Tea disease and insect pest recognition and early warning method and system based on unmanned aerial vehicle

Abstract

The invention discloses an unmanned aerial vehicle-based tea disease and insect pest recognition and early warning method and system, and belongs to the technical field of disease and insect pest recognition and early warning, and the method comprises an information obtaining module which obtains the image information of tea and the reflectivity information of the tea; the information analyzing and processing module is used for processing and analyzing the acquired information to judge whether the tea leaves have diseases and insect pests; the early warning module is used for carrying out early warning analysis on pest and disease damage conditions in the tea garden according to the acquired information; and the alarm module is used for carrying out corresponding instruction reminding according to the information analysis result. Related information of the tea leaves in the tea garden is collected through the unmanned aerial vehicle, corresponding processing and analysis are carried out to judge whether the tea leaves have pest and disease damage conditions or not, corresponding alarm reminding is carried out, the pest and disease damage conditions of the tea leaves can be obtained without manual inspection, the detection result is accurate and efficient, and the pest and disease damage conditions of the tea leaves can be effectively mastered; and the tea growth is ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of pest and disease identification and early warning, and in particular relates to a method and system for identifying and warning tea pests and diseases based on an unmanned aerial vehicle. Background Art

[0002] The outbreak of pests and diseases will cause abnormal growth of tea trees, thus affecting tea production and quality, and reducing the economic benefits of tea farmers. Therefore, the identification and early warning of tea pests and diseases can effectively deal with the pests and diseases, thereby ensuring the normal growth of tea trees and ensuring the quality and yield of tea.

[0003] Currently, the identification of tea plant pests and diseases is mostly carried out by manual detection. The types and quantities of pests and diseases on the tea trees are manually observed and recorded to determine the pest and disease situation. Although this method can obtain the pest and disease situation of the tea trees more accurately, it requires a lot of time and human resources to execute, resulting in low efficiency. In addition, manual observation and recording are easily affected by subjective factors, resulting in inaccurate final detection results. Summary of the invention

[0004] The purpose of the present invention is to provide a tea pest and disease identification and early warning method and system based on drones to solve the problems faced by the above-mentioned background technology.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A tea pest and disease identification and early warning method based on drones, the method comprising:

[0007] S1. Use a high-resolution drone to fly over the tea garden to take a large number of tea images at different time periods. The drone is equipped with a spectral sensor to obtain the reflectivity information of the tea leaves.

[0008] S2. Preprocess the collected images by image processing technology to obtain precision images and obtain pest and disease information on the precision images;

[0009] S3. Establish a pest and disease database, use a convolutional neural network to classify the extracted pest and disease information, and identify the number and type of pests and diseases in the precision image;

[0010] S4, using image processing technology to extract characteristic parameters of tea leaves in the precision image, combining the acquired reflectivity information to obtain a health status value of the tea leaves, and judging whether the tea leaves have pests and diseases according to the health status value;

[0011] S5. When pests and diseases exist, the severity of the pests and diseases is determined according to the health status value, and an alarm reminder is issued to a corresponding degree;

[0012] S6. When there are no pests and diseases, an early warning analysis of the pest and disease situation in the tea garden is carried out based on the information obtained. When the tea leaves show abnormal health conditions, an early warning reminder is issued.

[0013] Furthermore, the characteristic parameters include color characteristics, texture characteristics and shape characteristics of the tea leaves, and the reflectivity information includes the reflectivity of the tea leaves in a near-infrared band and a green band.

[0014] Furthermore, the method for determining whether the tea leaves are susceptible to pests and diseases in step S4 is as follows:

[0015] According to the acquired color features, texture features and shape features, the corresponding color feature score value B is obtained. c , Texture feature score B t and the shape feature score B s ;

[0016] So through the formula The health status value Hsv of tea leaves is obtained;

[0017] When Hsv<Hsv0, it is judged that the tea leaves are in danger of pests and diseases;

[0018] Among them, Hsv0 is the preset health status judgment threshold, μ1, μ2 and μ3 are proportional coefficients, R g is the reflectance of the green band, R r is the reflectivity in the near-infrared band.

[0019] Furthermore, the color feature score value B c , Texture feature score B t and the shape feature score B s The method to obtain is:

[0020] First, we build tea color feature scoring libraries, texture feature scoring libraries, and shape feature scoring libraries. Each scoring library has multiple scoring values, and each scoring value has enough feature information. The lower the score value in each scoring library, the better the tea condition.

[0021] The color features, texture features, and shape features in the acquired precision image are input into the corresponding scoring library to obtain the corresponding color feature score value B c , Texture feature score B t and the shape feature score B s ;

[0022] Furthermore, the method for determining the severity of pests and diseases in step S5 is:

[0023] The pest and disease conditions are divided into four levels: level one, level two, level three and level four. The higher the level, the more serious the pest and disease.

[0024] Compare the obtained health status value Hsv with each preset threshold interval;

[0025] When Hsv∈[Hsv z , Hsv0], the severity of the pests and diseases is judged to be level one, and a level one alarm instruction is generated;

[0026] When Hsv∈(Hsv y , Hsv z ), the severity of the pest is judged to be level 2, and a level 2 alarm instruction is generated;

[0027] When Hsv∈[Hsv x , Hsv y ], the severity of the pests and diseases is judged to be level three, and a level three alarm instruction is generated;

[0028] When Hsv∈(0, Hsv x ), the severity of the pest is judged to be level 4, and a level 4 alarm instruction is generated;

[0029] Among them, Hsv x 、Hsv y 、Hsv z The judgment threshold is set.

[0030] Furthermore, the method for performing early warning analysis on the pest and disease situation in the tea garden according to the acquired information in step S6 is:

[0031] When it is determined that there is no pest or disease, the health status value within n days is obtained, so as to formulate a curve f(x, Hsv) of the change of health status value with the number of days;

[0032] At the same time, we can obtain the types of pests and diseases detected every day, as well as the number of each pest and disease PC, through the formula Obtain the daily pest damage index V, and then formulate the pest damage index change curve f(x, V) with the number of days;

[0033] By formula The health outlier value DS is obtained;

[0034] When DS>DS0, it is judged that the tea leaves are in abnormal health condition, and an early warning instruction is generated;

[0035] Among them, j∈[1,m], PC j is the number of pests and diseases of the jth pest type, τ j is the weight coefficient of the jth pest type, x1 is the first day of detection, xn It is the last day of testing. is the health status value obtained on the first day, The health status value obtained on the last day, max V is the maximum pest damage index obtained, min V is the minimum pest damage index obtained, α1 and α2 are preset coefficients, and DS0 is the set health abnormality judgment threshold.

[0036] A tea pest and disease identification and early warning system based on drones, the system is implemented by the tea pest and disease identification and early warning method based on drones, and the early warning system comprises:

[0037] An information acquisition module, the information acquisition module is used to acquire image information and reflectivity information of the tea leaves;

[0038] An information transmission module, which transmits the acquired image information and reflectivity information to the information analysis and processing module through the Internet of Things technology;

[0039] An information analysis and processing module, which processes and analyzes the acquired information to determine whether the tea leaves are susceptible to pests and diseases;

[0040] An early warning module, when it is determined that there is no pest or disease condition, the early warning module performs early warning analysis on the pest and disease condition in the tea garden according to the acquired information;

[0041] The alarm module issues corresponding instruction reminders according to the analysis results of the information analysis and processing module and the early warning module.

[0042] Beneficial effects of the present invention:

[0043] The present invention collects relevant information of tea leaves in a tea garden through a drone, and performs corresponding processing and analysis to determine whether the tea leaves have diseases and insect pests, thereby making corresponding alarm reminders. The invention can obtain the insect pest situation of the tea leaves without manual inspection, and the detection result is accurate and efficient, and the insect pest situation of the tea leaves can be effectively grasped to ensure the growth of the tea leaves.

[0044] The present invention can conduct a comprehensive analysis based on the reflectivity of tea leaves in the near-infrared band and the green band as well as the color, shape, and texture characteristics of the tea leaves to determine whether the tea leaves have pests and diseases, thereby reducing detection errors and making the detection results more accurate.

[0045] When the present invention detects that there are no pests and diseases, it can also perform early warning analysis on the pest and disease status of tea leaves to determine whether the tea leaves have potential pests and diseases, thereby intervening in time, reducing the occurrence of pests and diseases, and ensuring the normal growth of tea leaves.

[0046] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0048] Figure 1 is a flow chart of the method of the present invention;

[0049] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0050] 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.

[0051] In one embodiment, a method for identifying and warning tea pests based on a drone is disclosed. Figure 1 As shown, the early warning method includes the following steps:

[0052] S1. Use a high-resolution drone to fly over the tea garden to take a large number of tea images at different time periods. The drone is equipped with a spectral sensor to obtain the reflectivity information of the tea leaves.

[0053] S2. Preprocess the collected images by image processing technology to obtain precision images and obtain pest and disease information on the precision images;

[0054] S3. Establish a pest database, use convolutional neural network to classify the extracted pest information, and identify the number and type of pests in the precision image;

[0055] S4, using image processing technology to extract characteristic parameters of tea leaves in the precision image, combining the acquired reflectivity information to obtain a health status value of the tea leaves, and judging whether the tea leaves have pests and diseases according to the health status value;

[0056] S5. When pests and diseases exist, the severity of the pests and diseases is determined according to the health status value, and an alarm reminder is issued to a corresponding degree;

[0057] S6. When there are no pests and diseases, an early warning analysis of the pest and disease situation in the tea garden is carried out based on the information obtained. When the tea leaves show abnormal health conditions, an early warning reminder is issued.

[0058] Through the above technical scheme, the present application adopts a high-resolution drone to fly over the tea garden, and takes a large number of tea images at different time periods. At the same time, the drone is equipped with a spectral sensor to obtain the reflectivity information of the tea leaves. Then, the collected images are pre-processed by image processing technology, including denoising, enhancement, cropping and other processing, to obtain precision images, and obtain pest information on the precision images; then a pest database is established, and a convolutional neural network is used to classify the extracted pest information, identify the number and type of pests in the precision image, and thus analyze the pest and disease situation of the tea leaves; image processing technology is then used to extract the characteristic parameters of the tea leaves in the precision image, including the color characteristics, texture characteristics and shape characteristics of the tea leaves, etc., and the health status value of the tea leaves is obtained in combination with the obtained reflectivity information (the reflectivity of the tea leaves in the near-infrared band and the green band). It is judged whether the tea leaves have pest and disease conditions according to the health status value. When pest and disease exist, the severity of the pest and disease is determined according to the size of the health status value, and an alarm reminder of the corresponding degree is given. When pest and disease do not exist, an early warning analysis of the pest and disease condition in the tea garden is performed according to the acquired information, and an early warning reminder is given when the tea leaves are analyzed to have an abnormal health condition. By collecting relevant information about tea in the tea garden through drones and conducting corresponding processing and analysis, it is possible to determine whether the tea has any diseases or insect pests, and thus issue corresponding alarm reminders. The pest situation of the tea can be obtained without manual inspections, and the detection results are accurate and efficient, which can effectively grasp the pest situation of the tea to ensure the growth of the tea. At the same time, a comprehensive analysis can be conducted based on the reflectivity of the tea in the near-infrared band and the green band and the characteristics of the tea to determine whether the tea has any diseases or insect pests, and thus take corresponding measures. If it is detected that there are no diseases or insect pests, an early warning analysis of the tea's disease and insect pest situation can be conducted to analyze whether there are potential diseases and insect pests, so as to intervene in time, reduce the occurrence of diseases and insect pests, and ensure the normal growth of tea.

[0059] As an embodiment of the present invention, the method for determining whether the tea leaves are in a pest and disease state in step S4 is:

[0060] First, we build tea color feature scoring libraries, texture feature scoring libraries, and shape feature scoring libraries. Each scoring library has multiple scoring values, and each scoring value has enough feature information. The lower the score value in each scoring library, the better the tea condition.

[0061] Then, the color features, texture features, and shape features in the acquired precision image are input into the corresponding scoring library respectively, so as to obtain the corresponding color feature scoring value Bc , Texture feature score B t and the shape feature score B s ;

[0062] So through the formula The health status value Hsv of tea leaves is obtained;

[0063] When Hsv<Hsv0, it is judged that the tea leaves are in danger of pests and diseases;

[0064] Among them, Hsv0 is the preset health status judgment threshold, μ1, μ2 and μ3 are proportional coefficients, R g is the reflectance of the green band, R r is the reflectivity in the near-infrared band.

[0065] Through the above technical solution, this embodiment provides a specific method for judging whether tea leaves have pests and diseases. When tea leaves have pests and diseases, the tea leaves infected by the pests and diseases may show characteristics such as color change, leaf deformation, and abnormal texture. Therefore, according to the relevant data in the big data, a color feature scoring library, a texture feature scoring library, and a shape feature scoring library of tea leaves are established. Each scoring library has multiple scoring values. The lower the score value in each scoring library, the better the condition of the tea leaves. Each scoring value has enough feature information. Then, the color features, texture features, and shape features in the acquired precision image are respectively input into the corresponding scoring library, so as to obtain the corresponding color feature scoring value B. c , Texture feature score B t and the shape feature score B s , so that through the formula The health status value Hsv of tea leaves is obtained; formula Indicates the spectrum of tea leaves. Generally speaking, healthy tea leaves have a lower reflectivity in the green band and a higher reflectivity in the near-infrared band. Therefore, when the formula The larger the value, the better the health of the tea leaves, and the lower the possibility of pests and diseases. As for color features, texture features, and shape features, when pests and diseases occur, the color of the tea leaves may change, such as from green to yellow or black, and the shape and texture may also be quite different from healthy tea leaves. Therefore, according to the established tea color feature scoring library, texture feature scoring library, and shape feature scoring library, the corresponding color feature scoring value B is obtained. c , Texture feature score B t and the shape feature score B s , when the score is lower, the tea condition is better, so when the formula The larger the value, the better the health of the tea leaves, and the lower the possibility of pests and diseases. Therefore, after obtaining the health value of the tea leaves, it is compared with the preset health judgment threshold Hsv0. When Hsv<Hsv0, it is judged that the tea leaves have pests and diseases. In this way, the drone collection and analysis processing method is used instead of the manual inspection method to check whether the tea leaves have pests and diseases, which can save human resources and improve detection efficiency. At the same time, the reflectivity of the tea leaves in the near-infrared band and the green band, as well as the color, shape, and texture characteristics of the tea leaves are comprehensively analyzed to judge whether the tea leaves have pests and diseases, which can be more accurate.

[0066] In the above technical solution, the preset health status judgment threshold Hsv0, proportional coefficients μ1, μ2 and μ3 can all be determined based on empirical data, which will not be described in detail here.

[0067] As an embodiment of the present invention, the method for determining the severity of pests and diseases in step S5 is:

[0068] The pest and disease conditions are divided into four levels: level one, level two, level three and level four. The higher the level, the more serious the pest and disease.

[0069] Compare the obtained health status value Hsv with each preset threshold interval;

[0070] When Hsv∈[Hsv z , Hsv0], the severity of the pests and diseases is judged to be level one, and a level one alarm instruction is generated;

[0071] When Hsv∈(Hsv y , Hsv z ), the severity of the pest is judged to be level 2, and a level 2 alarm instruction is generated;

[0072] When Hsv∈[Hsv x , Hsv y ], the severity of the pests and diseases is judged to be level three, and a level three alarm instruction is generated;

[0073] When Hsv∈(0, Hsv x ), the severity of the pest is judged to be level 4, and a level 4 alarm instruction is generated;

[0074] Among them, Hsv x 、Hsv y 、Hsv z The judgment threshold is set.

[0075] Through the above technical solution, this embodiment provides a specific method for judging the severity of pests and diseases. When it is judged that there are pests and diseases, the pests and diseases are divided into four levels: level one, level two, level three and level four according to historical experience data. The higher the level, the higher the severity of the pests and diseases. Then, three judgment thresholds Hsv are set according to historical data. x 、Hsv y 、Hsv z , and Hsv x <Hsv y <Hsv z , when Hsv∈Hsv z , Hsv0], the pests and diseases are relatively light, the severity of the pests and diseases is judged to be level one, and a level one alarm instruction is generated. y , Hsv z ), the pest situation is relatively light or slightly serious, then the severity of the pest is judged to be level 2, and a level 2 alarm instruction is generated; when Hsv∈[Hsv x , Hsv y ], the pest situation is serious, the severity of the pest is judged to be level three, and a level three alarm instruction is generated; when Hsv∈(0, Hsv x ), the pests and diseases are very serious, then the severity of the pests and diseases is judged to be level 4, and a level 4 alarm instruction is generated. In this way, the severity of the pests and diseases of the tea can be judged according to the generated health value, which is convenient for the management personnel to understand the condition of the tea in time and take corresponding measures to restore the health of the tea as much as possible and reduce the loss of tea.

[0076] As an embodiment of the present invention, the method for performing early warning analysis on the disease and insect pest status of tea according to the acquired information in step S6 is:

[0077] When it is determined that there is no pest or disease, the health status value within n days is obtained, so as to formulate a curve f(x, Hsv) of the change of health status value with the number of days;

[0078] At the same time, we can obtain the types of pests and diseases detected every day, as well as the number of each pest and disease PC, through the formula Obtain the daily pest damage index V, and then formulate the pest damage index change curve f(x, V) with the number of days;

[0079] By formula The health outlier value DS is obtained;

[0080] When DS>DS0, it is judged that the tea leaves are in abnormal health condition, and an early warning instruction is generated;

[0081] Among them, j∈[1,m], PC j is the number of pests and diseases of the jth pest type, τ j is the weight coefficient of the jth pest type, x1 is the first day of detection, x n It is the last day of testing. is the health status value obtained on the first day, The health status value obtained on the last day, max V is the maximum pest damage index obtained, min V is the minimum pest damage index obtained, α1 and α2 are preset coefficients, and DS0 is the set health abnormality judgment threshold.

[0082] Through the above technical solution, this embodiment provides a method for early warning analysis of the pest status of tea. When it is judged that there are no pests, although the pest situation is not displayed at this time, the occurrence of pests is gradually expanding from small to large. When pests are found, the tea has been contaminated by large pests. Therefore, when it is judged that there are no pests, this embodiment obtains the health status value within n days, so as to formulate the curve f(x, Hsv) of the change of health status value with the number of days, and at the same time obtains the type of pests m detected every day and the number PC of each pest, and through the formula The daily pest and disease damage index V is obtained, and the pest and disease damage index variation curve f(x, V) is proposed. It can be seen that the more types of pests and diseases appear in tea, the greater the risk of pests and diseases. Therefore, through the formula The health abnormal value DS is obtained; formula It indicates the change of tea health value within the testing days. If the value is larger, it means the tea health is better. If the value is smaller, it means the tea health is declining. The formula It indicates the difference between the health status value on the first day and the health status value on the last day. Generally speaking, the health status of the two is the same or the health status detected later is better than the health status detected earlier. Therefore, the larger the value, the worse the health status. The same formula It indicates the change of tea pest damage index within the detection days. The larger the value, the worse the health of tea. V -min V It indicates the difference between the maximum pest damage index and the minimum pest damage index within the detection days. If the difference is larger, it means that the trend of pests and diseases is increasing within the detection days. Therefore, the formula A comprehensive analysis is performed to obtain a health abnormality value DS, and the obtained health abnormality value DS is compared with the set health abnormality judgment threshold DS0. When DS>DS0, it is judged that the tea has an abnormal health condition, and an early warning instruction is generated. In this way, a comprehensive analysis can be performed based on the changes in the tea's pest damage index and the health status value in consecutive days to determine whether the tea has potential pests and diseases. Once a potential pest and disease trend is found, a corresponding early warning reminder is generated to remind management personnel to intervene in time to reduce the occurrence of pests and diseases and ensure the normal growth of tea.

[0083] In the above technical solution, the preset coefficients α1 and α2 and the set health abnormality judgment threshold DS0 can be determined based on empirical data, while the set length of n days can be artificially determined and the duration will not be described in detail here.

[0084] In one embodiment, a tea pest identification and early warning system based on drones is disclosed. Figure 2 As shown, the early warning system includes:

[0085] An information acquisition module, which is used to acquire image information and reflectivity information of tea leaves;

[0086] An information transmission module, which transmits the acquired image information and reflectivity information to the information analysis and processing module through the Internet of Things technology;

[0087] The information analysis and processing module processes and analyzes the acquired information to determine whether the tea leaves are susceptible to pests and diseases;

[0088] Early warning module: when it is determined that there is no pest or disease condition, the early warning module will conduct early warning analysis on the pest and disease condition in the tea garden based on the acquired information;

[0089] The alarm module issues corresponding instruction reminders based on the analysis results of the information analysis and processing module and the early warning module.

[0090] It is only an example and explanation of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A tea pest and disease identification and early warning method based on drones, characterized in that: The method comprises: S1. Use a high-resolution drone to fly over the tea garden to take a large number of tea images at different time periods. The drone is equipped with a spectral sensor to obtain the reflectivity information of the tea leaves. S2. Preprocess the collected images by image processing technology to obtain precision images and obtain pest and disease information on the precision images; S3. Establish a pest and disease database, use a convolutional neural network to classify the extracted pest and disease information, and identify the number and type of pests and diseases in the precision image; S4, using image processing technology to extract characteristic parameters of tea leaves in the precision image, combining the acquired reflectivity information to obtain a health status value of the tea leaves, and judging whether the tea leaves have pests and diseases according to the health status value; S5. When pests and diseases exist, the severity of the pests and diseases is determined according to the health status value, and an alarm reminder is issued to a corresponding degree; S6. When there are no pests and diseases, an early warning analysis of the pest and disease situation in the tea garden is carried out based on the information obtained. When the tea leaves show abnormal health conditions, an early warning reminder is issued.

2. The method for identifying and warning tea pests based on drones according to claim 1, characterized in that: The characteristic parameters include color characteristics, texture characteristics and shape characteristics of the tea leaves, and the reflectivity information includes the reflectivity of the tea leaves in the near-infrared band and the green band.

3. The method for identifying and warning tea pests based on drones according to claim 2, characterized in that: The method for determining whether the tea leaves are in a pest- and disease-prone state in step S4 is: According to the acquired color features, texture features and shape features, the corresponding color feature score value B is obtained. c , Texture feature score B t and the shape feature score B s ; So through the formula The health status value Hsv of tea leaves is obtained; When Hsv<Hsv0, it is judged that the tea leaves are in danger of pests and diseases; Among them, Hsv0 is the preset health status judgment threshold, μ1, μ2 and μ3 are proportional coefficients, R g is the reflectance of the green band, R r is the reflectivity in the near-infrared band.

4. The method for identifying and warning tea pests based on drones according to claim 3, characterized in that: The color feature score value B c , Texture feature score B t and the shape feature score B s The method to obtain is: First, we build tea color feature scoring libraries, texture feature scoring libraries, and shape feature scoring libraries. Each scoring library has multiple scoring values, and each scoring value has enough feature information. The lower the score value in each scoring library, the better the tea condition. The color features, texture features, and shape features in the acquired precision image are input into the corresponding scoring library to obtain the corresponding color feature score value B c , Texture feature score B t and the shape feature score B s。 5. The method for identifying and warning tea pests based on drones according to claim 3, characterized in that: The method for determining the severity of pests and diseases in step S5 is: The pest and disease conditions are divided into four levels: level one, level two, level three and level four. The higher the level, the more serious the pest and disease. Compare the obtained health status value Hsv with each preset threshold interval; When Hsv∈[Hsv z , Hsv0], the severity of the pests and diseases is judged to be level one, and a level one alarm instruction is generated; When Hsv∈(Hsv y , Hsv z ), the severity of the pest is judged to be level 2, and a level 2 alarm instruction is generated; When Hsv∈[Hsv x , Hsv y ], the severity of the pests and diseases is judged to be level three, and a level three alarm instruction is generated; When Hsv∈(0, Hsv x ), the severity of the pest is judged to be level 4, and a level 4 alarm instruction is generated; Among them, Hsv x 、Hsv y 、Hsv z The judgment threshold is set.

6. The method for identifying and warning tea pests based on drones according to claim 3, characterized in that: The method for performing early warning analysis on the pest and disease situation in the tea garden according to the acquired information in step S6 is: When it is determined that there is no pest or disease, the health status value within n days is obtained, so as to formulate a curve f(x, Hsv) of the change of health status value with the number of days; At the same time, we can obtain the types of pests and diseases detected every day, as well as the number of each pest and disease PC, through the formula Obtain the daily pest damage index V, and then formulate the pest damage index change curve f(x, V) with the number of days; By formula The health outlier value DS is obtained; When DS>DS0, it is judged that the tea leaves are in abnormal health condition, and an early warning instruction is generated; Among them, j∈[1,m], PC j is the number of pests and diseases of the jth pest type, τ j is the weight coefficient of the jth pest type, x1 is the first day of detection, x n It is the last day of testing. is the health status value obtained on the first day, The health status value obtained on the last day, max V is the maximum pest damage index obtained, min V is the minimum pest damage index obtained, α1 and α2 are preset coefficients, and DS0 is the set health abnormality judgment threshold.

7. A tea pest and disease identification and early warning system based on drones, the system being implemented by the tea pest and disease identification and early warning method based on drones according to any one of claims 1 to 6, characterized in that: The early warning system includes: An information acquisition module, the information acquisition module is used to acquire image information and reflectivity information of tea leaves; An information transmission module, which transmits the acquired image information and reflectivity information to the information analysis and processing module through the Internet of Things technology; An information analysis and processing module, which processes and analyzes the acquired information to determine whether the tea leaves are susceptible to pests and diseases; An early warning module, when it is determined that there is no pest or disease condition, the early warning module performs early warning analysis on the pest and disease condition in the tea garden according to the acquired information; The alarm module issues corresponding instruction reminders according to the analysis results of the information analysis and processing module and the early warning module.

Images