An intelligent early warning system for agricultural and forestry diseases and insect pests based on monitoring images

Abstract

The application relates to the technical field of prevention and treatment of plant diseases and insect pests, and provides an intelligent early warning system for plant diseases and insect pests based on monitoring images, which comprises a regional monitoring module, local devices such as humidity sensors, temperature sensors and monitoring camera probes connected to the regional monitoring module, a plurality of groups of the local devices equidistantly arranged in a plant region, and the regional monitoring module acquiring environmental data of the arranged region through the local devices; and a UAV base station control module mounted on a UAV base station. The environmental data and image information of the plant region can be acquired in real time, the disease and insect pest risk can be rapidly determined, an early warning signal can be timely sent, the disease and insect pest monitoring efficiency is improved, the type and position of the disease and insect pest can be accurately identified, the related data of each region are collected, the intelligent word items are compared, and the plant-related personnel can better trace the disease and insect pest source of the plant.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of prevention and cure of plant diseases and insect pests, in particular to an intelligent early warning system for plant diseases and insect pests based on monitoring images. BACKGROUND

[0002] With the modernization and scale of agricultural production and the increasing forest area in China, the harm of plant diseases and insect pests to crops and trees is becoming more and more serious, and it is particularly important to monitor and manage plant diseases and insect pests in a timely and effective manner.

[0003] According to the search, the patent number CN118887563A discloses an agricultural and forestry pest monitoring system and a monitoring method based on the Internet of Things, which realizes comprehensive monitoring and intelligent management of agricultural and forestry pests by constructing a complete Internet of Things architecture. The system is composed of a sensor network, a data acquisition module, an image acquisition device, a transmission system, a monitoring center / cloud platform, an intelligent analysis system, an early warning module, a geographic information system, a remote control terminal and a prevention and control device. Based on the above monitoring system, the present application also provides a forestry disease monitoring scheme based on unmanned aerial vehicles, including a multi-rotor unmanned aerial vehicle scheme and a fixed-wing unmanned aerial vehicle scheme. The agricultural and forestry pest monitoring system based on the Internet of Things and the monitoring method thereof realize comprehensive monitoring, intelligent analysis and early warning, as well as remote control and prevention and control of agricultural and forestry pests by integrating Internet of Things technology and unmanned aerial vehicle technology, improve the management efficiency and prevention and control effect of agricultural and forestry pests, and have important practical application value.

[0004] Although the detection method can realize intelligent analysis and early warning, as well as remote control and prevention and control, it still lacks the collection accuracy and accuracy of related data of the pest area, making it inconvenient for relevant personnel to further understand the root cause of the plant diseases and insect pests. Therefore, an intelligent early warning system for plant diseases and insect pests based on monitoring images is needed. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides an intelligent early warning system for plant diseases and insect pests based on monitoring images, which solves the problem of lack of collection accuracy and accuracy of related data of the pest area, making it inconvenient for relevant personnel to further understand the root cause of the plant diseases and insect pests.

[0006] To achieve the above purpose, the present application realizes the following technical solutions:

[0007] An intelligent early warning system for plant diseases and insect pests based on monitoring images, comprising:

[0008] A regional monitoring module is connected with local devices such as humidity sensors, temperature sensors, monitoring camera probes, etc., the number of the local devices is multiple groups, and the local devices are equidistantly arranged in the agricultural and forestry region, and the regional monitoring module obtains environmental data of the arranged region through the local devices;

[0009] A UAV base station control module is mounted on a UAV base station to connect a monitoring and shooting UAV to issue a cruise monitoring and shooting task and receive agricultural and forestry shooting image data;

[0010] An agricultural and forestry state determination module receives data transmitted by the regional monitoring module to determine whether the region has a disease and pest risk and is in a disease and pest state, and sends a cruise monitoring and shooting task to the UAV base station control module according to the determination result;

[0011] An alarm module outputs corresponding early warning signals and emergency alarm signals based on the determination result output by the agricultural and forestry state determination module;

[0012] A monitoring image post-processing module is used to process agricultural and forestry shooting image data transmitted by the UAV base station control module, label agricultural and forestry disease and pest image features, and generate a response report;

[0013] A cloud service module is used to transmit data and commands between local devices, is responsible for uploading and analyzing related data, downloading agricultural and forestry disease and pest data, and provides identification reference data for the agricultural and forestry state determination module and the monitoring image post-processing module.

[0014] Preferably, the regional monitoring module comprises:

[0015] An environmental monitoring unit is used to collect environmental data of the region, including humidity, temperature, agricultural and forestry images, and agricultural and forestry environmental images;

[0016] A data correction unit is used to determine and delete data with a large variation range before and after the data to exclude error and abnormal data collected by external human factors;

[0017] An environmental data error prevention unit obtains local weather data through a connected network to determine whether the data to be deleted by the data correction unit is affected by weather factors, so as to cancel the deletion of environmental data affected by weather factors.

[0018] Preferably, the UAV base station control module comprises:

[0019] A cruise route generation unit is used to generate a cruise shooting route of the UAV and exclude shooting locations that do not need to be cruised;

[0020] The monitoring photographing task dispatching unit is used to issue a photographing task of a required photographing area before the UAV takes off, so that the UAV reaches a specified range area to start image photographing and obtain image data and photographing coordinates in the air above the farmland and forest;

[0021] The cruising monitoring image splicing unit is used to integrate and splice the image data in the air above the farmland and forest based on the photographing coordinates;

[0022] The image splicing repair unit corrects and repairs the image data in the air above the farmland and forest spliced by the cruising monitoring image splicing unit based on an intelligent model of IOPaint as a repair engine.

[0023] Preferably, the farmland and forest state judgment module comprises:

[0024] The pest identification and judgment unit is used to identify and judge the environmental data output by the regional monitoring module in view of the farmland and forest tree features caused by pests, determine what kind of farmland and forest pests affect the farmland and forest, and issue an alarm message;

[0025] The disease identification and judgment unit is used to identify and judge the environmental data output by the regional monitoring module in view of the farmland and forest tree features caused by diseases, determine what kind of farmland and forest bacteria, farmland and forest fungi, and farmland and forest viruses affect the farmland and forest, and issue an alarm message;

[0026] The disease and pest environmental development trend judgment unit is used to identify and judge the environmental data output by the regional monitoring module, count and determine whether the environmental values meet the conditions of disease spread and the conditions of attracting pests, and issue a warning message;

[0027] The disease and pest feature download unit is used to download disease and pest identification feature data for disease and pest symptom identification and comparison verification work;

[0028] The abnormal area anchor point generation unit is used to generate the range coordinates of the area issuing the warning message and the alarm message, and transmit them to the UAV base station control module.

[0029] Preferably, the monitoring image post-processing module comprises:

[0030] The image separation unit is used to identify and separate normal farmland and forest and farmland and forest affected by diseases and pests in the image, independently render the farmland and forest image affected by diseases and pests, and improve the recognition degree;

[0031] The disease and pest spread trend prediction unit predicts the disease and pest spread direction based on the farmland and forest image affected by diseases and pests independently rendered by the image separation unit;

[0032] The report export unit is used to export the farmland and forest image affected by diseases and pests and the data of disease and pest spread direction prediction independently rendered.

[0033] Preferably, the cloud service module comprises:

[0034] A cloud pest characteristic database records and stores the characteristic data of the forest pests in a numbered form;

[0035] A data transmission unit for receiving and transmitting data and commands;

[0036] A cloud record storage unit for storing and backing up historical monitoring data to observe the development and changes of the forest and summarize the prevention and control experience.

[0037] A forest pest intelligent early warning method based on monitoring images, comprising the following steps:

[0038] Step one, the regional monitoring module obtains the environmental data of the forest area in real time through the connected local devices such as humidity sensors, temperature sensors, monitoring camera probes, etc., including humidity, temperature, forest images, forest environment images, etc., and transmits the obtained data to the forest state determination module;

[0039] Step two, the forest state determination module receives the data transmitted by the regional monitoring module, makes an abnormal change early warning determination, and when it is in a pest state, sends a cruise monitoring shooting task to the unmanned aerial vehicle base station control module and connects the alarm module to send an emergency alarm signal, otherwise, when it is determined that there is a pest risk, it connects the alarm module to send a warning signal;

[0040] Step three, after the unmanned aerial vehicle base station control module receives the shooting task, the cruise route generation unit generates the cruise shooting route of the unmanned aerial vehicle, excludes the shooting locations that do not need to be cruised, and the monitoring shooting task assignment unit assigns the shooting task of the required shooting area before the unmanned aerial vehicle takes off;

[0041] Step four, the unmanned aerial vehicle arrives at the specified range area to start image shooting, obtains the image data and shooting coordinates above the forest, the cruise monitoring image splicing unit integrates and splices the image data based on the shooting coordinates, the image splicing repair unit repairs the edges of the spliced image data, and the processed image data is transmitted to the monitoring image post-processing module;

[0042] Step five, the image separation unit of the monitoring image post-processing module identifies and separates the normal forest and the forest affected by pests in the image, independently renders the forest affected by pests, and improves the recognition degree;

[0043] Step six, the pest spread trend prediction unit predicts the pest spread direction of the independently rendered forest affected by pests, the report export unit exports the independently rendered forest affected by pests and the pest spread direction prediction data, and generates a scheme response report.

[0044] Preferably, the specific steps of the abnormal variation early warning determination in step two are as follows:

[0045] S1, download pest characteristic number and disease risk characteristic number through a cloud pest and disease characteristic database, interpret the pest characteristic number and the disease risk characteristic number according to the number entries built in the pest and disease identification determination unit;

[0046] S2, based on the hazard identification determination of the pest and disease identification determination unit on the environmental data output by the area monitoring module, accumulate the hazard variation based on the potential hazard identification determination unit of the disease;

[0047] S3, increase the abnormal variation value by 1 point each time the hazard identification determination is successful, and increase the early warning variation value by 1 point each time the potential hazard identification determination is successful;

[0048] S4, set the abnormal variation value threshold A, and set the early warning variation value threshold B;

[0049] S5, when the abnormal variation value is greater than or equal to A, output an alarm message, and when the abnormal variation value is less than A and the early warning variation value is greater than or equal to B, output an early warning message;

[0050] S6, when the abnormal variation value is less than A and the early warning variation value is less than B for more than one hour, the abnormal variation value and the early warning variation value are reduced by 1 point.

[0051] Preferably, the pest characteristic number in S1 includes the parent number of the agricultural and forestry leaf lesion characteristic, the mother number of the agricultural and forestry branch lesion characteristic, and the son number of the agricultural and forestry pest species.

[0052] Preferably, the disease risk characteristic number in S1 includes the parent number of the agricultural and forestry leaf lesion characteristic, the mother number of the agricultural and forestry branch lesion characteristic, and the son number of the agricultural and forestry bark surface adhering object species.

[0053] The present application provides an intelligent early warning system for agricultural and forestry diseases and pests based on monitoring images. It has the following beneficial effects:

[0054] 1. The present application can obtain real-time environmental data and image information of the agricultural and forestry area, quickly determine the disease and pest risk, and timely issue an early warning signal, thereby improving the efficiency of disease and pest monitoring, and accurately identifying the type and location of the disease and pest, and collecting relevant data of each area, and comparing with intelligent entries, so that agricultural and forestry related personnel can better trace the source of the disease and pest of the agricultural and forestry.

[0055] 2. The present application can generate an abnormal area anchor point to provide accurate shooting tasks for a drone, ensure the pertinence and effectiveness of monitoring, and automatically splice the pictures taken to generate a complete cruise image, without the need for manual review, quickly locate the disease and pest location of the agricultural and forestry, and separately divide the disease area for relevant personnel to quickly interpret and understand.

[0056] 2、The application can effectively eliminate the interference of external factors, adapt to complex agricultural and forestry environment, ensure the reliability of monitoring data, generate detailed pest spread trend prediction and response plan report, provide scientific decision support for agricultural and forestry managers, help to take preventive measures in time, and reduce the loss of agricultural and forestry resources caused by pests. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 It is a monitoring image-based agricultural and forestry pest intelligent early warning system diagram of the application;

[0058] Figure 2 It is a regional monitoring module system schematic diagram of the monitoring image-based agricultural and forestry pest intelligent early warning system of the application;

[0059] Figure 3 It is a UAV base station control module system schematic diagram of the monitoring image-based agricultural and forestry pest intelligent early warning system of the application;

[0060] Figure 4 It is a monitoring image post-processing module system schematic diagram of the monitoring image-based agricultural and forestry pest intelligent early warning system of the application;

[0061] Figure 5 It is a system schematic diagram of the agricultural and forestry state determination module of the monitoring image-based agricultural and forestry pest intelligent early warning system of the application;

[0062] Figure 6 It is a cloud service module system schematic diagram of the monitoring image-based agricultural and forestry pest intelligent early warning system of the application. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.

[0064] Embodiment:

[0065] As an aspect of the present application, please refer to the drawings Figure 1 -Appendix Figure 6 The embodiment of the application provides a monitoring image-based agricultural and forestry pest intelligent early warning system, which comprises:

[0066] The regional monitoring module is connected with local devices such as humidity sensors, temperature sensors, monitoring camera probes and the like, the number of local devices is multiple groups, and the local devices are equidistantly arranged in the agricultural and forestry region. The regional monitoring module obtains environmental data of the arranged region through the local devices, including:

[0067] The environmental monitoring unit is used for collecting environmental data of the region, including humidity, temperature, agricultural and forestry images, agricultural and forestry environment images and the like;

[0068] The data correction unit is used for determining and deleting data with a large variation range before and after, so as to exclude error abnormal data collected by external human factors;

[0069] The environmental data error prevention unit obtains local weather data through a connected network to determine whether the data to be deleted by the data correction unit is affected by weather factors, so as to cancel the deletion of environmental data affected by weather factors.

[0070] The unmanned aerial vehicle base station control module is mounted on the unmanned aerial vehicle base station to connect the monitoring and shooting unmanned aerial vehicle to issue a cruise monitoring and shooting task, and receive agricultural and forestry shooting image data, including:

[0071] The cruise route generation unit is used for generating a cruise shooting route of the unmanned aerial vehicle, and excluding shooting locations that do not need to be cruised;

[0072] The monitoring and shooting task dispatching unit is used for issuing a shooting task of a required shooting region before the unmanned aerial vehicle takes off, so that the unmanned aerial vehicle arrives at a specified range region to start image shooting, and obtains image data and shooting coordinates above the agricultural and forestry region;

[0073] The cruise monitoring image splicing unit is used for integrating and splicing the image data above the agricultural and forestry region based on the shooting coordinates;

[0074] The image splicing repair unit uses an intelligent model based on IOPaint as a repair engine to repair the image data above the agricultural and forestry region spliced by the cruise monitoring image splicing unit at the edge.

[0075] The agricultural and forestry state determination module receives data transmitted by the regional monitoring module to determine the data, determines whether the region has a pest risk and is in a pest state, and sends a cruise monitoring and shooting task to the unmanned aerial vehicle base station control module according to the determination result, including:

[0076] The pest identification determination unit is used for identifying and determining the environmental data output by the regional monitoring module according to the characteristics of the agricultural and forestry trees caused by pests, determining the influence of agricultural and forestry pests on the agricultural and forestry, and issuing an alarm message;

[0077] A disease identification and determination unit is configured to identify and determine the environmental data output by the area monitoring module with respect to the characteristics of the forest caused by the disease, determine the influence of the forest on the forest bacteria, the forest fungus and the forest virus, and issue an alarm message.

[0078] A disease and pest environmental development trend judgment unit is configured to identify and determine the environmental data output by the area monitoring module, count and determine whether the environmental values meet the conditions of disease spread and whether they meet the conditions of attracting pests, and issue a warning message.

[0079] A disease and pest characteristic download unit is configured to download disease and pest identification characteristic data for disease and pest symptom identification and comparison verification.

[0080] An abnormal area anchor point generation unit is configured to generate the range coordinates of the area issuing the warning message and the alarm message, and transmit them to the unmanned aerial vehicle base station control module.

[0081] An alarm module is configured to output corresponding warning signals and emergency alarm signals based on the determination results output by the forest state determination module.

[0082] A monitoring image post-processing module is configured to process the forest image data transmitted by the unmanned aerial vehicle base station control module, label the forest disease and pest image characteristics, and generate a response report, including:

[0083] An image separation unit is configured to identify and separate normal forests and forests affected by diseases and pests in the image, independently render the forest images affected by diseases and pests, and improve the recognition degree.

[0084] A disease and pest spread trend prediction unit is configured to predict the spread direction of the disease and pest based on the forest images affected by the disease and pest independently rendered by the image separation unit.

[0085] A report export unit is configured to export the independently rendered forest images affected by the disease and pest and the data of the disease and pest spread direction prediction.

[0086] A cloud service module is configured to provide data and command transmission between local devices, and is responsible for uploading and analyzing related data, downloading forest disease and pest data, providing identification reference data for the forest state determination module and the monitoring image post-processing module, including:

[0087] A cloud disease and pest characteristic database is configured to record and store forest disease and pest characteristic data in a numbered form.

[0088] A data transmission unit is configured to receive and transmit data and commands.

[0089] A cloud record saving unit is configured to save and backup historical monitoring data to observe the development and changes of the forest and summarize prevention and control experience.

[0090] Based on the above-mentioned one kind based on monitoring image's agriculture and forestry disease and pest intelligent early warning system, as another aspect of the present application, a kind of based on monitoring image's agriculture and forestry disease and pest intelligent early warning method, including the following steps:

[0091] Step one, area monitoring module passes through the local equipment such as humidity sensor, temperature sensor, monitoring camera probe connected, real-time obtains the environmental data of agricultural and forestry area, including humidity, temperature, agricultural and forestry image, agricultural and forestry environment image etc., and the data obtained are transmitted to agricultural and forestry state determination module;

[0092] Step two, agricultural and forestry state determination module receives the data transmitted by area monitoring module, carries out abnormal change early warning determination, when being in the state of disease and pest, then send the cruise monitoring shooting task to unmanned aerial vehicle base station control module, and connect alarm module to send emergency alarm signal, otherwise determine that there is disease and pest risk then connect alarm module to send early warning signal, the specific steps of abnormal change early warning determination are as follows:

[0093] S1, download pest characteristic number and disease risk characteristic number by cloud pest characteristic database, and interpret pest characteristic number and disease risk characteristic number according to the number of entries inbuilt in pest identification determination unit;

[0094] S2, based on the hazard identification determination unit and the disease identification determination unit, the environmental data output by area monitoring module are identified, and the potential hazard identification determination unit is accumulated based on hazard change;

[0095] S3, hazard identification determination increases 1 point abnormal change value each time determination is successful, and potential hazard identification determination increases 1 point early warning change value each time determination is successful;

[0096] S4, set abnormal change value threshold A, set early warning change value threshold B;

[0097] S5, when abnormal change value ≥A, output alarm message, when abnormal change value <A and early warning change value ≥B, output early warning message;

[0098] S6, when abnormal change value <A and early warning change value <B exceed one hour, abnormal change value and early warning change value are reduced by 1 point independently.

[0099] Wherein, pest characteristic number includes the parent number of agricultural and forestry leaf lesion characteristic, the mother number of agricultural and forestry branch lesion characteristic and the son number of agricultural and forestry pest species. Disease risk characteristic number includes the parent number of agricultural and forestry leaf lesion characteristic, the mother number of agricultural and forestry branch lesion characteristic and the son number of agricultural and forestry tree bark surface attachment species.

[0100] Step three, after the unmanned aerial vehicle base station control module receives the shooting task, the cruise route generation unit generates the cruise shooting route of the unmanned aerial vehicle, excludes the shooting places that do not need to cruise, and the monitoring shooting task distribution unit issues the shooting task of the required shooting area before the unmanned aerial vehicle takes off;

[0101] Step four, the unmanned aerial vehicle arrives at the specified range area to start image shooting, obtains the image data and shooting coordinates in the air above the farmland, the cruise monitoring image splicing unit integrates and splices the image data based on the shooting coordinates, the image splicing repair unit repairs the edges of the spliced image data, and the processed image data is transmitted to the monitoring image post-processing module;

[0102] Step five, the image separation unit of the monitoring image post-processing module identifies and separates the normal farmland and the farmland affected by pests and diseases in the image, independently renders the farmland image affected by pests and diseases, and improves the recognition degree;

[0103] Step six, the pest and disease spread trend prediction unit predicts the spread direction of the independently rendered farmland image affected by pests and diseases, the report export unit exports the independently rendered farmland image affected by pests and diseases and the data of the pest and disease spread direction prediction, and generates a scheme response report.

[0104] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. An intelligent warning method for agricultural and forestry diseases and pests based on monitoring images, based on an intelligent warning system for agricultural and forestry diseases and pests based on monitoring images, characterized in that, Comprise the following steps: Step one, the regional monitoring module obtains the environmental data of the agricultural and forestry area in real time through the connected humidity sensor, temperature sensor, monitoring camera probe, including humidity, temperature, agricultural and forestry image, agricultural and forestry environment image, and transmits the obtained data to the agricultural and forestry state determination module; Step two, the agricultural and forestry state determination module receives the data transmitted by the regional monitoring module, carries out abnormal change early warning judgment, is in the disease and pest state, then sends the cruise monitoring shooting task to the unmanned aerial vehicle base station control module, and connects the alarm module to send the emergency alarm signal, otherwise, when it is judged that there is a disease and pest risk, the alarm module is connected to send a warning signal; Step three, after receiving the shooting task, the unmanned aerial vehicle base station control module generates the cruise shooting route of the unmanned aerial vehicle, excludes the shooting place that does not need to cruise, and the monitoring shooting task assignment unit issues the shooting task of the required shooting area before the unmanned aerial vehicle takes off; Step four, the unmanned aerial vehicle arrives at the specified range area to start image shooting, obtains the image data and shooting coordinates above the agricultural and forestry, the cruise monitoring image splicing unit integrates and splices the image data based on the shooting coordinates, the image splicing repair unit repairs the edge of the spliced image data, and transmits the processed image data to the monitoring image post-processing module; Step five, the image separation unit of the monitoring image post-processing module identifies and separates the normal agricultural and forestry and the agricultural and forestry affected by the disease and pest in the image, independently renders the agricultural and forestry image affected by the disease and pest, and improves the identification degree; Step six, the disease and pest spread trend prediction unit predicts the disease and pest spread direction of the independently rendered agricultural and forestry image affected by the disease and pest, the report export unit exports the independently rendered agricultural and forestry image affected by the disease and pest and the disease and pest spread direction prediction data, and generates a scheme response report; The specific steps of the abnormal change early warning judgment in step two are: S1, download the pest characteristic number and the disease risk characteristic number through the cloud pest characteristic database, interpret the pest characteristic number and the disease risk characteristic number according to the number of entries in the built-in pest identification judgment unit; S2, based on the hazard identification judgment of the environmental data output by the regional monitoring module, the hazard identification judgment unit and the disease identification judgment unit, the potential hazard identification judgment unit accumulates the hazard change; S3, the hazard identification judgment increases 1 point of abnormal change value each time the judgment is successful, and the potential hazard identification judgment increases 1 point of early warning change value each time the judgment is successful; S4, set the abnormal change value threshold A, and set the early warning change value threshold B; S5, when the abnormal change value is greater than or equal to A, output the alarm message, when the abnormal change value is less than A and the early warning change value is greater than or equal to B, output the early warning message; S6, when the abnormal change value is less than A and the early warning change value is less than B for more than one hour, the abnormal change value and the early warning change value are reduced by 1 point; The regional monitoring module is connected with humidity sensor, temperature sensor, monitoring camera probe local device, the number of local devices is multiple groups, and is equidistantly arranged in the agricultural and forestry area, the regional monitoring module obtains the environmental data of the arranged area through the local device; The unmanned aerial vehicle base station control module is mounted on the unmanned aerial vehicle base station to connect and monitor the shooting unmanned aerial vehicle to issue a cruise monitoring and shooting task and receive the agricultural and forestry shooting image data. The agricultural and forestry state determination module receives and determines the data transmitted by the region monitoring module, determines whether the region has a disease and pest risk and is in a disease and pest state, and sends a cruise monitoring and shooting task to the unmanned aerial vehicle base station control module according to the determination result. The alarm module outputs corresponding early warning signals and emergency alarm signals based on the determination result output by the agricultural and forestry state determination module. The monitoring image post-processing module is used for processing the agricultural and forestry shooting image data transmitted by the unmanned aerial vehicle base station control module, labeling the agricultural and forestry disease and pest image features, and generating a scheme response report. The cloud service module is used for providing transmission of data and commands between local devices, is responsible for uploading and analyzing relevant data, downloading agricultural and forestry disease and pest data, and provides identification reference data for the agricultural and forestry state determination module and the monitoring image post-processing module. 2.The intelligent warning method for agricultural and forestry diseases and insect pests based on monitoring images according to claim 1, characterized in that, The S1 includes a parent number of an agricultural and forestry leaf lesion feature, a mother number of an agricultural and forestry branch lesion feature, and a child number of an agricultural and forestry pest species. 3.The intelligent warning method of agricultural and forestry diseases and insect pests based on monitoring images according to claim 1, characterized in that, The S1 includes a parent number of an agricultural and forestry leaf lesion feature, a mother number of an agricultural and forestry branch lesion feature, and a child number of an agricultural and forestry pest species.

Images