Camellia planting intelligent disaster prevention and control system and method based on deep learning

Abstract

The invention discloses a camellia planting intelligent disaster prevention and control system and method based on deep learning, and relates to the technical field of planting management, and the system comprises a prevention and control collection module, a prevention and control analysis module, a prediction module and an early warning processing module. The prevention and control acquisition module is used for acquiring biological data and environmental data in a camellia planting area; the prevention and control analysis module is used for analyzing the biological data and the environmental data to obtain biological state data and environmental state data; the prediction module performs prediction based on the biological state data and the environment state data to obtain prediction data; the early warning processing module performs prevention and treatment on the camellia planting area based on the prediction data; the method is used for solving the problems of untimely monitoring and inaccurate prevention and control in the existing planting prevention and control technology, and the problems of insect attack and disease diffusion and yield reduction and even no harvest of camellia.

Description

Technical Field

[0001] The present invention relates to the field of planting management technology, and specifically to a camellia planting intelligent disaster prevention and control system and method based on deep learning. Background Art

[0002] Camellia is shade-tolerant, moisture-loving, warm-loving, and afraid of the sun, cold, and waterlogging, and has very demanding requirements for its growing environment. During the planting process, it may encounter pests and diseases, including aphids, scale insects, mites, stem borers, and tea yellow moths, which can cause the leaves to deform, curl, or even die branches. Environmental and pest factors have a serious impact on the growth, flowering, and fruiting of camellia, affecting both yield and quality.

[0003] Existing prevention and control technologies usually identify and compare pests and diseases to determine their types and conditions, and then prevent and control them. However, this prevention and control method often requires that pests and diseases show obvious characteristics before they can be collected and accurately judged. For example, a patent application with publication number CN117196685A discloses an intelligent management method and system for agricultural product planting. The solution: screens agricultural products according to climate data, reasons for the formation of characteristic industries, characteristic crops and restrictive factors, and obtains recommended agricultural products; in the process of recommending agricultural product planting, obtains pest and disease videos, The characteristics of pests and diseases are identified through neural networks, and the types of pests and diseases are determined based on the characteristics of pests and diseases; pest and disease images are obtained and compared with the pest and disease images in the database for similarity to obtain the degree of damage; the conditions for the occurrence of pest and disease types are retrieved, and the occurrence probability and spread probability of the pest and disease types are determined in combination with the types of recommended agricultural products and agricultural operations during the planting process; if the occurrence probability and the spread probability both exceed the threshold, prevention and control methods are given according to the pest and disease type and the degree of damage. This method has the problems of untimely monitoring and inaccurate prevention and control, which will lead to the spread of pests and diseases, resulting in a reduction in camellia production or even a complete loss of harvest. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to a certain extent, by collecting and analyzing biological data and environmental data in a camellia planting area to obtain biological status data and environmental status data, and then making predictions to obtain prediction data. Finally, based on the prediction data, prevention and control treatment is carried out on the camellia planting area to solve the problems of untimely monitoring and inaccurate prevention and control in existing planting prevention and control technologies, which may lead to the spread of pests and diseases, resulting in a reduction in camellia production or even a total loss of production.

[0005] To achieve the above objectives, in the first aspect, the present application provides a camellia planting smart disaster prevention and control system based on deep learning, including a prevention and control collection module, a prevention and control analysis module, a prediction module and an early warning and disposal module;

[0006] The control and collection module is used to collect biological data and environmental data in the camellia planting area;

[0007] The prevention and control analysis module is used to analyze biological data and environmental data to obtain biological state data and environmental state data;

[0008] The prediction module performs prediction based on the biological state data and the environmental state data to obtain prediction data;

[0009] The early warning and disposal module performs prevention and control measures on the camellia planting area based on the prediction data.

[0010] Further, the control collection module includes a biological collection unit, the biological collection unit is configured with a biological collection strategy, the biological collection strategy includes setting an automatic insect situation light in the camellia planting area, the automatic insect situation light is provided with a built-in sensor, the built-in sensor is used to count the insects entering the automatic insect situation light, and the number of insects entering the automatic insect situation light is obtained once every first time interval, and set as regional insect data;

[0011] A spore catcher is set in the camellia planting area, and the spore density in the area is obtained from the spore catcher once every first time interval, and is set as regional spore data;

[0012] A plurality of camera devices are arranged in the camellia planting area, the camera devices are used to collect camellia images in different directions, and a camellia image in the area is obtained once at a second time interval, and is set as camellia image data;

[0013] The regional insect data, regional spore data and camellia image data are collectively referred to as biological data.

[0014] Further, the prevention and control collection module also includes an environment collection unit, the environment collection unit is configured with an environment collection strategy, the environment collection strategy includes setting a soil sensor in the camellia planting area, the soil sensor is used to monitor the soil moisture content and soil temperature in the soil and transmit the data in real time, and record it as real-time soil data;

[0015] An intelligent meteorological instrument is installed in the camellia planting area, and the intelligent meteorological instrument is used to monitor the regional temperature and regional humidity in real time and record them as real-time meteorological data;

[0016] The real-time soil data and real-time meteorological data are collectively referred to as environmental data.

[0017] Further, the control analysis module includes a pest analysis unit, the pest analysis unit is configured with a pest analysis strategy, the pest analysis strategy includes obtaining regional insect data and regional spore data within a continuous period of time to draw a pest data change curve chart, and obtaining a pest data change rate chart from the pest data change curve chart;

[0018] Obtain a pest data change rate graph within the third period before the pest outbreak from historical data, set it as a critical point change rate graph, perform feature extraction on the critical point change rate graph, obtain the average rate of the critical point change rate graph, and set it as a critical value;

[0019] The average rate of the insect pest data change rate graph is obtained once every third time period and is set as the insect pest change characteristic value.

[0020] Furthermore, the prevention and control analysis module also includes an image analysis unit, and the image analysis unit is configured with an image analysis strategy, which includes performing contour extraction on camellia image data to obtain a number of camellia leaf contours, performing feature extraction on the image within each camellia leaf contour, and the feature extraction includes using RGB channels to represent each pixel point of the image within the camellia leaf contour to obtain a leaf color distribution map;

[0021] Use RGB channels to represent each pixel of the normal camellia leaf image to obtain a standard color distribution map, and obtain the pixel with the smallest RGB channel and the pixel with the largest RGB channel from the standard color distribution map, and set them as the standard channel interval;

[0022] The pixels whose pixel values ​​in the leaf color distribution map are not in the standard channel interval are recorded as special pixels, and the number of special pixels and RGB channels are set as color features;

[0023] A pathological characteristic database is established, wherein the pathological characteristic database is used to store color characteristics corresponding to camellia leaves with different pathological conditions.

[0024] Furthermore, the prevention and control analysis module also includes an environmental analysis unit, which is configured with an environmental analysis strategy, which includes obtaining soil moisture content and regional humidity, and substituting them into the regional humidity influence formula In the calculation, the regional humidity impact value is obtained, where A1 is the regional humidity impact value, H0 is the soil moisture content, and S0 is the regional humidity;

[0025] Get the soil temperature and regional temperature and substitute them into the regional temperature influence formula The regional temperature impact value is calculated, where A2 is the regional temperature impact value, T t is the soil temperature, T q is the regional temperature;

[0026] Obtain regional temperature impact values ​​and regional humidity impact values ​​in real time, and draw regional impact curves;

[0027] The regional impact curve is segmented to obtain the time period type, and the section where the temperature impact value and humidity impact value are both greater than 1 is set as the high temperature and humidity period.

[0028] The section where both the temperature impact value and the humidity impact value are less than 1 is set as the low-temperature drying period.

[0029] Set the section where the temperature impact value is less than 1 and the humidity impact value is greater than or equal to 1 as the low temperature and humid period.

[0030] The section where the temperature influence value is greater than 1 and the humidity influence value is less than or equal to 1 is set as the high-temperature drying period.

[0031] Furthermore, the prediction module is configured with a prediction strategy, which includes monitoring the pest change characteristic value, and before the pest change characteristic value reaches a critical point, obtaining the symptom corresponding to the color feature of the camellia leaf from the symptom feature database, and obtaining the appropriate time period type of the symptom;

[0032] When the time period type at this time is an appropriate time period type for the symptom, the average duration of the time period type at this time is obtained and set as the predicted high incidence duration.

[0033] Furthermore, the early warning and handling module is configured with an early warning and handling strategy, and the early warning and handling strategy includes obtaining and activating a preset coping method for any symptom from a database when a predicted high-incidence duration of the symptom is obtained.

[0034] In a second aspect, the present invention provides a method for intelligent disaster prevention and control of camellia planting based on deep learning, comprising the following steps:

[0035] Collect biological and environmental data in camellia growing areas;

[0036] Analyze the biological data and the environmental data to obtain biological state data and environmental state data;

[0037] Perform prediction based on biological state data and environmental state data to obtain prediction data;

[0038] Carry out prevention and control measures in camellia growing areas based on predicted data.

[0039] Beneficial effects of the present invention: The present invention uses a variety of sensors to collect data on camellia planting areas, obtains biological data and environmental data in the camellia planting areas, can perform multi-dimensional data analysis and geographic information processing, and provide data support for precision agriculture;

[0040] The present invention obtains biological state data and environmental state data by analyzing biological data and environmental data, utilizes deep learning models, performs efficient training and analysis based on historical data and real-time data, and then performs prediction based on the biological state data and environmental state data to obtain prediction data, which can provide early warning of possible disasters, continuously optimize prevention and control strategies, and improve the accuracy of disaster prevention and mitigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a functional block diagram of the system of the present invention;

[0042] Figure 2 It is the insect pest data variation curve diagram of the present invention;

[0043] Figure 3 The figure is a flow chart of the steps of the method of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Example 1, please refer to Figure 1 As shown, the present application provides a camellia planting intelligent disaster prevention and control system based on deep learning, including a prevention and control collection module, a prevention and control analysis module, a prediction module and an early warning and disposal module;

[0046] The prevention and control collection module is used to collect biological and environmental data in camellia planting areas;

[0047] The prevention and control collection module includes a biological collection unit, which is configured with a biological collection strategy. The biological collection strategy includes setting an automatic insect situation lamp in the camellia planting area, and the automatic insect situation lamp is provided with a built-in sensor, which is used to count the insects entering the automatic insect situation lamp, and the number of insects entering the automatic insect situation lamp is obtained once at a first time interval, and is set as regional insect data; in the specific implementation process, the first time interval is set to 24 hours, and the activity time of different insects is different, but usually 24 hours is a cycle, so the number of insects is obtained once at a time interval of 24 hours, which can ensure that different types of insects in the area are counted;

[0048] A spore catcher is set in the camellia planting area, and the spore density in the area is obtained from the spore catcher once every first time interval, and is set as regional spore data;

[0049] Several cameras are set up in the camellia planting area. The cameras are used to collect camellia images in different directions. The camellia images in the area are obtained once every second time interval, which are set as camellia image data. In the specific implementation process, the second time interval is set to 6 hours. The camellia images collected under different light conditions will be different. If the collection interval is too long, the moment of clear imaging may be missed. If the collection interval is too short, the amount of data is too large, which will lead to excessive calculation.

[0050] Regional insect data, regional spore data, and camellia image data are collectively referred to as biological data.

[0051] The prevention and control collection module also includes an environment collection unit, which is configured with an environment collection strategy, which includes setting up soil sensors in the camellia planting area, the soil sensors are used to monitor the soil moisture content and soil temperature in the soil and transmit data in real time, and record them as real-time soil data;

[0052] Intelligent meteorological instruments are installed in camellia planting areas to monitor regional temperature and humidity in real time and record them as real-time meteorological data. In the specific implementation process, soil temperature and humidity have a greater impact on insect pests and diseases, and soil temperature and humidity are affected by the weather to a certain extent, so it is necessary to monitor meteorological data and soil data in real time.

[0053] Real-time soil data and real-time meteorological data are collectively referred to as environmental data.

[0054] The prevention and control analysis module is used to analyze biological data and environmental data to obtain biological status data and environmental status data;

[0055] See also Figure 2 As shown, the prevention and control analysis module includes a pest analysis unit, which is configured with a pest analysis strategy. The pest analysis strategy includes obtaining regional insect data and regional spore data within a continuous period of time to draw a pest data change curve chart, and obtaining a pest data change rate chart from the pest data change curve chart;

[0056] Obtain a pest data change rate graph within the third period before the pest outbreak from historical data, set it as a critical point change rate graph, perform feature extraction on the critical point change rate graph, obtain the average rate of the critical point change rate graph, and set it as a critical value;

[0057] The average rate of the pest data change rate graph is obtained once every third time period, and is set as the pest change characteristic value. In the specific implementation process, the third time period is set to 7 days;

[0058] The prevention and control analysis module also includes an image analysis unit, which is configured with an image analysis strategy. The image analysis strategy includes performing contour extraction on camellia image data to obtain a number of camellia leaf contours, performing feature extraction on the image within each camellia leaf contour, and the feature extraction includes using RGB channels to represent each pixel point of the image within the camellia leaf contour to obtain a leaf color distribution map;

[0059] Use RGB channels to represent each pixel of the normal camellia leaf image to obtain a standard color distribution map, and obtain the pixel with the smallest RGB channel and the pixel with the largest RGB channel from the standard color distribution map, and set them as the standard channel interval;

[0060] Pixels whose pixel values ​​in the leaf color distribution map are not in the standard channel interval are recorded as special pixels, and the number of special pixels and RGB channels are set as color features; in the specific implementation process, diseased leaves usually have spots that are obviously different from normal leaves. For example, in the early stage of the disease of camellia trees suffering from tea cake disease, small light yellow or light red translucent spots appear on the front of the leaves. Therefore, judging the color of camellia leaves can quickly screen out diseased camellia leaves;

[0061] Establish a disease feature database, which is used to store the color features corresponding to camellia leaves with different disease conditions. In the specific implementation process, the color distribution of different disease conditions varies greatly, so the disease condition can be judged by the distribution color performance; for example, the leaves of camellia trees with red leaf blight will have dark green spots in the early stage, while the leaves of camellia trees with rust will have needle-shaped gray-green dots in the early stage;

[0062] The prevention and control analysis module also includes an environmental analysis unit, which is equipped with an environmental analysis strategy. The environmental analysis strategy includes obtaining soil moisture content and regional humidity, and substituting them into the regional humidity influence formula. In the calculation, the regional humidity impact value is obtained, where A1 is the regional humidity impact value, H0 is the soil moisture content, and S0 is the regional humidity. In the specific implementation process, for example, at a certain moment, the soil humidity is 60%, and the regional humidity is 72%, then the regional temperature impact value is obtained to be 1.2;

[0063] Get the soil temperature and regional temperature and substitute them into the regional temperature influence formula The regional temperature impact value is calculated, where A2 is the regional temperature impact value, T t is the soil temperature, T q is the regional temperature; in the specific implementation process, for example, at a certain moment, the soil temperature obtained is 25°C and the regional temperature is 30°C, then the regional temperature influence value is obtained to be 1.2;

[0064] Obtain regional temperature impact values ​​and regional humidity impact values ​​in real time, and draw regional impact curves;

[0065] The regional impact curve is segmented to obtain the time period type.

[0066] The section where both the temperature impact value and the humidity impact value are greater than 1 is set as a high temperature and humid period. In the specific implementation process, when both the temperature impact value and the humidity impact value are greater than 1, the regional temperature is higher than the soil temperature, and the regional humidity is higher than the soil humidity;

[0067] The section where both the temperature impact value and the humidity impact value are less than 1 is set as the low-temperature drying period. In the specific implementation process, when both the temperature impact value and the humidity impact value are less than 1, the regional temperature is lower than the soil temperature, and the regional humidity is lower than the soil humidity;

[0068] The section where the temperature impact value is less than 1 and the humidity impact value is greater than or equal to 1 is set as a low temperature and humid period. In the specific implementation process, when the temperature impact value is less than 1 and the humidity impact value is greater than or equal to 1, the regional temperature is lower than the soil temperature, and the regional humidity is higher than or equal to the soil humidity;

[0069] The section where the temperature impact value is greater than 1 and the humidity impact value is less than or equal to 1 is set as a high-temperature drying period. In the specific implementation process, when the temperature impact value is less than 1 and the humidity impact value is greater than or equal to 1, the regional temperature is higher than or equal to the soil temperature, and the regional humidity is less than the soil humidity;

[0070] The prediction module performs prediction based on the biological state data and the environmental state data to obtain prediction data;

[0071] The prediction module is equipped with a prediction strategy, which includes monitoring the characteristic value of pest changes. When the characteristic value of pest changes reaches a critical point, the symptom corresponding to the color characteristics of camellia leaves is obtained from the symptom characteristic database, and the appropriate time period type of the symptom is obtained. In the specific implementation process, the development of the symptom is greatly affected by the environmental state, so the outbreak time of the symptom can be predicted by the environmental state at this time.

[0072] When the time period type at this time is an appropriate time period type for the symptom, the average duration of the time period type at this time is obtained and set as the predicted high incidence duration;

[0073] The early warning and disposal module carries out prevention and control measures in camellia-growing areas based on the predicted data.

[0074] The early warning and handling module is configured with an early warning and handling strategy, which includes obtaining and activating a preset response method for any symptom from a database when a predicted high-incidence duration of the symptom is obtained.

[0075] Example 2, please refer to Figure 3As shown, the present invention provides a method for intelligent disaster prevention and control of camellia planting based on deep learning, comprising the following steps:

[0076] Step S1: Collecting biological data and environmental data in the camellia planting area, including the following sub-steps:

[0077] Step S101: An automatic insect light is set in the camellia planting area. The automatic insect light is provided with a built-in sensor, and the built-in sensor is used to count the insects entering the automatic insect light. The number of insects entering the automatic insect light is obtained once every first time interval and set as regional insect data;

[0078] Step S102: a spore trap is set in the camellia planting area, and the spore density in the area is obtained from the spore trap once every first time interval, and is set as regional spore data;

[0079] Step S103: several cameras are set up in the camellia planting area, and the cameras are used to collect camellia images in different directions. The camellia image in the area is acquired once every second time period and set as camellia image data. The regional insect data, regional spore data and camellia image data are collectively referred to as biological data.

[0080] Step S104: a soil sensor is installed in the camellia planting area, the soil sensor is used to monitor the soil moisture content and soil temperature in the soil and transmit the data in real time, and record it as real-time soil data;

[0081] Step S105: an intelligent meteorological instrument is set in the camellia planting area, and the intelligent meteorological instrument is used to monitor the regional temperature and regional humidity in real time and record them as real-time meteorological data. The real-time soil data and the real-time meteorological data are collectively referred to as environmental data;

[0082] Step S2: Analyze the biological data and environmental data to obtain biological state data and environmental state data, including the following sub-steps:

[0083] Step S201: obtaining regional insect data and regional spore data within a continuous time period to draw a pest data change curve graph, and obtaining a pest data change rate graph from the pest data change curve graph;

[0084] Step S202: obtaining a pest data change rate graph within the third period before the pest outbreak from historical data, setting it as a critical point change rate graph, performing feature extraction on the critical point change rate graph, obtaining an average rate of the critical point change rate graph, and setting it as a critical value;

[0085] Step S203: obtaining an average rate of the insect pest data change rate graph once every third time period, and setting it as an insect pest change characteristic value.

[0086] Step S204: extracting the contour of the camellia image data to obtain a number of camellia leaf contours, and extracting features from the image within each camellia leaf contour;

[0087] Feature extraction includes using RGB channels to represent each pixel of the image within the outline of the camellia leaf to obtain a leaf color distribution map;

[0088] Use RGB channels to represent each pixel of the normal camellia leaf image to obtain a standard color distribution map, and obtain the pixel with the smallest RGB channel and the pixel with the largest RGB channel from the standard color distribution map, and set them as the standard channel interval;

[0089] The pixels whose pixel values ​​in the leaf color distribution map are not in the standard channel interval are recorded as special pixels, and the number of special pixels and RGB channels are set as color features;

[0090] Step S205: establishing a pathological characteristic database, the pathological characteristic database is used to store the color characteristics corresponding to camellia leaves with different pathological conditions;

[0091] Step S206: Obtain soil moisture content and regional humidity, and substitute into the regional humidity influence formula In the calculation, the regional humidity impact value is obtained, where A1 is the regional humidity impact value, H0 is the soil moisture content, and S0 is the regional humidity;

[0092] Step S207: Obtain soil temperature and regional temperature, and substitute into the regional temperature influence formula The regional temperature impact value is calculated, where A2 is the regional temperature impact value, T t is the soil temperature, T q is the regional temperature;

[0093] Step S207: acquiring the regional temperature impact value and the regional humidity impact value in real time, and drawing a regional impact curve diagram;

[0094] Step S208: segment the regional impact curve to obtain the time period type, and set the section where the temperature impact value and humidity impact value are both greater than 1 as the high temperature and humidity period.

[0095] The section where both the temperature impact value and the humidity impact value are less than 1 is set as the low-temperature drying period.

[0096] Set the section where the temperature impact value is less than 1 and the humidity impact value is greater than or equal to 1 as the low temperature and humid period.

[0097] The section where the temperature impact value is greater than 1 and the humidity impact value is less than or equal to 1 is set as a high-temperature drying period;

[0098] Step S3: Predicting based on the biological state data and the environmental state data to obtain predicted data, including the following sub-steps:

[0099] Step S301: monitoring the pest change characteristic value, and before the pest change characteristic value reaches the critical point, obtaining the disease corresponding to the color characteristic of the camellia leaf from the disease characteristic database, and obtaining the appropriate time period type of the disease;

[0100] Step S302: when the time period type at this time is an appropriate time period type for the symptom, the average duration of the time period type at this time is obtained and set as the predicted high incidence duration;

[0101] Step S4: Performing prevention and control measures on the camellia planting area based on the predicted data, including obtaining and activating a preset response method for any disease condition from a database when a predicted high-incidence duration of the disease condition is obtained.

[0102] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0103] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A camellia planting intelligent disaster prevention and control system based on deep learning, characterized in that: It includes prevention and control collection module, prevention and control analysis module, prediction module and early warning and disposal module; The control and collection module is used to collect biological data and environmental data in the camellia planting area; The prevention and control analysis module is used to analyze biological data and environmental data to obtain biological state data and environmental state data; The prediction module performs prediction based on the biological state data and the environmental state data to obtain prediction data; The early warning and disposal module performs prevention and control measures on the camellia planting area based on the prediction data.

2. The deep learning-based camellia planting intelligent disaster prevention system according to claim 1, characterized in that: The prevention and control collection module includes a biological collection unit, and the biological collection unit is configured with a biological collection strategy. The biological collection strategy includes setting an automatic insect situation lamp in the camellia planting area, and setting a built-in sensor in the automatic insect situation lamp. The built-in sensor is used to count the insects entering the automatic insect situation lamp, and the number of insects entering the automatic insect situation lamp is obtained once every first time interval, and set as regional insect data; A spore catcher is set in the camellia planting area, and the spore density in the area is obtained from the spore catcher once every first time interval, and is set as regional spore data; A plurality of camera devices are arranged in the camellia planting area, the camera devices are used to collect camellia images in different directions, and a camellia image in the area is obtained once at a second time interval, and is set as camellia image data; The regional insect data, regional spore data and camellia image data are collectively referred to as biological data.

3. The deep learning-based camellia planting intelligent disaster prevention and control system according to claim 2, characterized in that: The prevention and control collection module also includes an environment collection unit, the environment collection unit is configured with an environment collection strategy, the environment collection strategy includes setting a soil sensor in the camellia planting area, the soil sensor is used to monitor the soil moisture content and soil temperature in the soil and transmit the data in real time, and record it as real-time soil data; An intelligent meteorological instrument is installed in the camellia planting area, and the intelligent meteorological instrument is used to monitor the regional temperature and regional humidity in real time and record them as real-time meteorological data; The real-time soil data and real-time meteorological data are collectively referred to as environmental data.

4. The deep learning-based camellia planting intelligent disaster prevention and control system according to claim 3, characterized in that: The control analysis module includes a pest analysis unit, and the pest analysis unit is configured with a pest analysis strategy, and the pest analysis strategy includes obtaining regional insect data and regional spore data within a continuous period of time to draw a pest data change curve chart, and obtaining a pest data change rate chart from the pest data change curve chart; Obtain a pest data change rate graph within the third period before the pest outbreak from historical data, set it as a critical point change rate graph, perform feature extraction on the critical point change rate graph, obtain the average rate of the critical point change rate graph, and set it as a critical value; The average rate of the insect pest data change rate graph is obtained once every third time period and is set as the insect pest change characteristic value.

5. The deep learning-based camellia planting intelligent disaster prevention and control system according to claim 4, characterized in that: The prevention and control analysis module also includes an image analysis unit, which is configured with an image analysis strategy, which includes performing contour extraction on camellia image data to obtain a number of camellia leaf contours, performing feature extraction on the image within each camellia leaf contour, and the feature extraction includes using RGB channels to represent each pixel point of the image within the camellia leaf contour to obtain a leaf color distribution map; Use RGB channels to represent each pixel of the normal camellia leaf image to obtain a standard color distribution map, and obtain the pixel with the smallest RGB channel and the pixel with the largest RGB channel from the standard color distribution map, and set them as the standard channel interval; The pixels whose pixel values ​​in the leaf color distribution map are not in the standard channel interval are recorded as special pixels, and the number of special pixels and RGB channels are set as color features; A pathological characteristic database is established, wherein the pathological characteristic database is used to store color characteristics corresponding to camellia leaves with different pathological conditions.

6. The deep learning-based camellia planting intelligent disaster prevention and control system according to claim 5, characterized in that: The prevention and control analysis module also includes an environmental analysis unit, which is configured with an environmental analysis strategy, which includes obtaining soil moisture content and regional humidity, and substituting them into the regional humidity influence formula. In the calculation, the regional humidity impact value is obtained, where A1 is the regional humidity impact value, H0 is the soil moisture content, and S0 is the regional humidity; Get the soil temperature and regional temperature and substitute them into the regional temperature influence formula The regional temperature impact value is calculated, where A2 is the regional temperature impact value, T t is the soil temperature, T q is the regional temperature; Obtain regional temperature impact values ​​and regional humidity impact values ​​in real time, and draw regional impact curves; The regional impact curve is segmented to obtain the time period type, and the section where the temperature impact value and humidity impact value are both greater than 1 is set as the high temperature and humidity period. The section where both the temperature impact value and the humidity impact value are less than 1 is set as the low-temperature drying period. Set the section where the temperature impact value is less than 1 and the humidity impact value is greater than or equal to 1 as the low temperature and humid period. The section where the temperature influence value is greater than 1 and the humidity influence value is less than or equal to 1 is set as the high-temperature drying period.

7. The deep learning-based camellia planting intelligent disaster prevention and control system according to claim 6, characterized in that: The prediction module is configured with a prediction strategy, which includes monitoring the pest change characteristic value, obtaining the symptom corresponding to the color characteristic of the camellia leaves from the symptom characteristic database before the pest change characteristic value reaches the critical point, and obtaining the appropriate time period type of the symptom; When the time period type at this time is an appropriate time period type for the symptom, the average duration of the time period type at this time is obtained and set as the predicted high incidence duration.

8. The deep learning-based camellia planting intelligent disaster prevention and control system according to claim 7, characterized in that: The early warning and handling module is configured with an early warning and handling strategy, which includes obtaining and activating a preset coping method for any symptom from a database when a predicted high-incidence duration of the symptom is obtained.

9. A method for intelligent disaster prevention and control of camellia planting based on deep learning, applicable to a system for intelligent disaster prevention and control of camellia planting based on deep learning according to any one of claims 1 to 8, characterized in that: The steps include: Collect biological and environmental data in camellia growing areas; Analyze the biological data and the environmental data to obtain biological state data and environmental state data; Perform prediction based on biological state data and environmental state data to obtain prediction data; Carry out prevention and control measures in camellia growing areas based on predicted data.

Images