A rapid identification method for jujube tree diseases and insect pests

Abstract

The present invention relates to a rapid identification method for jujube pests and diseases, and the rapid identification method for jujube pests and diseases comprises the following steps: drawing a jujube forest outline map, evenly setting sampling points on the jujube forest outline map, and the jujube trees in the sampling points are sample jujube trees; establishing a pest and disease identification model and a control database; regularly taking photos and sampling the sample jujube trees in all the sampling points to obtain sample photos; pre-processing the sample photos and inputting them into the pest and disease identification model for identification, outputting the plant's pest and disease information, and the pest and disease information is simultaneously input into the control database to output the control information. The present invention evenly sets sampling points in the jujube forest, and the sampling in the sampling points represents the pest and disease status of the jujube trees in the entire independent area, thereby avoiding the problem of difficulty in sampling a large area of ​​jujube forests; and by establishing a pest and disease model to identify the photographed pest and disease photos, the accuracy is higher and the speed is faster.

Description

Technical Field

[0001] The invention belongs to the technical field of agriculture, and in particular relates to a method for quickly identifying diseases and insect pests of jujube trees. Background Art

[0002] Jujube is one of the oldest varieties of fruit tree cultivation in my country and is also an important endemic fruit tree in my country. Jujube has the advantages of strong adaptability, early fruiting, fast income, long life, easy management, and cold and drought resistance, as well as tolerance to barrenness. Jujube fruit is rich in nutrition, has a strong tonic effect, and has high commercial value. Jujube also has the functions of preventing wind, fixing sand, reducing wind speed, regulating temperature, preventing and reducing the harm of dry and hot winds, and is an ideal farmland forest network tree species and forest-grain intercropping tree species. Jujube often suffers from serious production reduction due to pests and diseases. Therefore, pest and disease control is an important task in the cultivation and management of jujube. The first step in pest and disease control is to quickly identify pests and diseases.

[0003] At present, the management of pests and diseases in large jujube orchards is very difficult. Firstly, due to the large area of ​​jujube plantations and the large number of jujube trees, pests and diseases cannot be discovered in time, that is, the sampling mechanism has big problems; secondly, the identification of jujube tree pests and diseases mostly relies on human visual observation and judgment based on experience, which requires a high level of work experience for the staff, and has large errors, and is prone to misjudgment.

[0004] Therefore, it is necessary to provide a new method for rapid identification of diseases and insect pests of jujube trees to solve the above technical problems. Summary of the invention

[0005] The purpose of the present invention is to provide a method for rapid identification of diseases and insect pests of jujube trees in order to solve the above-mentioned problems.

[0006] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0007] A rapid identification method for jujube tree diseases and insect pests comprises the following steps:

[0008] S1: draw a jujube forest outline map, and evenly set sampling points on the jujube forest outline map, and the jujube trees in the sampling points are sample jujube trees;

[0009] S2: Establish pest and disease identification model and control database;

[0010] S3: regularly taking photos of the sample jujube trees in all the sampling points to obtain sample photos;

[0011] S4: pre-processing the sample photos and inputting them into a pest and disease recognition model for recognition, outputting plant pest and disease information, and simultaneously inputting the pest and disease information into a prevention and control database, outputting prevention and control information.

[0012] As a further optimization scheme of the present invention, the step S1 specifically includes the following steps:

[0013] Taking aerial photos of the jujube tree plantation to obtain aerial photos;

[0014] Describing the edge line of the jujube plantation in the aerial photograph to obtain the jujube forest outline map;

[0015] To estimate the area of ​​the date palm plantations;

[0016] Selecting grid coordinates of appropriate size according to the estimated area of ​​the jujube tree plantation;

[0017] Put the jujube forest outline map into the grid coordinates;

[0018] Select all the complete small squares in the jujube forest outline map as an independent area;

[0019] Draw a circle with a suitable radius at the center of the independent area, the inside of the circle is the sampling area, and the outside of the circle is the edge area;

[0020] The jujube trees within the sampling area are taken as sample jujube trees.

[0021] As a further optimization scheme of the present invention, the area of ​​the date forest outline map is not less than the sum of the areas of the four small squares.

[0022] As a further optimization scheme of the present invention, the area of ​​the small square located in the jujube forest outline map is greater than or equal to 50% of its own area, and the small square can be called a complete small square and can be used as an independent area.

[0023] As a further optimization scheme of the present invention, there are no less than 5 jujube trees in the sampling area.

[0024] As a further optimization scheme of the present invention, the step S2 specifically includes the following steps:

[0025] Establishing the pest and disease identification model: pre-training a feature extraction network using a preset source data set, connecting the pre-trained feature extraction network and a preset prediction network to obtain a preliminary pest and disease identification model;

[0026] Using a target data set to train the preliminary pest and disease identification model to obtain the pest and disease identification model, the sample data in the target data set includes images marked with historical pest and disease information;

[0027] Establishing the control database: storing all pest type information in the control database, wherein the pest type information includes pathogen information, transmission route information, onset time and condition information and solution information.

[0028] As a further optimization scheme of the present invention, the specific steps of obtaining a preliminary pest identification model include:

[0029] Extract common features from sample data of the target dataset;

[0030] Selecting the source data set from a preset data set, wherein the sample data in the source data set and the sample data in the target data set have the common feature;

[0031] Performing a transformation operation on the feature dimension on the pre-trained feature extraction network;

[0032] The feature extraction network after performing the transformation operation is used as the input of the preset prediction network to form the preliminary pest and disease identification model.

[0033] As a further optimization solution of the present invention, step S4 specifically includes the following steps:

[0034] Performing geometric distortion correction on the sample photos to remove geometric deformation of the photos caused by the shooting process, selecting a uniform size for the images to perform cropping operations, unifying the image size and format, and obtaining a standard image;

[0035] Inputting the standard image into the pest and disease recognition model to extract image features, thereby obtaining image features of the standard image;

[0036] Based on the image features, a plurality of attribute features for characterizing the attributes of pests and diseases in the standard image are obtained;

[0037] Generating a joint attribute feature to be identified according to a plurality of the attribute features;

[0038] Identifying the joint attribute features to be identified to obtain pest and disease information in the standard image;

[0039] The pest information is simultaneously input into a control database, and control information is output.

[0040] The beneficial effects of the present invention are:

[0041] The present invention evenly sets sampling points in the jujube forest so that the sampling points cover the entire jujube forest. The sampling at the sampling points represents the disease and pest status of the jujube trees in the entire independent area, thereby avoiding the problem of difficulty in sampling in a large area of ​​jujube forests. The present invention also establishes a disease and pest model to identify the captured disease and pest photos, thereby replacing manual judgment based on experience, with higher accuracy, faster speed, high intelligence, and greatly improved recognition efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of the present invention;

[0043] Figure 2 is an aerial photograph of the present invention;

[0044] Figure 3 Draw the outline map of the jujube forest in the present invention;

[0045] Figure 4 It is the outline diagram of the jujube forest of the present invention;

[0046] Figure 5 The jujube forest outline map in the present invention is located in a grid coordinate map;

[0047] Figure 6 It is a small square grid diagram of the square grid coordinates of the present invention;

[0048] Figure 7 It is the pathogen information diagram of the present invention;

[0049] Figure 8 It is a diagram of jujube rust of the present invention;

[0050] Fig. 9 It is an anthrax diagram of the present invention;

[0051] Fig.10 is a spot disease diagram of the present invention;

[0052] Fig.11 It is a bacterial scab diagram of the present invention;

[0053] Fig.12 It is a diagram of a green stink bug of the present invention;

[0054] Fig.13 It is a red spider diagram of the present invention;

[0055] Fig.14 It is a diagram of the jujube looper of the present invention;

[0056] Fig.15 It is a scale insect diagram of the present invention. DETAILED DESCRIPTION

[0057] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0058] Example 1

[0059] like Figure 1-15 As shown, a method for rapid identification of diseases and insect pests of jujube trees comprises the following steps:

[0060] S1: draw a jujube forest outline map, and evenly set sampling points on the jujube forest outline map, and the jujube trees in the sampling points are sample jujube trees;

[0061] The step S1 specifically includes the following steps:

[0062] The jujube tree plantation is aerially photographed to obtain an aerial photograph, such as Figure 2 As shown;

[0063] The edge line of the jujube plantation in the aerial photograph is depicted, such as Figure 3 As shown, the outline of the jujube forest is obtained, such as Figure 4 As shown;

[0064] The area of ​​the jujube tree plantation is estimated, and the accuracy of the estimation is determined by the area of ​​the jujube tree plantation. The larger the area of ​​the jujube tree plantation, the lower the accuracy. The error between the estimated area and the actual area is within 10%;

[0065] Selecting a grid coordinate of appropriate specifications according to the estimated area of ​​the jujube plantation forest, that is, the area of ​​the jujube forest outline map is not less than the sum of the areas of the four small squares in the grid coordinates, that is, ensuring that there are at least four sampling points;

[0066] Put the jujube forest outline map into the grid coordinates;

[0067] Select all the complete small squares in the jujube forest outline map as an independent area. It should be noted that the area of ​​the small square in the jujube forest outline map is greater than or equal to 50% of its own area. The small square can be called a complete small square and can be used as an independent area.

[0068] Draw a circle with a suitable radius at the center of the independent area, the inside of the circle is the sampling area, and the outside of the circle is the edge area, such as Figure 6 As shown, the jujube trees in the sampling area are taken as sample jujube trees, wherein there are no less than 5 jujube trees in the sampling area. If the number of jujube trees in the sampling area of ​​the set radius is less than 5, the radius of the sampling area is expanded until there are no less than 5 jujube trees in the sampling area.

[0069] S2: Establish pest and disease identification model and control database;

[0070] The step S2 specifically includes the following steps:

[0071] Establishing the pest and disease identification model: pre-training a feature extraction network using a preset source data set, connecting the pre-trained feature extraction network and a preset prediction network to obtain a preliminary pest and disease identification model;

[0072] Using a target data set to train the preliminary pest and disease identification model to obtain the pest and disease identification model, the sample data in the target data set includes images marked with historical pest and disease information;

[0073] Establishing the control database: storing all pest type information in the control database, wherein the pest type information includes pathogen information, transmission route information, onset time and condition information and solution information.

[0074] Pathogen information includes disease information and pest information, such as Figure 7 As shown, the disease information includes leaf disease information and fruit disease information. The leaf disease information includes jujube madness information, rust information, bacterial scab information, etc.; the fruit disease information includes fruit shrinkage disease information, anthracnose information, fruit cracking disease information, spot disease information, etc.;

[0075] The pest information includes bud-feeding pest information and leaf-feeding pest information. The bud-feeding pest information includes weevils information and green stink bug information, etc.; the leaf-feeding pest information includes loopers information, date gall midge information, red spider information, date armyworm information, date leaf mite information, spiny moth information, date leaf tick information, date wax scale information and date mealybug information, etc.

[0076] The specific steps to obtain a preliminary pest and disease identification model include:

[0077] Extract common features from sample data of the target dataset;

[0078] Selecting the source data set from a preset data set, wherein the sample data in the source data set and the sample data in the target data set have the common feature;

[0079] Performing a transformation operation on the feature dimension on the pre-trained feature extraction network;

[0080] The feature extraction network after performing the transformation operation is used as the input of the preset prediction network to form the preliminary pest and disease identification model.

[0081] S3: Regularly take photos of the sample jujube trees in all the sampling points to obtain sample photos, such as Figure 8-Figure 15 .

[0082] S4: pre-processing the sample photos and inputting them into a pest and disease recognition model for recognition, outputting plant pest and disease information, and simultaneously inputting the pest and disease information into a prevention and control database, and outputting prevention and control information;

[0083] The step S4 specifically comprises the following steps:

[0084] Performing geometric distortion correction on the sample photos to remove geometric deformation of the photos caused by the shooting process, selecting a uniform size for the images to perform cropping operations, unifying the image size and format, and obtaining a standard image;

[0085] Inputting the standard image into the pest and disease recognition model to extract image features, thereby obtaining image features of the standard image;

[0086] Based on the image features, a plurality of attribute features for characterizing the attributes of pests and diseases in the standard image are obtained;

[0087] Generating a joint attribute feature to be identified according to a plurality of the attribute features;

[0088] Identifying the joint attribute features to be identified to obtain pest and disease information in the standard image;

[0089] The pest information is simultaneously input into a control database, and control information is output;

[0090] When the pest model extracts scattered light green, then light gray-brown, and yellow-brown spots on the back of the leaves in the standard image, forming a protruding uredia, and there are green spots facing the uredia, which later become yellow-brown angular spots, it can be determined to be jujube rust. Figure 8 As shown, the control database also outputs control information, specifically: Jujube rust is a parasitic disease, and the disease is serious in jujube gardens with many cypress crops around. Disease conditions and time: The disease is prone to occur when there is a lot of rain and high humidity. In July, summer spores germinate and invade through the stomata of the leaves. Symptoms appear in about 10 days, and leaves fall off. In mid-to-late August, a large number of leaves fall off. The early and late onset and severity of the disease are closely related to the rainfall, temperature and humidity of the year. Fundamental conditions: intercropping of tall crops under the tree, poor ventilation and light transmission, early and severe disease. The disease starts from the lower and middle parts of the crown and gradually spreads to the top.

[0091] Also includes solutions:

[0092] 1. Agricultural measures:

[0093] 1. Clean the garden before winter and remove dead branches and fallen leaves; plant plants at a reasonable density; and prune plants appropriately to ensure good ventilation and light transmission.

[0094] 2. Chemical control

[0095] 1. When cleaning the garden, spray 5000 times diluted 50% tebuconazole to kill the overwintering fungi.

[0096] 2. After germination, use 800 times diluted 80% mancozeb for prevention.

[0097] 3. Before the disease occurs in early to mid-July, use 80% mancozeb 600 times liquid spray to prevent and control winter jujube spot disease. Use 50% tebuconazole 5000-7000 times liquid spray to prevent and control in the early stage of the disease. Spray once every 7-10 days.

[0098] An embodiment of the present invention provides a computer program, which can execute the identification method in any of the above method embodiments.

[0099] An embodiment of the present invention provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed on a computer, the computer executes the identification method in any of the above method embodiments.

[0100] The algorithm or display provided here is not inherently related to any specific computer, virtual system or other equipment. Various general systems can also be used together with the teaching based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the embodiment of the present invention is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described here, and the description of the above specific language is for disclosing the best mode of the present invention.

[0101] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0102] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

[0103] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0104] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0105] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A rapid identification method for jujube tree diseases and insect pests, characterized in that: The following steps are involved: S1: draw a jujube forest outline map, and evenly set sampling points on the jujube forest outline map, and the jujube trees in the sampling points are sample jujube trees; S2: Establish pest and disease identification model and control database; S3: regularly taking photos of the sample jujube trees in all the sampling points to obtain sample photos; S4: pre-processing the sample photos and inputting them into the pest and disease recognition model for recognition, outputting the pest and disease information of the plants, and simultaneously inputting the pest and disease information into the prevention and control database, outputting the prevention and control information, The step S1 specifically includes the following steps: Take aerial photos of the date palm plantation and obtain aerial photographs; Describing the edge line of the jujube plantation in the aerial photograph to obtain the jujube forest outline map; To estimate the area of ​​the date palm plantations; Selecting grid coordinates of appropriate size according to the estimated area of ​​the jujube tree plantation; Put the jujube forest outline map into the grid coordinates; Select all the complete small squares in the jujube forest outline map as an independent area; Draw a circle with a suitable radius at the center of the independent area, the inside of the circle is the sampling area, and the outside of the circle is the edge area; The jujube trees within the sampling area are taken as sample jujube trees.

2. A rapid identification method for jujube tree diseases and insect pests according to claim 1, characterized in that: The area of ​​the date forest outline map is not less than the sum of the areas of the four small squares.

3. A rapid identification method for jujube tree diseases and insect pests according to claim 1, characterized in that: The area of ​​the small square within the jujube forest outline map is greater than or equal to 50% of its own area. The small square can be called a complete small square and can be used as an independent area.

4. The method for rapid identification of diseases and insect pests of jujube trees according to claim 1, characterized in that: There are no less than 5 jujube trees in the sampling area.

5. The method for rapid identification of diseases and insect pests of jujube trees according to claim 1, characterized in that: The step S2 specifically includes the following steps: Establishing the pest and disease identification model: pre-training a feature extraction network using a preset source data set, connecting the pre-trained feature extraction network and a preset prediction network to obtain a preliminary pest and disease identification model; Using a target data set to train the preliminary pest and disease identification model to obtain the pest and disease identification model, the sample data in the target data set includes images marked with historical pest and disease information; Establishing the control database: storing all pest type information in the control database, wherein the pest type information includes pathogen information, transmission route information, onset time and condition information and solution information.

6. A rapid identification method for jujube tree diseases and insect pests according to claim 5, characterized in that: The specific steps to obtain a preliminary pest and disease identification model include: Extract common features from sample data of the target dataset; Selecting the source data set from a preset data set, wherein the sample data in the source data set and the sample data in the target data set have the common feature; Performing a transformation operation on the feature dimension on the pre-trained feature extraction network; The feature extraction network after performing the transformation operation is used as the input of the preset prediction network to form the preliminary pest and disease identification model.

7. The method for rapid identification of diseases and insect pests of jujube trees according to claim 1, characterized in that: The step S4 specifically comprises the following steps: Performing geometric distortion correction on the sample photos to remove geometric deformation of the photos caused by the shooting process, selecting a uniform size for the images to perform cropping operations, unifying the image size and format, and obtaining a standard image; Inputting the standard image into the pest and disease recognition model to extract image features, thereby obtaining image features of the standard image; Based on the image features, a plurality of attribute features for characterizing the attributes of pests and diseases in the standard image are obtained; Generating a joint attribute feature to be identified according to a plurality of the attribute features; Identifying the joint attribute features to be identified to obtain pest and disease information in the standard image; The pest information is simultaneously input into a control database, and control information is output.

Images