts at different growth stages within each cycle's data acquisition framework. It preprocesses the real-time acquired plant image data, filtering out unqualified images, and categorizes and summarizes the acquired images according to different growth stages and block information. It extracts features from different blocks within the same growth stage, including peanut plant leaf color, texture, and morphology data. A pre-trained disease spot learning model is constructed, and the first acquired plant canopy image data for each block within each growth stage is pre-extracted. Using these images as baseline images for different peanut plant growth cycles, the baseline images of different blocks under a certain growth cycle are input into a pre-trained disease spot learning model to detect plant disease spots. The disease spot images of the baseline images under each block are identified, and the disease type and lesion area of ​​the baseline images of different blocks under the current growth cycle are identified. The disease type of different blocks under the current growth cycle is identified and marked. The lesion area of ​​the baseline images under the blocks with disease markings under the current growth cycle is identified. The administrator can pre-set the threshold of the total lesion area of ​​different blocks under different growth cycles through the plant data integrated management platform. Pre-determine the growth cycle period to which the disease-marked blocks belong. If the proportion of lesion area in the baseline image of the disease-marked blocks in a certain growth cycle is greater than the pre-set threshold for lesion area in different blocks in the current growth cycle, mark the block with a severe warning and send a severe warning to the plant data management platform. If the percentage of lesion area in the baseline image of a disease-marked block during a certain growth cycle is less than a pre-set threshold for lesion area in different blocks during the current growth cycle, continuously collected plant canopy image data for the disease-marked block during the current growth cycle is input into a pre-trained disease spot learning model. This model monitors the percentage of lesion area within different continuously collected plant canopy images of that block. The percentage of lesion area within different continuously collected plant canopy images of that block during the current growth cycle is set as follows: Set a pre-defined threshold for lesion area in different regions during the current growth cycle. If the percentage of lesion area inside the canopy images of different plants continuously collected in this block during the current growth cycle... ,and The area is marked as a severe warning and a severe warning is sent to the plant data management platform; if the proportion of lesion area inside different plant canopy images continuously collected in this area during the current growth cycle is... ,and The area is marked as having a moderate warning level, and a moderate warning is sent to the plant data management platform; if the maximum percentage of lesion area within the canopy images of different plants continuously collected during the current growth cycle is... ,in A mild warning is marked for the block based on a pre-set adjustment coefficient, and a mild warning is sent to the plant data management platform. The warning signals of different blocks under different growth cycles are then collected and uploaded to the plant data management platform.

[0022] The plant application trajectory data mapping module summarizes the information on the areas requiring application, disease data, and early warning markers under different growth cycles, constructs multi-block application files under different cycles, and generates application prescriptions and application trajectories for multi-block applications under different cycles for effective application data monitoring. It should be specifically noted that the plant application trajectory data mapping module includes a sub-module for optimizing application trajectory data in different blocks and a sub-module for recording and monitoring effective application data. The sub-module for optimizing application trajectory data in different blocks acquires information on each block of the target peanut field at different growth stages, block disease data, and block warning markers. It summarizes the block location, disease type, and warning signal of the blocks that need to be treated at different growth stages, constructs multi-block application files for different stages, and generates application prescriptions and application trajectories for multiple blocks at different stages. The application prescription includes the type and concentration of pesticide solution for each block at different stages, and the application trajectory includes the operation speed and number of applications for each block at different stages. The administrator needs to predefine the blocks with different warning markers through the plant data integrated management platform, and plan different threshold operation speeds and different number of applications for blocks with severe, moderate, and mild warnings. For blocks without warnings, the application equipment is automatically shut down, and the application prescription and application trajectory are sent to the plant data integrated management platform for manual review. The effective pesticide application data recording and monitoring submodule acquires the number of pesticide applications, pesticide application data, and pesticide application trajectory for different blocks in different periods of the target peanut field. The pesticide application data includes the type of pesticide solution, concentration of pesticide solution, start and end interval of pesticide application, and meteorological data of pesticide application, which are summarized into effective pesticide application data for different blocks under different plant growth cycles.

[0023] The multi-block plant disease status intervention and analysis module evaluates different blocks at multiple time periods after each application, sets up control image data before application, tracks and compares the disease status of plants in multiple blocks before and after application, and analyzes the treatment effect of application in each block. To further explain, the multi-block plant disease status intervention analysis module includes a multi-block drug treatment intervention data association submodule and a plant status tracking and comparison submodule before and after intervention. The multi-block drug treatment intervention data association submodule obtains effective drug treatment data for different blocks under different cycles, records the number of drug treatment operations in different blocks, extracts the end time of each drug treatment operation in different blocks under different cycles, and marks it as a drug treatment intervention. At any given time, an initial short-term assessment and a secondary assessment are conducted for different blocks after each application operation. The initial short-term assessment period is set as follows: The second evaluation period is set as ,in, , The system is pre-configured by the administrator to obtain plant canopy image data collected by the multidimensional growth data acquisition module for different application blocks during the initial short-term evaluation period and the secondary evaluation period after each application. The plant canopy image data collected during the initial short-term evaluation period and the secondary evaluation period after each application in each block are correlated with the current application data and summarized into application intervention evaluation image data for different blocks under different growth cycles.

[0024] Further details are needed regarding the submodule for tracking and comparing plant status before and after intervention application. This submodule acquires image data of each intervention application in multiple blocks at different growth stages, and simultaneously extracts the data for each intervention application in different blocks at different growth stages. The last canopy image data collected by the multidimensional growth data acquisition module of the previous time block is used to label the pesticide application control image data. The pesticide application intervention evaluation image data and pesticide application control image data of each pesticide application operation in different blocks under different growth cycles are correlated, and the pesticide application control image data of each pesticide application operation in different blocks are extracted separately. Time-lapse image data Real-time image data is input into a pre-trained lesion spot learning model. The model identifies lesion area from images extracted during the current application of pesticides in a given area. It also includes control images extracted before and after a specific application in a given area. Time-lapse image data The lesion areas identified from the time-lapse image data are respectively , , The formula is used to determine the disease status of the plants in the current pesticide application area: in, For pesticide application operations The image recognition system for lesion area reduction sets a threshold. When the image extracted from the current application of pesticides in a certain area shows that the plant disease area status meets the above formula, the application is marked as valid. When the image extracted from the current application of pesticides in a certain area shows that the plant disease area status does not meet the above formula, the application is marked as invalid.

[0025] The quantitative qualification assessment module for plant disease control indicators summarizes and analyzes the number of effective and ineffective pesticide application markers in different blocks of the target peanut field under different growth stages, and makes a qualification assessment for the pesticide control data of each block.

[0026] To further explain, the quantitative qualification judgment module for plant disease control indicators includes a sub-module for quantitative analysis of plant control effectiveness indicators and a sub-module for qualification assessment of regional blocks. The sub-module for quantitative analysis of plant control effectiveness indicators summarizes the number of effective and ineffective pesticide application marks in different blocks of the target peanut field under different growth stages. It analyzes the proportion of the number of effective pesticide application marks in different blocks under different growth stages to the total number of pesticide applications. If the proportion of the number of effective pesticide application marks in a certain block under different growth stages of peanut plants to the total number of pesticide applications is greater than or equal to a set threshold, the block information is sent to the sub-module for qualification assessment of regional blocks. The sub-module for qualification assessment of regional blocks marks the block as qualified. If the proportion of the number of effective pesticide application marks in a certain block under different growth stages of peanut plants to the total number of pesticide applications is less than the set threshold, the sub-module for qualification assessment of regional blocks marks the block as unqualified. The sub-module for qualification assessment of regional blocks summarizes the block information of qualified and unqualified blocks in the target peanut field and sends it to the plant data integrated management platform for manual review.

[0027] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A data acquisition system for disease control based on peanut growth data, characterized in that: The system includes a multi-block plant segmentation and monitoring module, a multi-dimensional growth data acquisition module for plant segments, a plant application trajectory data mapping module, a multi-block plant disease status intervention and analysis module, a plant disease control indicator quantitative qualification judgment module, and a plant data comprehensive management platform. The multi-block division monitoring module collects spatial data, environmental data, area data, and basic soil data of peanut plant planting, and summarizes them into data information for the target peanut field. It dynamically divides the target peanut field into multiple blocks and establishes a multi-block peanut planting file. The multidimensional growth data acquisition module of the partitioned blocks collects growth data of peanut planting in different blocks at different time dimensions according to different partitions, performs preprocessing, identifies disease types and lesion areas in different blocks under different growth cycles, constructs a disease data early warning rule library, and provides intelligent early warning for plant disease data in different blocks; The plant application trajectory data mapping module summarizes the information on the areas requiring application, disease data, and early warning markers under different growth cycles, constructs multi-block application files under different cycles, and generates application prescriptions and application trajectories for multi-block applications under different cycles for effective application data monitoring. The multi-block plant disease status intervention and analysis module evaluates different blocks at multiple time periods after each application, sets up control image data before application, tracks and compares the disease status of plants in multiple blocks before and after application, and analyzes the treatment effect of application in each block. The quantitative qualification assessment module for plant disease control indicators summarizes and analyzes the number of effective and ineffective pesticide application markers in different blocks of the target peanut field under different growth stages, and makes a qualification assessment for the pesticide control data of each block.

2. The disease control data acquisition system based on peanut growth data according to claim 1, characterized in that... The plant multi-block division monitoring module includes a basic spatial information acquisition and survey sub-module and a multi-factor calibration dynamic block division and archiving sub-module. The basic spatial information acquisition and survey sub-module includes a GPS positioning unit, a UAV data connection and acquisition unit, and a historical data import unit. The GPS positioning unit obtains the geographical boundary coordinates of the target peanut field. The UAV data connection and acquisition unit connects to an external UAV to collect images of the target peanut field. The historical data import unit imports the historical planting data of the target peanut field, including historical environmental data, historical plant disease and pest data, historical yield maps, and historical soil sampling and monitoring data. These data are then summarized into data information of the target peanut field and sent to the multi-factor calibration dynamic block division and archiving sub-module.

3. The disease control data acquisition system based on peanut growth data according to claim 2, characterized in that... The multi-factor calibration dynamic block division and archiving submodule pre-calculates the area of ​​the target peanut field based on its geographical boundary coordinates, pre-divides it according to the same management area, and summarizes it into initial blocks. Based on images of peanut plants within each initial block collected by a drone, it determines the row spacing and plant spacing of the peanut plants within each initial block, and estimates the number of peanut plants within each initial block. If the number of peanut plants within an initial block exceeds a set threshold, it is further divided according to equal areas to obtain different blocks of the target peanut field. Simultaneously, it collects historical planting data... If all the locations corresponding to the divided blocks are high-incidence areas for plant diseases and pests, they are marked. If only some of the locations corresponding to the divided blocks are high-incidence areas for plant diseases and pests, the high-incidence areas within the current block are divided into separate blocks, and each block is assigned a unique identifier. All block information within the current target peanut field is summarized, including block identifier, block location, block area, estimated number of plants in the block, and high-incidence marking for diseases and pests in the block. Based on the historical data import unit, environmental topographic data and soil data are matched for each block information.

4. The disease control data acquisition system based on peanut growth data according to claim 1, characterized in that... The multi-dimensional growth data acquisition module for the peanut field includes several multispectral cameras, environmental sensors, a multi-temporal-dimensional differentiated data acquisition sub-module, and a plant growth and disease data fusion and evaluation sub-module. The multi-temporal-dimensional differentiated data acquisition sub-module pre-constructs a data acquisition framework for each cycle according to the peanut growth cycle within the target peanut field. The administrator pre-sets the basic data acquisition frequency of the multi-temporal-dimensional differentiated data acquisition sub-module through the plant data integrated management platform. Combined with several multispectral cameras, environmental sensors, and drone data connection acquisition units, it acquires plant canopy image data and key environmental data for each block within the target peanut field according to different cycle periods. Simultaneously, it imports historical plant disease and pest data from different cycles of the peanut plants through the historical data import unit to determine whether the current cycle period falls within a historically high-incidence period for plant diseases and pests. If so, the multi-temporal-dimensional differentiated data acquisition... The submodule automatically increases the data collection frequency of all divided blocks within the time period to a high-frequency mode until the marked historical high-incidence period of plant diseases and pests ends. If not, the administrator uses the plant data integrated management platform to pre-establish a threshold database of high-incidence environmental conditions for diseases, obtains the real-time environmental monitoring data for the current period, and compares the real-time environmental monitoring data with any environmental data in the high-incidence environmental condition threshold database. If any environmental data is greater than the corresponding high-incidence environmental data threshold, the environmental data for the current period is determined to be a high-incidence environment for diseases. The multi-time dimension differentiated data collection submodule automatically increases the data collection frequency of all divided blocks within the time period to a high-frequency mode until the real-time monitored environmental data are all below the high-incidence environmental condition threshold, and then returns to the basic collection frequency. The peanut growth data collected according to the data collection framework for each period is summarized and archived.

5. A disease control data acquisition system based on peanut growth data according to claim 4, characterized in that... The plant growth and disease data fusion and evaluation submodule acquires real-time plant canopy image data of peanuts at different growth stages within each cycle's data acquisition framework. It preprocesses the real-time acquired plant image data, filtering out unqualified images, and classifies and summarizes the acquired images according to different growth stages and block information. It extracts features from different blocks within the same growth stage, including peanut plant leaf color, texture, and morphology data. A pre-trained disease spot learning model is constructed, and the first acquired plant canopy image data for each block within each growth stage is pre-extracted. As a reference image for different peanut plant growth cycles, the reference images of different blocks under a certain growth cycle are input into a pre-trained disease spot learning model to detect plant disease spots. The disease spot images of the reference images under each block are identified, the disease type and lesion area of ​​the reference images of different blocks under the current growth cycle are identified, the disease type of different blocks under the current growth cycle is identified and disease is marked, and the lesion area of ​​the reference images under the blocks with disease markings under the current growth cycle is identified. The administrator can pre-set the threshold of the total lesion area of ​​different blocks under different growth cycles through the plant data integrated management platform. Pre-determine the growth cycle period to which the disease-marked blocks belong. If the proportion of lesion area in the baseline image of the disease-marked blocks in a certain growth cycle is greater than the pre-set threshold for lesion area in different blocks in the current growth cycle, mark the block with a severe warning and send a severe warning to the plant data management platform. If the percentage of lesion area in the baseline image of a disease-marked block during a certain growth cycle is less than a pre-set threshold for lesion area in different blocks during the current growth cycle, continuously collected plant canopy image data for the disease-marked block during the current growth cycle is input into a pre-trained disease spot learning model. This model monitors the percentage of lesion area within different continuously collected plant canopy images of that block. The percentage of lesion area within different continuously collected plant canopy images of that block during the current growth cycle is set as follows: Set a pre-defined threshold for lesion area in different regions during the current growth cycle. If the percentage of lesion area inside the canopy images of different plants continuously collected in this block during the current growth cycle... ,and The area is marked as a severe warning and a severe warning is sent to the plant data management platform; if the proportion of lesion area inside different plant canopy images continuously collected in this area during the current growth cycle is... ,and The area is marked as having a moderate warning level, and a moderate warning is sent to the plant data management platform; if the maximum percentage of lesion area within the canopy images of different plants continuously collected during the current growth cycle is... ,in A mild warning is marked for the block based on a pre-set adjustment coefficient, and a mild warning is sent to the plant data management platform. The warning signals of different blocks under different growth cycles are then collected and uploaded to the plant data management platform.

6. A disease control data acquisition system based on peanut growth data according to claim 1, characterized in that... The plant application trajectory data mapping module includes a sub-module for optimizing application trajectory data in different blocks and a sub-module for recording and monitoring effective application data. The sub-module for optimizing application trajectory data in different blocks acquires information on each block of the target peanut field at different growth cycles, disease data, and warning markers. It summarizes the location, disease type, and warning signals of the blocks that need to be treated at different growth cycles, constructs application files for multiple blocks at different cycles, and generates application prescriptions and application trajectories for multiple blocks at different cycles. The application prescription includes the type and concentration of pesticide solution for each block at different cycles, and the application trajectory includes the operating speed and number of applications for each block at different cycles. The administrator needs to predefine the blocks with different warning markers through the plant data integrated management platform, and plan different threshold operating speeds and different number of applications for blocks with severe, moderate, and mild warnings. For blocks without warnings, the application equipment is automatically shut down, and the application prescription and application trajectory are sent to the plant data integrated management platform for manual review. The effective pesticide application data recording and monitoring submodule acquires the number of pesticide applications, pesticide application data, and pesticide application trajectory for different blocks in different periods of the target peanut field. The pesticide application data includes the type of pesticide solution, concentration of pesticide solution, start and end interval of pesticide application, and meteorological data of pesticide application, which are summarized into effective pesticide application data for different blocks under different plant growth cycles.

7. A disease control data acquisition system based on peanut growth data according to claim 1, characterized in that... The multi-block plant disease status intervention analysis module includes a multi-block drug treatment intervention data association submodule and a plant status tracking and comparison submodule before and after intervention. The multi-block drug treatment intervention data association submodule acquires effective drug treatment data for different blocks under different cycles, records the number of drug treatment operations for different blocks, extracts the end time of each drug treatment operation for different blocks under different cycles, and marks it as a drug treatment intervention. At any given time, an initial short-term assessment and a secondary assessment are conducted for different blocks after each application operation. The initial short-term assessment period is set as follows: The second evaluation period is set as ,in, , The system is pre-configured by the administrator to obtain plant canopy image data collected by the multidimensional growth data acquisition module for different application blocks during the initial short-term evaluation period and the secondary evaluation period after each application. The plant canopy image data collected during the initial short-term evaluation period and the secondary evaluation period after each application in each block are correlated with the current application data and summarized into application intervention evaluation image data for different blocks under different growth cycles.

8. A disease control data acquisition system based on peanut growth data according to claim 7, characterized in that... The plant status tracking and comparison submodule before and after intervention and application acquires image data of each application and intervention in multiple blocks at different growth stages, and simultaneously extracts the data for each application and intervention in different blocks at different growth stages. The last canopy image data collected by the multidimensional growth data acquisition module of the previous time block is used to label the pesticide application control image data. The pesticide application intervention evaluation image data and pesticide application control image data of each pesticide application operation in different blocks under different growth cycles are correlated, and the pesticide application control image data of each pesticide application operation in different blocks are extracted separately. Time-lapse image data Real-time image data is input into a pre-trained lesion spot learning model. The model identifies lesion area from images extracted during the current application of pesticides in a given area. It also includes control images extracted before and after a specific application in a given area. Time-lapse image data The lesion areas identified from the time-lapse image data are respectively , , The formula is used to determine the disease status of the plants in the current pesticide application area: in, For pesticide application operations The image recognition system for lesion area reduction sets a threshold. When the image extracted from the current application of pesticides in a certain area shows that the plant disease area status meets the above formula, the application is marked as valid. When the image extracted from the current application of pesticides in a certain area shows that the plant disease area status does not meet the above formula, the application is marked as invalid.

9. A disease control data acquisition system based on peanut growth data according to claim 1, characterized in that... The plant disease control index quantification and qualification judgment module includes a plant control effect index quantification analysis submodule and a block qualification assessment submodule. The plant control effect index quantification analysis submodule summarizes the number of effective and ineffective pesticide application operation markers in different blocks of the target peanut field under different growth stages, and analyzes the proportion of the number of effective pesticide application operation markers in different blocks under different growth stages to the total number of pesticide applications. If the proportion of the number of effective pesticide application operation markers in a certain block under different growth stages of peanut plants to the total number of pesticide applications is greater than or equal to a set threshold, the block information is sent to the block qualification assessment submodule, which marks the block as qualified. If the proportion of the number of effective pesticide application markers in a certain block under different growth stages of peanut plants to the total number of pesticide applications is less than the set threshold, the block qualification assessment submodule marks the block as unqualified. The block qualification assessment submodule summarizes the block information of qualified and unqualified blocks in the target peanut field and sends it to the plant data integrated management platform for manual review.

Images