Method and device for identifying plant pathology type based on big data

Abstract

The application relates to the technical field of artificial intelligence, and discloses a method and device for identifying plant pathological types based on big data, which comprises the following steps: collecting a plant image of a preset plant, generating a labeled image of the plant image; performing image correction on the labeled image to obtain a corrected image of the labeled image; generating structural features of the preset plant according to the corrected image; performing image classification on the corrected image according to the structural features to obtain a classified image of the corrected image, generating a feature database of the classified image; acquiring a real-time image of a target plant, and generating a pathological type of the target plant according to the real-time image, a preset similarity algorithm and the feature database. The application can improve the identification efficiency of plant pathological types.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method and apparatus for identifying plant pathological types based on big data. Background Technology

[0002] Plant diseases not only affect yield but also seriously impact the quality of agricultural products. Some diseases can cause spoilage, decreased nutritional value, and even the production of toxins, posing a threat to human health. Timely identification of plant pathological types allows for early prevention and control, ensuring the quality and safety of agricultural products.

[0003] Current plant pathology type identification technology mainly relies on manually labeled data. However, this method requires a lot of time, and the accuracy of the labeling results is also subject to certain errors. Therefore, how to improve the efficiency of plant pathology type identification has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and apparatus for identifying plant pathological types based on big data, the main purpose of which is to solve the problem of low efficiency in identifying plant pathological types.

[0005] To achieve the above objectives, this invention provides a method for identifying plant pathological types based on big data, comprising:

[0006] Collect plant images of preset plants and generate labeled images of the plant images;

[0007] The labeled image is corrected to obtain the corrected image of the labeled image;

[0008] Generate the structural features of the preset plant based on the corrected image;

[0009] Based on the structural features, the corrected image is classified to obtain a classified image of the corrected image, and a feature database of the classified image is generated.

[0010] A real-time image of the target plant is acquired. Based on the real-time image, a preset similarity algorithm, and the feature database, a pathological type of the target plant is generated. The preset similarity algorithm is as follows:

[0011]

[0012] Where d(x,y) is the feature similarity between the real-time feature and the corresponding feature in the feature database, x i It is the i-th feature element of the real-time feature, y iIt is the i-th feature element of the comparison feature, where i is the element identifier of the feature element, and n is the total number of feature elements in the real-time feature.

[0013] Optionally, the acquisition of plant images of preset plants includes:

[0014] Generate plant images of preset plants using preset digital devices;

[0015] Recording parameters for generating the plant image;

[0016] A plant image of the preset plant is generated based on the recorded parameters and the plant image.

[0017] Optionally, the annotation image for generating the plant image includes:

[0018] Generate annotation information for the preset plant based on the plant image;

[0019] The plant image is labeled using the annotation information to obtain an annotated image of the plant image.

[0020] Optionally, the step of performing image correction on the labeled image to obtain a corrected image of the labeled image includes:

[0021] The labeled image is subjected to illumination correction to obtain a first-level image of the labeled image;

[0022] Perform color correction on the primary image to obtain a color-corrected primary image;

[0023] Geometric correction is performed on the color-corrected primary image to obtain the corrected image of the labeled image.

[0024] Optionally, generating the structural features of the preset plant based on the corrected image includes:

[0025] The corrected image is segmented to obtain a segmented image of the corrected image;

[0026] Generate the predefined plant spot distribution features based on the segmented image;

[0027] Generate the spot density characteristics of the preset plant;

[0028] The distribution characteristics and density characteristics of the spots are combined to form the structural characteristics of the preset plant.

[0029] Optionally, generating the spot density of the preset plant includes:

[0030] The spot density of the preset plant is generated using a preset density algorithm, wherein the preset density algorithm is:

[0031]

[0032] Where m is the spot density of the preset plant, S is the total number of spots of the preset plant, and A is the sampling area of ​​the preset plant.

[0033] Optionally, the step of classifying the corrected image based on the structural features to obtain a classified image of the corrected image includes:

[0034] The structural features are classified to obtain the classification features of the structural features;

[0035] The corrected image is classified using the classification features to obtain a classified image of the corrected image.

[0036] Optionally, the feature database for generating the classified images includes:

[0037] Generate feature labels for the classification images based on the classification features corresponding to the classification images;

[0038] Generate the feature index of the classified image;

[0039] The feature database of the classified image is generated based on the feature index and the feature label.

[0040] Optionally, the step of generating the pathological type of the target plant based on the real-time image and the feature database includes:

[0041] Extract real-time features from the real-time image;

[0042] Calculate the feature similarity between each real-time feature and the corresponding feature in the feature database;

[0043] The feature attributes of the real-time features are generated based on the feature similarity and the preset similarity threshold;

[0044] The pathological type of the target plant is generated based on the aforementioned characteristic attributes.

[0045] To address the aforementioned problems, the present invention also provides a device for identifying plant pathological types based on big data, the device comprising:

[0046] The image acquisition module is used to acquire plant images of preset plants and generate labeled images of the plant images;

[0047] An image correction module is used to perform image correction on the labeled image to obtain a corrected image of the labeled image;

[0048] A structural feature generation module is used to generate the structural features of the preset plant based on the corrected image;

[0049] An image classification module is used to classify the corrected image according to the structural features, obtain a classified image of the corrected image, and generate a feature database of the classified image;

[0050] The pathology type generation module is used to acquire real-time images of the target plant, and generate the pathology type of the target plant based on the real-time images, a preset similarity algorithm, and the feature database. The preset similarity algorithm is as follows:

[0051]

[0052] Where d(x,y) is the feature similarity between the real-time feature and the corresponding feature in the feature database, x i It is the i-th feature element of the real-time feature, y i It is the i-th feature element of the comparison feature, where i is the element identifier of the feature element, and n is the total number of feature elements in the real-time feature.

[0053] This invention, through the acquisition of plant images and the generation of labeled images, can automatically collect a large amount of data. Then, image correction and feature extraction are performed on the labeled images to obtain clear and accurate corrected images and structural feature information. Using this feature information, real-time images of target plants can be quickly and accurately classified and identified. Finally, pathological type identification is performed through a preset similarity algorithm and feature database, which can improve identification efficiency and accuracy. The entire process combines multiple technologies such as image processing, machine learning, and data mining, and can make full use of the advantages of big data. Therefore, this invention proposes a method and device for identifying plant pathological types based on big data, which can solve the problem of low identification efficiency of plant pathological types. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating a method for identifying plant pathological types based on big data, according to an embodiment of the present invention.

[0055] Figure 2 This is a schematic diagram of the image correction process for an annotated image provided in an embodiment of the present invention;

[0056] Figure 3 This is a schematic diagram of a process for generating the structural features of a preset plant according to an embodiment of the present invention;

[0057] Figure 4 This is a functional module diagram of a plant pathology type identification device based on big data provided in an embodiment of the present invention;

[0058] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0059] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0060] This application provides a method for identifying plant pathology types based on big data. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for identifying plant pathology types based on big data can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0061] Reference Figure 1 The diagram shown is a flowchart illustrating a method for identifying plant pathology types based on big data, according to an embodiment of the present invention. In this embodiment, the method for identifying plant pathology types based on big data includes:

[0062] S1. Collect plant images of preset plants and generate labeled images of the plant images.

[0063] In this embodiment of the invention, the acquisition of plant images of preset plants includes:

[0064] Generate plant images of preset plants using preset digital devices;

[0065] Recording parameters for generating the plant image;

[0066] A plant image of the preset plant is generated based on the recorded parameters and the plant image.

[0067] In detail, generating a plant image of a preset plant using a preset digital device refers to taking a picture of the preset plant using a digital camera or smartphone and generating a plant image of the preset plant.

[0068] In detail, the recording parameters of the plant images include: shooting time, shooting location, shooting light, camera parameters, and other information.

[0069] In detail, generating the plant image of the preset plant based on the recorded parameters and the plant image means associating and storing the recorded parameters and the plant image to obtain the plant image of the preset plant. This association storage can be implemented in various ways, such as saving the recorded parameters as part of the image file name, or saving the recorded parameters as metadata associated with the image file.

[0070] In addition, the generated record parameters can be used for subsequent steps such as image correction, feature extraction, and classification to improve the accuracy and reliability of pathological diagnosis.

[0071] In this embodiment of the invention, generating the labeled image of the plant image includes:

[0072] Generate annotation information for the preset plant based on the plant image;

[0073] The plant image is labeled using the annotation information to obtain an annotated image of the plant image.

[0074] In detail, the labeling information includes: the growth status, morphological characteristics, and leaf color of the preset plant.

[0075] In detail, by labeling the growth status of preset plants, it is possible to understand which stage of the plant's life cycle it is in, such as the seedling stage, flowering stage, fruiting stage, etc., which is very important for judging the plant's health status and the possibility of pathological problems.

[0076] In detail, by labeling the morphological characteristics of preset plants, the external structure and shape of plants can be described in detail, such as root morphology, stem thickness, and the arrangement of branches and leaves. These characteristics can provide more information and help to analyze the growth status and potential diseases of plants.

[0077] In detail, by marking the leaf colors of preset plants, changes in the hue, saturation, and brightness of plant leaves can be observed. Changes in leaf color may indicate the plant's nutritional status, pathological problems, or environmental stress.

[0078] S2. Perform image correction on the labeled image to obtain the corrected image of the labeled image.

[0079] In this embodiment of the invention, the reference Figure 2 As shown, the step of performing image correction on the labeled image to obtain the corrected image of the labeled image includes:

[0080] S21. Perform illumination correction on the labeled image to obtain a first-level image of the labeled image;

[0081] S22. Perform color correction on the primary image to obtain a color-corrected primary image;

[0082] S23. Perform geometric correction on the color-corrected first-level image to obtain the corrected image of the labeled image.

[0083] In detail, the illumination correction aims to eliminate brightness differences in the image caused by uneven light sources or shadows. Through illumination correction, the overall brightness of the image can be made more uniform, the details clearer, and the subsequent processing will not be affected by lighting conditions.

[0084] In detail, the color correction aims to adjust the color balance of the image, eliminate color deviations caused by the color of the light source or the camera equipment itself, and ensure that the presented colors are more consistent with the actual scene, which helps to accurately describe characteristics such as the color of plant leaves.

[0085] In detail, the geometric correction aims to eliminate distortion problems that may exist in the image due to shooting angle or lens distortion, so that the objects in the image are presented more realistically and accurately in terms of geometry, which helps to accurately describe the morphological characteristics of plants.

[0086] S3. Generate the structural features of the preset plant based on the corrected image.

[0087] In this embodiment of the invention, the reference Figure 3 As shown, generating the structural features of the preset plant based on the corrected image includes:

[0088] S31. Perform image segmentation on the corrected image to obtain a segmented image of the corrected image;

[0089] S32. Generate the spot distribution features of the preset plant based on the segmented image;

[0090] S33. Generate the spot density characteristics of the preset plant;

[0091] S34. The distribution characteristics of the spots and the density characteristics of the spots are combined to form the structural characteristics of the preset plant.

[0092] In detail, the image segmentation is the process of dividing the pixels in an image into different regions, thereby enabling the analysis and processing of the image.

[0093] In this embodiment of the invention, by performing image segmentation on the corrected image, different parts of the plant (such as leaves, branches, flowers, etc.) can be separated, laying the foundation for subsequent feature extraction and analysis.

[0094] In detail, the distribution characteristics of the spots can reflect information such as the plant's health status, growth status, and pathological problems. The distribution of the spots can be described by statistical analysis of the positional relationship and mutual distance of the spots.

[0095] In detail, spot density characteristics refer to the number of spots per unit area.

[0096] Specifically, generating the spot density of the preset plant includes:

[0097] The spot density of the preset plant is generated using a preset density algorithm, wherein the preset density algorithm is:

[0098]

[0099] Where m is the spot density of the preset plant, S is the total number of spots of the preset plant, and A is the sampling area of ​​the preset plant.

[0100] In detail, firstly, it is necessary to determine the sampling area of ​​the preset plant, which can be the entire area of ​​the plant or the area of ​​a specific part of the plant; then, spot detection or segmentation needs to be performed on the preset plant to obtain the total number of spots S, where spot detection or segmentation can be performed using algorithms such as threshold segmentation, morphological processing, and machine learning; according to the preset density algorithm, the spot density m is equal to the total number of spots S divided by the sampling area A; finally, the calculated spot density m can be used as one of the structural features of the preset plant for further analysis, identification, or processing.

[0101] S4. Based on the structural features, classify the corrected image to obtain a classified image of the corrected image, and generate a feature database of the classified image.

[0102] In this embodiment of the invention, the step of classifying the corrected image based on the structural features to obtain a classified image of the corrected image includes:

[0103] The structural features are classified to obtain the classification features of the structural features;

[0104] The corrected image is classified using the classification features to obtain a classified image of the corrected image.

[0105] In detail, the structural features can be classified using the K-means clustering algorithm to obtain the classification features of the structural features.

[0106] Furthermore, K-means clustering is a commonly used unsupervised learning algorithm for dividing a dataset into K distinct clusters. For the step of classifying structural features to obtain their classification features, the K-means clustering algorithm can be used, with the following specific steps: randomly select K points as initial cluster centers; calculate the distance between each sample point and the K cluster centers, and assign the sample point to the cluster corresponding to the nearest cluster center; for each cluster, recalculate the average value of the sample points within that cluster, updating the cluster centers, until a termination condition is met, such as reaching the maximum number of iterations or the cluster centers no longer changing; finally, each sample point is assigned to a cluster, and its classification feature can be represented by its cluster affiliation.

[0107] In detail, during the feature extraction stage, feature vectors are obtained by calculating the structural features of the corrected image, such as speckle density and speckle distribution. Then, the K-means clustering algorithm is used to divide these feature vectors into K different clusters, thus obtaining the classification features of the structural features.

[0108] It is important to note that when applying the K-means clustering algorithm, it is necessary to select an appropriate K value and distance metric, and to perform appropriate convergence checks.

[0109] In this embodiment of the invention, the feature database for generating the classified image includes:

[0110] Generate feature labels for the classification images based on the classification features corresponding to the classification images;

[0111] Generate the feature index of the classified image;

[0112] The feature database of the classified image is generated based on the feature index and the feature label.

[0113] In detail, based on the classification features corresponding to the classification images, feature labels are generated for each classification image. The feature labels can be strings, numbers, or other forms of identifiers used to uniquely identify and represent the features of each classification image.

[0114] In detail, the feature index is a data structure used to store and organize feature data of classified images. The feature index can be implemented using hash tables, tree structures, or other indexing methods to improve data retrieval efficiency.

[0115] In detail, the feature database is a collection that stores categorized image feature data, which can be quickly retrieved and queried using feature indexes and feature labels.

[0116] S5. Obtain a real-time image of the target plant, and generate the pathological type of the target plant based on the real-time image and the feature database.

[0117] In this embodiment of the invention, real-time images of the target plant can be acquired using a camera.

[0118] In this embodiment of the invention, the step of generating the pathological type of the target plant based on the real-time image and the feature database includes:

[0119] Extract real-time features from the real-time image;

[0120] Calculate the feature similarity between each real-time feature and the corresponding feature in the feature database;

[0121] The feature attributes of the real-time features are generated based on the feature similarity and the preset similarity threshold;

[0122] The pathological type of the target plant is generated based on the aforementioned characteristic attributes.

[0123] Specifically, the real-time features refer to the distribution and density of spots on the target plant.

[0124] Furthermore, real-time features are used to determine the pathological type of the target plant because different pathological types typically produce specific spot distribution patterns in different parts of the plant. For example, some diseases may produce spots on the edges of leaves, while others may be evenly distributed across the entire leaf surface. Generally, the more severe the disease, the higher the spot density.

[0125] In detail, the step of calculating the feature similarity between each real-time feature and the corresponding feature in the feature database includes:

[0126] The feature similarity between the real-time feature and the corresponding feature in the feature database is calculated one by one using a preset similarity algorithm, wherein the preset similarity algorithm is:

[0127]

[0128] Where d(x,y) is the feature similarity between the real-time feature and the corresponding feature in the feature database, x i It is the i-th feature element of the real-time feature, y i It is the i-th feature element of the comparison feature, where i is the element identifier of the feature element, and n is the total number of feature elements in the real-time feature.

[0129] In detail, the feature attribute can be a degree attribute. For example, based on a preset similarity threshold, the similarity calculation result can be divided into different feature attributes, such as normal, mild infection, moderate infection, severe infection, etc.

[0130] Furthermore, if the feature attribute of the real-time feature corresponds to "minor infection", then the pathological type of the target plant can be determined as a minor infection disease type.

[0131] Similarly, if the characteristic attribute corresponds to "severe infection", then the pathological type of the target plant can be identified as a severely infected disease type.

[0132] This invention, through the acquisition of plant images and the generation of labeled images, can automatically collect a large amount of data. Then, image correction and feature extraction are performed on the labeled images to obtain clear and accurate corrected images and structural feature information. Using this feature information, real-time images of target plants can be quickly and accurately classified and identified. Finally, pathological type identification is performed through a preset similarity algorithm and feature database, which can improve identification efficiency and accuracy. The entire process combines multiple technologies such as image processing, machine learning, and data mining, and can make full use of the advantages of big data. Therefore, this invention proposes a method for identifying plant pathological types based on big data, which can solve the problem of low identification efficiency of plant pathological types.

[0133] like Figure 4 The diagram shown is a functional block diagram of a plant pathology type identification device based on big data provided in an embodiment of the present invention.

[0134] The plant pathology type identification device 100 based on big data described in this invention can be installed in an electronic device. Depending on the functions implemented, the plant pathology type identification device 100 based on big data may include an image acquisition module 101, an image correction module 102, a structural feature generation module 103, an image classification module 104, and a pathology type generation module 105. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0135] In this embodiment, the functions of each module / unit are as follows:

[0136] The image acquisition module 101 is used to acquire plant images of preset plants and generate labeled images of the plant images;

[0137] The image correction module 102 is used to perform image correction on the labeled image to obtain a corrected image of the labeled image;

[0138] The structural feature generation module 103 is used to generate the structural features of the preset plant based on the corrected image;

[0139] The image classification module 104 is used to classify the corrected image according to the structural features, obtain the classified image of the corrected image, and generate the feature database of the classified image;

[0140] The pathology type generation module 105 is used to acquire real-time images of the target plant, and generate the pathology type of the target plant based on the real-time images, a preset similarity algorithm, and the feature database. The preset similarity algorithm is as follows:

[0141]

[0142] Where d(x,y) is the feature similarity between the real-time feature and the corresponding feature in the feature database, x i It is the i-th feature element of the real-time feature, y i is the i-th feature element of the comparison feature, where i is the element identifier of the feature element, and n is the total number of feature elements in the real-time feature. In the several embodiments provided by this invention, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for example, the division of modules is merely a logical functional division, and other division methods may exist in actual implementation.

[0143] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0144] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0145] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for identifying plant pathological types based on big data, characterized in that, The method includes: Collect plant images of preset plants and generate labeled images of the plant images; The labeled image is corrected to obtain the corrected image of the labeled image; Generate the structural features of the preset plant based on the corrected image; Based on the structural features, the corrected image is classified to obtain a classified image of the corrected image, and a feature database of the classified image is generated. A real-time image of the target plant is acquired. Based on the real-time image, a preset similarity algorithm, and the feature database, a pathological type of the target plant is generated. The preset similarity algorithm is as follows: Where d(x,y) is the feature similarity between the real-time feature and the corresponding feature in the feature database, x i It is the i-th feature element of the real-time feature, y i It is the i-th feature element of the comparison feature, where i is the element identifier of the feature element, and n is the total number of feature elements in the real-time feature.

2. The method for identifying plant pathological types based on big data as described in claim 1, characterized in that, The acquisition of plant images of preset plants includes: Generate plant images of preset plants using preset digital devices; Recording parameters for generating the plant image; A plant image of the preset plant is generated based on the recorded parameters and the plant image.

3. The method for identifying plant pathological types based on big data as described in claim 1, characterized in that, The labeled image used to generate the plant image includes: Generate annotation information for the preset plant based on the plant image; The plant image is labeled using the annotation information to obtain an annotated image of the plant image.

4. The method for identifying plant pathological types based on big data as described in claim 1, characterized in that, The step of performing image correction on the labeled image to obtain the corrected image of the labeled image includes: The labeled image is subjected to illumination correction to obtain a first-level image of the labeled image; Perform color correction on the primary image to obtain a color-corrected primary image; Geometric correction is performed on the color-corrected primary image to obtain the corrected image of the labeled image.

5. The method for identifying plant pathological types based on big data as described in claim 1, characterized in that, The step of generating the structural features of the preset plant based on the corrected image includes: The corrected image is segmented to obtain a segmented image of the corrected image; Generate the predefined plant spot distribution features based on the segmented image; Generate the spot density characteristics of the preset plant; The distribution characteristics and density characteristics of the spots are combined to form the structural characteristics of the preset plant.

6. The method for identifying plant pathological types based on big data as described in claim 5, characterized in that, Generating the preset plant spot density includes: The spot density of the preset plant is generated using a preset density algorithm, wherein the preset density algorithm is: Where m is the spot density of the preset plant, S is the total number of spots of the preset plant, and A is the sampling area of ​​the preset plant.

7. The method for identifying plant pathological types based on big data as described in claim 1, characterized in that, The step of classifying the corrected image based on the structural features to obtain a classified image of the corrected image includes: The structural features are classified to obtain the classification features of the structural features; The corrected image is classified using the classification features to obtain a classified image of the corrected image.

8. The method for identifying plant pathological types based on big data as described in claim 1, characterized in that, The feature database for generating the classified images includes: Generate feature labels for the classification images based on the classification features corresponding to the classification images; Generate the feature index of the classified image; The feature database of the classified image is generated based on the feature index and the feature label.

9. The method for identifying plant pathological types based on big data as described in any one of claims 1 to 8, characterized in that, The step of generating the pathological type of the target plant based on the real-time image and the feature database includes: Extract real-time features from the real-time image; Calculate the feature similarity between each real-time feature and the corresponding feature in the feature database; The feature attributes of the real-time features are generated based on the feature similarity and the preset similarity threshold; The pathological type of the target plant is generated based on the aforementioned characteristic attributes.

10. A device for identifying plant pathological types based on big data, characterized in that, The device includes: The image acquisition module is used to acquire plant images of preset plants and generate labeled images of the plant images; An image correction module is used to perform image correction on the labeled image to obtain a corrected image of the labeled image; A structural feature generation module is used to generate the structural features of the preset plant based on the corrected image; An image classification module is used to classify the corrected image according to the structural features, obtain a classified image of the corrected image, and generate a feature database of the classified image; The pathology type generation module is used to acquire real-time images of the target plant, and generate the pathology type of the target plant based on the real-time images, a preset similarity algorithm, and the feature database. The preset similarity algorithm is as follows: Where d(x,y) is the feature similarity between the real-time feature and the corresponding feature in the feature database, x i It is the i-th feature element of the real-time feature, y i It is the i-th feature element of the comparison feature, where i is the element identifier of the feature element, and n is the total number of feature elements in the real-time feature.

Images