A method and device for investigating rice planthoppers in a field based on AR glasses

Abstract

The application discloses a field rice planthopper investigation method and device based on AR glasses, and the method comprises the following steps: acquiring a white plate image collected by the AR glasses; inputting the white plate image into a convolutional neural network model to obtain a white plate image marked with rice planthopper types; performing statistical processing on the white plate image marked with the rice planthopper types to obtain the quantity of each rice planthopper type, and sending the white plate image, each rice planthopper type and the corresponding quantity to a client to display each rice planthopper type and the corresponding quantity in the white plate image by the client. The rice planthoppers flying on the white plate image are identified by the convolutional neural network model, which can effectively avoid the low efficiency, high professional skill and high labor intensity caused by traditional manual plate beating and counting. In addition, the identification and counting results can be fed back to the client to realize the interaction between the AR glasses and the mobile terminal, thereby improving the user experience.

Description

Technical Field

[0001] This application belongs to the field of intelligent agricultural pest survey technology, and specifically relates to a method and device for surveying rice planthoppers in the field based on AR glasses. Background Technology

[0002] Rice is one of the most important food crops, with about half the world's population relying on it as their staple food. Rice planthoppers, a major pest of rice, cause significant yield and economic losses to the rice industry every year. They are mainly divided into three types: brown planthoppers, white-backed planthoppers, and gray planthoppers. These three types of planthoppers not only suck rice sap but also transmit viral diseases. Therefore, accurate monitoring and forecasting of rice planthoppers is a prerequisite for achieving precise and green pest control.

[0003] The current "Specification for Rice Planthopper Monitoring and Survey (GB / T 15794-2009)" details the survey methods, time and location, survey items, and data reporting format. This specification stipulates that a white tray is used as the carrier. When checking for insects, the tray is gently inserted into the rice row, with the lower edge close to the base of the rice clump on the water surface. The tray is then quickly tapped on the middle and lower parts of the plant, three times in succession, with a count at each point. The number of adult planthoppers of different wing types, as well as the number of older and younger nymphs, is counted. This survey method requires two people to work together: one person goes into the field to tap, identify, and count the insects, while the other stands at the field edge to record the data. This method has certain limitations. When different planthopper species are mixed, the surveyor needs to accurately identify the planthopper species and quickly count the older and younger nymphs, requiring a high level of professional skill. When the insect population density is high, the surveyor often uses a zone-based method to give an estimate to prevent planthopper escape, which is inaccurate. Furthermore, there is currently a shortage of grassroots plant protection personnel, and the field survey workload is heavy; the traditional manual tray tapping and counting method can no longer meet the needs of modern agriculture. Therefore, it is necessary to develop a new monitoring and survey method and supporting equipment that is fast, simple, and conforms to the monitoring and survey specifications for rice planthoppers. Summary of the Invention

[0004] To address the limitations of traditional manual sampling and counting methods in modern agriculture, this application proposes a field survey method and device for rice planthoppers based on AR glasses. The technical solution is as follows:

[0005] In a first aspect, embodiments of this application provide a method for surveying rice planthoppers in the field based on AR glasses, including:

[0006] Acquire the white disk image captured by the AR glasses;

[0007] The white disk images are transmitted to a convolutional neural network model to obtain white disk images labeled with rice planthopper types; the convolutional neural network model is trained from multiple sample images labeled with rice planthopper types;

[0008] Statistical processing is performed on white disk images labeled with rice planthopper types to obtain the quantity of each rice planthopper type. The white disk images, each rice planthopper type, and the corresponding quantity are then sent to the client so that the client can display each rice planthopper type and the corresponding quantity in the white disk images.

[0009] In one alternative of the first aspect, acquiring the white disk image captured by the AR glasses includes:

[0010] Send a location request command to the AR glasses to obtain the current location fed back by the AR glasses;

[0011] When the current location is detected to be within the specified area, an image request command is generated;

[0012] The image request command is sent to the AR glasses, which then generate corresponding prompts for the user based on the image request command. The AR glasses also capture the white disk image based on the user's first voice command.

[0013] Acquire the white disk image captured by the AR glasses.

[0014] In another alternative to the first aspect, after acquiring the white disk image captured by the AR glasses, the method further includes:

[0015] Determine whether the edge of the white disk in the white disk image is within the display screen of the AR glasses;

[0016] When it is detected that part of the white disk edge is not within the display screen of the AR glasses, zoom prompt information is generated based on the position of the white disk in the white disk image;

[0017] The zoom prompt information is sent to the AR glasses, which then display the zoom prompt information to the user through the screen. The AR glasses also perform optical zoom processing based on the user's second voice command and capture a first image. The edge of the white disk in the first image is within the display screen of the AR glasses.

[0018] Acquire the first image captured by the AR glasses and replace the white disk image with the first image.

[0019] In another alternative to the first aspect, after acquiring the white disk image captured by the AR glasses, the method further includes:

[0020] Gaussian blurring is applied to the white disk image to obtain a blurred image corresponding to the white disk image;

[0021] Calculate the first gray level difference between any two adjacent pixels in the white image and the second gray level difference between any two adjacent pixels in the blurred image.

[0022] The first grayscale difference and the second grayscale difference are normalized, and when the processing result is detected to be within a preset range, an image is generated to obtain prompt information.

[0023] The image acquisition prompt is sent to the AR glasses, which then display the prompt to the user on the screen. The AR glasses also acquire a second image based on the user's third voice command.

[0024] Acquire the second image captured by the AR glasses and replace the white disk image with the second image.

[0025] In another alternative to the first aspect, the white disk image is transmitted to a convolutional neural network model to obtain a white disk image labeled with the rice planthopper type, including:

[0026] The white disk image is divided into blocks to obtain m sub-images with the same area; where m is a positive integer greater than or equal to 4.

[0027] Each sub-image is fed into a convolutional neural network model to obtain sub-images labeled with the rice planthopper type;

[0028] Image synthesis processing was performed on all sub-images labeled with the rice planthopper type to obtain white disc images labeled with the rice planthopper type.

[0029] In another alternative to the first aspect, after statistically processing the white disc images labeled with rice planthopper types to obtain the number of each rice planthopper type, the method further includes:

[0030] The current time corresponding to the AR glasses is obtained, and a change curve is generated based on the current location, the current time, the number of each type of rice planthopper, and historical records; wherein, the historical records include at least two historical times corresponding to the current location, and the number of each type of rice planthopper in the white disk image corresponding to each historical time.

[0031] The change curve is sent to the client so that the user can view the change curve on the client.

[0032] In another alternative of the first aspect, the sample images marked with rice planthopper types include any one of the following: long-winged adult white-backed planthopper, short-winged adult white-backed planthopper, older nymph white-backed planthopper, long-winged adult brown planthopper, short-winged adult brown planthopper, older nymph brown planthopper, long-winged adult gray planthopper, short-winged adult gray planthopper, older nymph gray planthopper, young nymph rice planthopper, spider, stink bug, rove beetle, leafhopper adult, and leafhopper nymph.

[0033] Secondly, embodiments of this application provide a field rice planthopper survey device based on AR glasses, comprising:

[0034] The image acquisition module is used to acquire the white disk image captured by the AR glasses;

[0035] The image processing module is used to transmit the white disk image to the convolutional neural network model to obtain the white disk image labeled with the rice planthopper type; wherein, the convolutional neural network model is trained from multiple sample images labeled with the rice planthopper type;

[0036] The image display module is used to perform statistical processing on white disk images marked with rice planthopper types, obtain the quantity of each rice planthopper type, and send the white disk image, each rice planthopper type, and the corresponding quantity to the client so that the client can display each rice planthopper type and the corresponding quantity in the white disk image.

[0037] In one alternative embodiment of the second aspect, the image acquisition module is specifically used for:

[0038] Send a location request command to the AR glasses to obtain the current location fed back by the AR glasses;

[0039] When the current location is detected to be within the specified area, an image request command is generated;

[0040] The image request command is sent to the AR glasses, which then generate corresponding prompts for the user based on the image request command. The AR glasses also capture the white disk image based on the user's first voice command.

[0041] Acquire the white disk image captured by the AR glasses.

[0042] In another alternative solution of the second aspect, the image acquisition module is specifically used for:

[0043] After acquiring the white disk image captured by the AR glasses, determine whether the edge of the white disk in the white disk image is within the display screen of the AR glasses;

[0044] When it is detected that part of the white disk edge is not within the display screen of the AR glasses, zoom prompt information is generated based on the position of the white disk in the white disk image;

[0045] The zoom prompt information is sent to the AR glasses, which then display the zoom prompt information to the user through the screen. The AR glasses also perform optical zoom processing based on the user's second voice command and capture a first image. The edge of the white disk in the first image is within the display screen of the AR glasses.

[0046] Acquire the first image captured by the AR glasses and replace the white disk image with the first image.

[0047] In another alternative solution of the second aspect, the image acquisition module is specifically used for:

[0048] After acquiring the white disk image captured by the AR glasses, the white disk image is subjected to Gaussian blur processing to obtain a blurred image corresponding to the white disk image;

[0049] Calculate the first gray level difference between any two adjacent pixels in the white image and the second gray level difference between any two adjacent pixels in the blurred image.

[0050] The first grayscale difference and the second grayscale difference are normalized, and when the processing result is detected to be within a preset range, an image is generated to obtain prompt information.

[0051] The image acquisition prompt is sent to the AR glasses, which then display the prompt to the user on the screen. The AR glasses also acquire a second image based on the user's third voice command.

[0052] Acquire the second image captured by the AR glasses and replace the white disk image with the second image.

[0053] In another alternative embodiment of the second aspect, the image processing module is specifically used for:

[0054] The white disk image is divided into blocks to obtain m sub-images with the same area; where m is a positive integer greater than or equal to 4.

[0055] Each sub-image is fed into a convolutional neural network model to obtain sub-images labeled with the rice planthopper type;

[0056] Image synthesis processing was performed on all sub-images labeled with the rice planthopper type to obtain white disc images labeled with the rice planthopper type.

[0057] In another alternative embodiment of the second aspect, the device further includes:

[0058] After statistically processing the white disk images labeled with rice planthopper types to obtain the number of each type, the current time corresponding to the AR glasses is obtained, and a change curve is generated based on the current location, the current time, the number of each type of rice planthopper, and historical records. The historical records include at least two historical times corresponding to the current location, and the number of each type of rice planthopper in the white disk image corresponding to each historical time.

[0059] The change curve is sent to the client so that the user can view the change curve on the client.

[0060] In another alternative of the second aspect, the sample images marked with rice planthopper types include any one of the following: long-winged adult white-backed planthopper, short-winged adult white-backed planthopper, older nymph white-backed planthopper, long-winged adult brown planthopper, short-winged adult brown planthopper, older nymph brown planthopper, long-winged adult gray planthopper, short-winged adult gray planthopper, older nymph gray planthopper, young nymph rice planthopper, spider, stink bug, rove beetle, leafhopper adult, and leafhopper nymph.

[0061] Thirdly, this application also provides a field rice planthopper survey device based on AR glasses, including a processor and a memory;

[0062] The processor is connected to the memory;

[0063] Memory, used to store executable program code;

[0064] The processor runs a program corresponding to the executable program code stored in the memory to implement the field rice planthopper survey method based on AR glasses provided in the first aspect or any implementation of the first aspect of the embodiments of this application.

[0065] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which includes program instructions. When executed by a processor, the program instructions can implement the field rice planthopper survey method based on AR glasses provided by the first aspect or any implementation of the first aspect of this application.

[0066] In this embodiment, during a survey of rice planthoppers in the field, white disc images collected by AR glasses can be obtained. These white disc images are then transmitted to a convolutional neural network model to obtain white disc images labeled with different rice planthopper types. Statistical processing is performed on these labeled white disc images to determine the quantity of each planthopper type. The white disc images, along with each planthopper type and its corresponding quantity, are then sent to a client, allowing the client to display each planthopper type and its quantity within the white disc images. This method of identifying rice planthoppers swatting at the white disc images using a convolutional neural network model effectively avoids the inefficiencies, high skill requirements, and heavy labor intensity associated with traditional manual swatting and counting. Furthermore, by feeding back the identification and counting results to the client, interaction between the AR glasses and the mobile terminal is achieved, thereby enhancing the user experience. Attached Figure Description

[0067] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 A flowchart illustrating an AR glasses-based method for surveying rice planthoppers in the field, as provided in this application embodiment;

[0069] Figure 2 A schematic diagram of the architecture of a field rice planthopper survey system based on AR glasses is provided for an embodiment of this application;

[0070] Figure 3 A schematic diagram illustrating the prediction performance of a convolutional neural network model provided in an embodiment of this application;

[0071] Figure 4 A schematic diagram of a client display interface provided in an embodiment of this application;

[0072] Figure 5 A schematic diagram of a field rice planthopper survey device based on AR glasses provided in this application embodiment;

[0073] Figure 6 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation

[0074] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0075] In the following description, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The following description provides multiple embodiments of this application, which can be substituted or combined with each other. Therefore, this application can also be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then this application should also be considered to include embodiments containing one or more other possible combinations of A, B, C, and D, even if such embodiments are not explicitly described in the following text.

[0076] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the function and arrangement of the described elements without departing from the scope of this application. Various processes or components may be appropriately omitted, substituted, or added to the examples. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

[0077] Please see Figure 1 , Figure 1 The diagram shows an overall flowchart of a field rice planthopper survey method based on AR glasses, as provided in an embodiment of this application.

[0078] like Figure 1 As shown, this AR glasses-based method for surveying rice planthoppers in the field may include at least the following steps:

[0079] Step 102: Obtain the white disk image captured by the AR glasses.

[0080] In this embodiment of the application, the AR glasses-based method for surveying rice planthoppers in the field can be applied to, but is not limited to, servers; see reference here. Figure 2 The diagram shown is an architectural schematic of a field rice planthopper survey system based on AR glasses, provided in an embodiment of this application. Figure 2 As shown, the server can establish communication connections with both the AR glasses and the mobile terminal corresponding to the client.

[0081] The AR glasses are worn by the user and, based on the user's voice commands, acquire images of a white disc in the field. This white disc image can be understood as the image captured by the AR glasses of a white disc (coated with a nano-coating or made of non-reflective white plastic) where the user pats the rice plants to cause various types of rice planthoppers to fall onto the disc. All the rice planthoppers on the white disc are live insects. It is understood that, in the embodiments of this application, the AR glasses may specifically include an eyeglass frame, a camera, a monocular display, a power board, a main control device, a communication module, a voice control module, and an integrated circuit unit. The camera, monocular display, main control device, communication module, voice control module, and integrated circuit unit are all installed inside the AR glasses. The main control device is connected to the camera, monocular display, communication module, and voice control module respectively through the integrated circuit unit. The eyeglass frame is for the user to wear and can be fitted with different prescription lenses or sunglasses. The camera is used to capture images of rice planthoppers in a white dish. The monocular display is used to display the images captured by the camera to the user in real time. The voice control module is used to perform operations such as taking photos, recording videos, and zooming in and out of objects according to the user's voice control, thereby achieving "hands-free" operation. The power board provides power to the camera and AR imaging system. The communication module is used to establish a communication connection with the server to transmit the captured images to the server. The main control device is used to control the operation and shutdown of all components in the AR glasses. The integrated circuit unit is used to connect different modules inside the glasses' camera.

[0082] The server can receive white disk images captured by AR glasses, transmit these images to a trained convolutional neural network model, and use the model to identify different types of rice planthoppers in the images. It can also perform statistical processing on the different types of planthoppers, and send the planthopper types, their numbers, and the white disk image to the corresponding mobile terminal of the client, allowing users to quickly view the information on their mobile devices. Specifically, the server may include a rice planthopper identification algorithm module, a business application layer software system, and an IoT communication module. The rice planthopper identification algorithm module is connected to the IoT communication module for communication between the AR glasses and the AI ​​artificial intelligence system platform. This module can also be connected to the business application layer software system. The IoT communication module can connect to the processor of the AI ​​artificial intelligence system platform, the IoT communication platform, and agricultural IoT device terminals. Here, the rice planthopper identification algorithm module can, but is not limited to, accurately identify and count 15 indicators, including, but not limited to, long-winged adults of white-backed planthoppers, short-winged adults of white-backed planthoppers, older nymphs of white-backed planthoppers, long-winged adults of brown planthoppers, short-winged adults of brown planthoppers, older nymphs of brown planthoppers, long-winged adults of gray planthoppers, short-winged adults of gray planthoppers, older nymphs of gray planthoppers, young nymphs of rice planthoppers, spiders, stink bugs, rove beetles, adult leafhoppers, and leafhopper nymphs.

[0083] The mobile terminal corresponding to the client can, but is not limited to, having a rice planthopper monitoring and survey app installed. This app is used to receive images of different types of rice planthoppers, their quantities, and white plate images sent by the server, and to display these images to the user. Here, the user can log into their account on the rice planthopper monitoring and survey app and receive images of different types of rice planthoppers, their quantities, and white plate images sent by the server, either manually or automatically. It is understood that the mobile terminal can also obtain real-time time, weather, and geographic information, and users can manually enter information about the field, rice variety, and survey personnel.

[0084] Specifically, during a survey of rice planthoppers in the field, the server receives white disk images captured by AR glasses. During the image acquisition process, the AR glasses can be used by a user, who can first wear them and tap the rice paddies to cause various types of planthoppers to fall into a white disk held by the user. Then, the AR glasses' camera can be pointed at the white disk, and the user can control the AR glasses to capture images via voice commands. It is understood that the AR glasses can also be used to adjust the shooting direction by the user turning their head, or automatically adjust the shooting direction and image capture via voice commands. Furthermore, the real-time captured white disk images can be displayed to the user on the AR glasses' monocular display, to meet the user's shooting needs.

[0085] As an optional embodiment of this application, acquiring the white disk image captured by AR glasses includes:

[0086] Send a location request command to the AR glasses to obtain the current location fed back by the AR glasses;

[0087] When the current location is detected to be within the specified area, an image request command is generated;

[0088] The image request command is sent to the AR glasses, which then generate corresponding prompts for the user based on the image request command. The AR glasses also capture the white disk image based on the user's first voice command.

[0089] Acquire the white disk image captured by the AR glasses.

[0090] Specifically, during the process of acquiring the white disk image captured by the AR glasses, the server can first send a location request command to the AR glasses to obtain the current location of the user wearing the AR glasses. Here, after receiving the location request command, the AR glasses can, but are not limited to, obtain the current location through positioning technology and promptly feed the current location back to the server. It is understood that, in this embodiment of the application, the server can also, but is not limited to, send a location request command to the mobile terminal corresponding to the user, where the mobile terminal corresponding to the user is in the same location as the AR glasses worn by the user.

[0091] Furthermore, after receiving the current location from the AR glasses, the server can determine whether the current location is within a designated area, which can be understood as a rice-growing region. If the server detects that the current location is within the designated area, it indicates that the user wearing the AR glasses is currently in a rice-growing area, and the server can proceed with the subsequent operation of capturing a white disk image, thereby generating an image request command to send to the AR glasses. Conversely, if the server detects that the current location is not within the designated area, it indicates that the user wearing the AR glasses is not yet in a rice-growing area, and the server needs to promptly remind the user to move to the appropriate area to avoid capturing invalid white disk images.

[0092] Furthermore, after sending the image request command to the AR glasses, the AR glasses can generate corresponding prompts based on the image request command. These prompts can be displayed to the user through the AR glasses' monocular display, but are not limited to, reminding the user to control the AR glasses to take pictures using the first voice command. In this embodiment, the first voice command can be, but is not limited to, voice commands such as "start taking pictures," "start," or "take pictures," indicating the shooting process. After receiving the first voice command from the user, the AR glasses can perform voice recognition processing and acquire a white disk image based on the recognition result. In this embodiment, the method of generating prompts through the AR glasses and controlling the AR glasses to acquire white disk images based on the user's voice commands avoids errors caused by excessive manual operation (such as manually controlling the AR glasses to take pictures or adjust them), thereby improving the efficiency of white disk image acquisition.

[0093] It should be noted that after the AR glasses capture the white disk image, the AR glasses can also synchronize the white disk image on the monocular display so that the user can judge whether the white disk image is valid. Here, the user can judge whether the white disk image is valid by judging whether there are rice planthoppers in the white disk image from a direct viewing angle, or by the AR glasses automatically executing a program to judge whether there are rice planthoppers in the white disk image through image recognition. It is not limited to these methods.

[0094] Furthermore, after the AR glasses capture a white disk image based on the user's first voice command, the server can receive the white disk image captured by the AR glasses.

[0095] As another optional embodiment of this application, after acquiring the white disk image captured by the AR glasses, the method further includes:

[0096] Determine whether the edge of the white disk in the white disk image is within the display screen of the AR glasses;

[0097] When it is detected that part of the white disk edge is not within the display screen of the AR glasses, zoom prompt information is generated based on the position of the white disk in the white disk image;

[0098] The zoom prompt information is sent to the AR glasses, which then display the zoom prompt information to the user through the screen. The AR glasses also perform optical zoom processing based on the user's second voice command and capture a first image. The edge of the white disk in the first image is within the display screen of the AR glasses.

[0099] Acquire the first image captured by the AR glasses and replace the white disk image with the first image.

[0100] To ensure the validity of the white disk image, after acquiring the white disk image captured by the AR glasses, the server can, but is not limited to, use image recognition technology to determine whether the edge of the white disk in the white disk image is within the display screen of the AR glasses. Here, the position of the AR glasses' display screen can be preset or planned within the server. After receiving the white disk image, the server can import the white disk image into a designated area of ​​the display screen containing the AR glasses to determine whether the entire edge of the white disk in the white disk image is within the display screen of the AR glasses.

[0101] Furthermore, when it is detected that part of the white disk's edge is not within the AR glasses' display screen, it indicates that the shooting angle or focus of the white disk image is incorrect. In this case, the server can combine the white disk's position in the white disk image to generate corresponding zoom prompt information. For example, but not limited to, when the white disk's position in the white disk image is in the lower right corner, the generated zoom prompt information can indicate zooming towards the lower right corner. After the AR glasses receive this zoom prompt information, they can automatically control the camera to perform optical zoom towards the lower right corner. Of course, the zoom prompt information can also include the specific coordinates of the white disk's position in the white disk image, which will not be elaborated on here.

[0102] Furthermore, after the server sends the zoom prompt information to the AR glasses, the AR glasses can display the zoom prompt information to the user via a monocular display to remind the user to issue a second voice command to control the AR glasses to capture the first image. This second voice command can be, but is not limited to, commands such as "zoom back" or "zoom in the xx direction," which are used to characterize optical zoom. After receiving the second voice command from the user, the AR glasses can perform voice recognition processing and, based on the recognition result, control the camera to re-zoom optically to acquire the white disk image again. It is understood that the first image can be a white disk image whose entire edge is within the AR glasses' display screen; alternatively, it can be a white disk image where half of the white disk is within the AR glasses' display screen. This is not a limitation.

[0103] In this embodiment of the application, the method of generating prompt information through AR glasses and controlling the AR glasses to perform optical zoom and acquire white disk images according to the voice commands issued by the user can avoid errors caused by excessive manual operation (such as manually controlling the AR glasses to shoot or adjust), thereby improving the acquisition efficiency of white disk images.

[0104] Furthermore, after the AR glasses capture the first image, the server can receive the first image and replace the white disk image with the first image to ensure the validity of the white disk image that needs to be transmitted to the convolutional neural network model for recognition.

[0105] As another optional embodiment of this application, after acquiring the white disk image captured by the AR glasses, the method further includes:

[0106] Gaussian blurring is applied to the white disk image to obtain a blurred image corresponding to the white disk image;

[0107] Calculate the first gray level difference between any two adjacent pixels in the white image and the second gray level difference between any two adjacent pixels in the blurred image.

[0108] The first grayscale difference and the second grayscale difference are normalized, and when the processing result is detected to be within a preset range, an image is generated to obtain prompt information.

[0109] The image acquisition prompt is sent to the AR glasses, which then display the prompt to the user on the screen. The AR glasses also acquire a second image based on the user's third voice command.

[0110] Acquire the second image captured by the AR glasses and replace the white disk image with the second image.

[0111] To ensure the clarity of the white disk image, after acquiring the white disk image captured by the AR glasses, the server can, but is not limited to, blurring the white disk image to obtain a blurred image corresponding to it. This blurring can, but is not limited to, filtering the white disk image based on different types of filters. It is understandable that if an image is already blurred, then blurring it again will not significantly change the high-frequency components; conversely, if an image is clear, then filtering it will significantly change the high-frequency components.

[0112] Furthermore, after obtaining the blurred image, the server can calculate the first grayscale difference between any two adjacent pixels in the white image (i.e., the grayscale difference between any two adjacent pixels in the white image), and the second grayscale difference between any two adjacent pixels in the blurred image (i.e., the grayscale difference between any two adjacent pixels in the blurred image). Then, according to the corresponding relationship, the first grayscale difference and the second grayscale difference corresponding to the two adjacent pixels are normalized. When the processing result is detected to be within a preset range, it indicates that the white image is relatively blurred, meaning the processing result is large, and an image acquisition prompt can be generated. Here, the preset range can be, but is not limited to, between 0.6 and 1. When the processing result is detected to be below this preset range, it indicates that the white image is relatively clear, meaning the processing result is small, and the white image can be used directly.

[0113] Furthermore, after generating the image acquisition prompt information, the server can send the image acquisition prompt information to the AR glasses, which can then display the prompt information to the user via a monocular display, reminding the user to control the AR glasses to take pictures using a third voice command. In this embodiment, the first voice command can be, but is not limited to, voice commands such as "take a picture again," "start," or "take a picture" to indicate the shooting action. After receiving the third voice command from the user, the AR glasses can perform voice recognition processing and acquire the white disk image based on the recognition result. In this embodiment, the method of generating prompt information through the AR glasses and controlling the AR glasses to acquire white disk images based on the user's voice commands avoids errors caused by excessive manual operation (such as manually controlling the AR glasses to take pictures or adjust them), thereby improving the efficiency of white disk image acquisition.

[0114] Furthermore, after the AR glasses capture the second image, the server can receive the second image and replace the white disk image with the second image to ensure the validity of the white disk image that needs to be transmitted to the convolutional neural network model for recognition.

[0115] Step 104: Transmit the white disk image to the convolutional neural network model to obtain a white disk image labeled with the rice planthopper type.

[0116] Specifically, after receiving the white disk image captured by the AR glasses, the server can transmit the white disk image to a trained convolutional neural network model, so that the convolutional neural network model can predict the white disk image labeled with the rice planthopper type. Here, the white disk image labeled with the rice planthopper type can be understood as the white disk image with the outline of each rice planthopper and the type corresponding to each rice planthopper marked. The convolutional neural network model can be, but is not limited to, being trained from sample images of multiple rice planthopper types manually labeled. In this embodiment, the predicted target of the convolutional neural network model and the real target are separated by Gaussian Wasserstein distance, and the accuracy and recall of different types of rice planthopper identification are used to evaluate the convolutional neural network model and select the optimal convolutional neural network model.

[0117] Here, the multiple types of rice planthoppers manually labeled can be, but are not limited to, any two of the following: white-backed planthopper, brown planthopper, gray planthopper, long-winged type, short-winged type, older nymphs, and younger nymphs. The sample image can be an image of rice planthoppers in white discs collected from the field under various lighting conditions.

[0118] See here. Figure 3 The diagram shown illustrates the prediction performance of a convolutional neural network model provided in an embodiment of this application. Figure 3As shown in the diagram, the white disc image is marked with different types of rice planthoppers, and each type of rice planthopper is marked differently, for example, but not limited to different shapes or sizes of the marks. This is not a limitation here.

[0119] As another optional embodiment of this application, the white disk image is transmitted to a convolutional neural network model to obtain a white disk image labeled with the rice planthopper type, including:

[0120] The white disk image is divided into blocks to obtain m sub-images with the same area; where m is a positive integer greater than or equal to 4.

[0121] Each sub-image is fed into a convolutional neural network model to obtain sub-images labeled with the rice planthopper type;

[0122] Image synthesis processing was performed on all sub-images labeled with the rice planthopper type to obtain white disc images labeled with the rice planthopper type.

[0123] To further improve the prediction accuracy of the convolutional neural network model, the server can divide the white disk image into blocks to obtain m sub-images of equal area. Each sub-image is then transmitted to the convolutional neural network model to predict each sub-image labeled with the rice planthopper type. The server can, but is not limited to, setting a fixed block window, with a block size of 1120 pixels × 868 pixels. Furthermore, during the block division of the white disk image, there is a 50-pixel overlap between any two adjacent sub-images to improve stitching accuracy and overall efficiency.

[0124] Furthermore, after obtaining each sub-image labeled with the rice planthopper type, the server can also perform image synthesis processing on all sub-images labeled with the rice planthopper type to stitch all sub-images labeled with the rice planthopper type into a white plate image labeled with the rice planthopper type.

[0125] It should be noted that, in the embodiments of this application, the convolutional neural network model can be trained by, but is not limited to, sample sub-images of multiple rice planthopper types manually labeled. In the embodiments of this application, the predicted target and the real target of the convolutional neural network model are separated by Gaussian Wasserstein distance, and the accuracy and recall of different types of rice planthoppers are used to evaluate the convolutional neural network model and select the optimal convolutional neural network.

[0126] Step 106: Perform statistical processing on the white disk images marked with rice planthopper types to obtain the quantity of each rice planthopper type, and send the white disk images, each rice planthopper type and the corresponding quantity to the client so that the client can display each rice planthopper type and the corresponding quantity in the white disk images.

[0127] Specifically, after predicting the white disk images marked with rice planthopper types, the server can perform statistical processing on the white disk images marked with rice planthopper types to obtain the quantity of each type of rice planthopper, and send the white disk images, each type of rice planthopper and the corresponding quantity to the client, so that the mobile terminal corresponding to the client can display them to the user on the corresponding rice planthopper monitoring and survey APP.

[0128] As another optional embodiment of this application, after statistically processing the white disc images marked with rice planthopper types to obtain the number of each rice planthopper type, the method further includes:

[0129] The current time corresponding to the AR glasses is obtained, and a change curve is generated based on the current location, the current time, the number of each type of rice planthopper, and historical records; wherein, the historical records include at least two historical times corresponding to the current location, and the number of each type of rice planthopper in the white disk image corresponding to each historical time.

[0130] The change curve is sent to the client so that the user can view the change curve on the client.

[0131] To facilitate users' more intuitive observation and understanding of rice planthoppers, after obtaining the number of each type of rice planthopper, the server obtains the current time corresponding to the AR glasses, and combines the current location, current time, number of each type of rice planthopper, and historical records to generate a change curve. This change curve can be used to characterize the change in the number of each type of rice planthopper at different times at the same moment, so that users can quickly make corresponding predictions and judgments while observing, and ensure data traceability.

[0132] Understandably, after generating the change curve, the server can send the change curve to the client so that the user can view it at any time on the mobile terminal corresponding to that client.

[0133] See here. Figure 4 The illustrated diagram shows a client display interface provided in an embodiment of this application, such as... Figure 4 As shown in the diagram, the display interface, from left to right, represents the login interface, information selection interface, white disk image display interface, and query display interface of the rice planthopper monitoring and survey app. It can be understood that after logging in to the rice planthopper monitoring and survey app, users can filter and select information on the information selection interface to view the generated white disk image on the white disk image display interface. In addition, they can use the query display interface to query the change curves within a specified time interval.

[0134] Please see Figure 5 , Figure 5A schematic diagram of a field rice planthopper survey device based on AR glasses, provided in an embodiment of this application, is shown.

[0135] like Figure 5 As shown, the AR glasses-based field rice planthopper survey device may include at least an image acquisition module 501, an image processing module 502, and an image display module 503, wherein:

[0136] Image acquisition module 501 is used to acquire the white disk image captured by AR glasses;

[0137] Image processing module 502 is used to transmit the white disk image to the convolutional neural network model to obtain the white disk image labeled with the rice planthopper type; wherein, the convolutional neural network model is trained from multiple sample images labeled with the rice planthopper type;

[0138] The image display module 503 is used to perform statistical processing on white disk images marked with rice planthopper types, obtain the quantity of each rice planthopper type, and send the white disk images, each rice planthopper type and the corresponding quantity to the client so that the client can display each rice planthopper type and the corresponding quantity in the white disk images.

[0139] In some possible embodiments, the image acquisition module is specifically used for:

[0140] Send a location request command to the AR glasses to obtain the current location fed back by the AR glasses;

[0141] When the current location is detected to be within the specified area, an image request command is generated;

[0142] The image request command is sent to the AR glasses, which then generate corresponding prompts for the user based on the image request command. The AR glasses also capture the white disk image based on the user's first voice command.

[0143] Acquire the white disk image captured by the AR glasses.

[0144] In some possible embodiments, the image acquisition module is further used for:

[0145] After acquiring the white disk image captured by the AR glasses, determine whether the edge of the white disk in the white disk image is within the display screen of the AR glasses;

[0146] When it is detected that part of the white disk edge is not within the display screen of the AR glasses, zoom prompt information is generated based on the position of the white disk in the white disk image;

[0147] The zoom prompt information is sent to the AR glasses, which then display the zoom prompt information to the user through the screen. The AR glasses also perform optical zoom processing based on the user's second voice command and capture a first image. The edge of the white disk in the first image is within the display screen of the AR glasses.

[0148] Acquire the first image captured by the AR glasses and replace the white disk image with the first image.

[0149] In some possible embodiments, the image acquisition module is further used for:

[0150] After acquiring the white disk image captured by the AR glasses, the white disk image is subjected to Gaussian blur processing to obtain a blurred image corresponding to the white disk image;

[0151] Calculate the first gray level difference between any two adjacent pixels in the white image and the second gray level difference between any two adjacent pixels in the blurred image.

[0152] The first grayscale difference and the second grayscale difference are normalized, and when the processing result is detected to be within a preset range, an image is generated to obtain prompt information.

[0153] The image acquisition prompt is sent to the AR glasses, which then display the prompt to the user on the screen. The AR glasses also acquire a second image based on the user's third voice command.

[0154] Acquire the second image captured by the AR glasses and replace the white disk image with the second image.

[0155] In some possible embodiments, the image processing module is specifically used for:

[0156] The white disk image is divided into blocks to obtain m sub-images with the same area; where m is a positive integer greater than or equal to 4.

[0157] Each sub-image is fed into a convolutional neural network model to obtain sub-images labeled with the rice planthopper type;

[0158] Image synthesis processing was performed on all sub-images labeled with the rice planthopper type to obtain white disc images labeled with the rice planthopper type.

[0159] In some possible embodiments, the apparatus further includes:

[0160] After statistically processing the white disk images labeled with rice planthopper types to obtain the number of each type, the current time corresponding to the AR glasses is obtained, and a change curve is generated based on the current location, the current time, the number of each type of rice planthopper, and historical records. The historical records include at least two historical times corresponding to the current location, and the number of each type of rice planthopper in the white disk image corresponding to each historical time.

[0161] The change curve is sent to the client so that the user can view the change curve on the client.

[0162] In some possible embodiments, the sample images labeled with rice planthopper types include any one of the following: long-winged adult white-backed planthopper, short-winged adult white-backed planthopper, older nymph white-backed planthopper, long-winged adult brown planthopper, short-winged adult brown planthopper, older nymph brown planthopper, long-winged adult gray planthopper, short-winged adult gray planthopper, older nymph gray planthopper, young nymph rice planthopper, spider, stink bug, rove beetle, leafhopper adult, and leafhopper nymph.

[0163] Those skilled in the art will clearly understand that the technical solutions of the embodiments of this application can be implemented by means of software and / or hardware. In this specification, "unit" and "module" refer to software and / or hardware that can independently complete or cooperate with other components to complete a specific function, wherein the hardware may be, for example, a field-programmable gate array (FPGA), an integrated circuit (IC), etc.

[0164] Please see Figure 6 , Figure 6 A schematic diagram of the structure of a server provided in an embodiment of this application is shown.

[0165] like Figure 6 As shown, the server 600 may include at least a processor 601, at least one network interface 604, a user interface 603, a memory 605, and at least one communication bus 602.

[0166] The communication bus 602 can be used to realize the connection and communication of the above components.

[0167] The user interface 603 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.

[0168] The network interface 604 may include, but is not limited to, Bluetooth modules, NFC modules, Wi-Fi modules, etc.

[0169] The processor 601 may include one or more processing cores. The processor 601 connects to various parts within the server 600 via various interfaces and lines, and performs various functions of the routing server 600 and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 605, and by calling data stored in the memory 605. Optionally, the processor 601 may be implemented using at least one hardware form of DSP, FPGA, or PLA. The processor 601 may integrate one or more of the following: CPU, GPU, and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 601 and may be implemented as a separate chip.

[0170] The memory 605 may include RAM or ROM. Optionally, the memory 605 may include a non-transitory computer-readable medium. The memory 605 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 605 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 605 may also be at least one storage device located remotely from the aforementioned processor 601. Figure 6 As shown, the memory 605, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an AR glasses-based field rice planthopper survey application.

[0171] Specifically, the processor 601 can be used to invoke the AR glasses-based field rice planthopper survey application stored in the memory 605, and specifically perform the following operations:

[0172] Acquire the white disk image captured by the AR glasses;

[0173] The white disk images are transmitted to a convolutional neural network model to obtain white disk images labeled with rice planthopper types; the convolutional neural network model is trained from multiple sample images labeled with rice planthopper types;

[0174] Statistical processing is performed on white disk images labeled with rice planthopper types to obtain the quantity of each rice planthopper type. The white disk images, each rice planthopper type, and the corresponding quantity are then sent to the client so that the client can display each rice planthopper type and the corresponding quantity in the white disk images.

[0175] In some possible embodiments, acquiring the white disk image captured by the AR glasses includes:

[0176] Send a location request command to the AR glasses to obtain the current location fed back by the AR glasses;

[0177] When the current location is detected to be within the specified area, an image request command is generated;

[0178] The image request command is sent to the AR glasses, which then generate corresponding prompts for the user based on the image request command. The AR glasses also capture the white disk image based on the user's first voice command.

[0179] Acquire the white disk image captured by the AR glasses.

[0180] In some possible embodiments, after acquiring the white disk image captured by the AR glasses, the method further includes:

[0181] Determine whether the edge of the white disk in the white disk image is within the display screen of the AR glasses;

[0182] When it is detected that part of the white disk edge is not within the display screen of the AR glasses, zoom prompt information is generated based on the position of the white disk in the white disk image;

[0183] The zoom prompt information is sent to the AR glasses, which then display the zoom prompt information to the user through the screen. The AR glasses also perform optical zoom processing based on the user's second voice command and capture a first image. The edge of the white disk in the first image is within the display screen of the AR glasses.

[0184] Acquire the first image captured by the AR glasses and replace the white disk image with the first image.

[0185] In some possible embodiments, after acquiring the white disk image captured by the AR glasses, the method further includes:

[0186] Gaussian blurring is applied to the white disk image to obtain a blurred image corresponding to the white disk image;

[0187] Calculate the first gray level difference between any two adjacent pixels in the white image and the second gray level difference between any two adjacent pixels in the blurred image.

[0188] The first grayscale difference and the second grayscale difference are normalized, and when the processing result is detected to be within a preset range, an image is generated to obtain prompt information.

[0189] The image acquisition prompt is sent to the AR glasses, which then display the prompt to the user on the screen. The AR glasses also acquire a second image based on the user's third voice command.

[0190] Acquire the second image captured by the AR glasses and replace the white disk image with the second image.

[0191] In some possible embodiments, the white disk image is transmitted to a convolutional neural network model to obtain a white disk image labeled with the rice planthopper type, including:

[0192] The white disk image is divided into blocks to obtain m sub-images with the same area; where m is a positive integer greater than or equal to 4.

[0193] Each sub-image is fed into a convolutional neural network model to obtain sub-images labeled with the rice planthopper type;

[0194] Image synthesis processing was performed on all sub-images labeled with the rice planthopper type to obtain white disc images labeled with the rice planthopper type.

[0195] In some possible embodiments, after statistically processing the white disc images labeled with rice planthopper types to obtain the number of each rice planthopper type, the method further includes:

[0196] The current time corresponding to the AR glasses is obtained, and a change curve is generated based on the current location, the current time, the number of each type of rice planthopper, and historical records; wherein, the historical records include at least two historical times corresponding to the current location, and the number of each type of rice planthopper in the white disk image corresponding to each historical time.

[0197] The change curve is sent to the client so that the user can view the change curve on the client.

[0198] In some possible embodiments, the sample images labeled with rice planthopper types include any one of the following: long-winged adult white-backed planthopper, short-winged adult white-backed planthopper, older nymph white-backed planthopper, long-winged adult brown planthopper, short-winged adult brown planthopper, older nymph brown planthopper, long-winged adult gray planthopper, short-winged adult gray planthopper, older nymph gray planthopper, young nymph rice planthopper, spider, stink bug, rove beetle, leafhopper adult, and leafhopper nymph.

[0199] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0200] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0201] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0202] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0203] The units described as separate components may or may not be physically separate. The components shown as units 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0204] Furthermore, the functional units in the various embodiments of this application 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 as a software functional unit.

[0205] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0206] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0207] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of embodiments of this disclosure upon considering the specification and practicing the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for surveying rice planthoppers in the field based on AR glasses, characterized in that, include: Acquire the white disk image captured by the AR glasses; The white disk image is transmitted to a convolutional neural network model to obtain white disk images labeled with rice planthopper types; wherein, the convolutional neural network model is trained from multiple sample images labeled with rice planthopper types; the rice planthopper types in the sample images labeled with rice planthopper types include any at least one of the following: white-backed planthopper long-winged adult, white-backed planthopper short-winged adult, white-backed planthopper late-stage nymph, brown planthopper long-winged adult, brown planthopper short-winged adult, brown planthopper late-stage nymph, gray planthopper long-winged adult, gray planthopper short-winged adult, gray planthopper late-stage nymph, rice planthopper early-stage nymph, spider, stink bug, rove beetle, leafhopper adult, and leafhopper nymph; the predicted target and the real target of the convolutional neural network model are separated by Gaussian Wasserstein distance; The white disk images labeled with rice planthopper types are statistically processed to obtain the quantity of each rice planthopper type. The white disk images, each rice planthopper type, and the corresponding quantity are then sent to the client so that the client can display each rice planthopper type and the corresponding quantity in the white disk images. The white disc image is transmitted to a convolutional neural network model to obtain a white disc image labeled with the rice planthopper type, including: The white disk image is divided into blocks to obtain m sub-images with the same area; where m is a positive integer greater than or equal to 4; the block size of each sub-image is 1120 pixels × 868 pixels, and there is a 50-pixel overlap between any two adjacent sub-images; Each of the sub-images is transmitted to a convolutional neural network model to obtain the sub-images labeled with the rice planthopper type; Image synthesis processing is performed on all the sub-images labeled with the rice planthopper type to obtain the white disc image labeled with the rice planthopper type.

2. The method according to claim 1, characterized in that, The acquisition of the white disk image captured by the AR glasses includes: Send a location request command to the AR glasses to obtain the current location fed back by the AR glasses; When the current location is detected to be within the specified area, an image request command is generated; The image request instruction is sent to the AR glasses, so that the AR glasses generate corresponding prompt information to the user based on the image request instruction, and the AR glasses capture the white disk image based on the user's first voice instruction; Acquire the white disk image captured by the AR glasses.

3. The method according to claim 2, characterized in that, After acquiring the white disk image captured by the AR glasses, the method further includes: Determine whether the edge of the white disk in the white disk image is within the display screen of the AR glasses; When it is detected that part of the white disk edge is not within the display screen of the AR glasses, zoom prompt information is generated based on the position of the white disk in the white disk image; The zoom prompt information is sent to the AR glasses, so that the AR glasses can display the zoom prompt information to the user through the display screen, and the AR glasses can perform optical zoom processing and capture a first image according to the user's second voice command; wherein, the edge of the white disk in the first image is within the display screen of the AR glasses; The first image captured by the AR glasses is acquired, and the white disk image is replaced with the first image.

4. The method according to claim 2, characterized in that, After acquiring the white disk image captured by the AR glasses, the method further includes: Gaussian blur processing is applied to the white disk image to obtain a blurred image corresponding to the white disk image; Calculate the first grayscale difference between any two adjacent pixels in the white disk image and the second grayscale difference between any two adjacent pixels in the blurred image. The first grayscale difference and the second grayscale difference are normalized, and when the processing result is detected to be within a preset range, an image is generated to obtain prompt information. The image acquisition prompt is sent to the AR glasses, so that the AR glasses can display the image acquisition prompt to the user through the display screen, and the AR glasses can acquire a second image according to the user's third voice command; The second image captured by the AR glasses is acquired, and the white disk image is replaced with the second image.

5. The method according to claim 2, characterized in that, After statistically processing the white disc images labeled with rice planthopper types to obtain the quantity of each rice planthopper type, the method further includes: The current time corresponding to the AR glasses is obtained, and a change curve is generated based on the current location, the current time, the number of each type of rice planthopper, and historical records; wherein, the historical records include at least two historical times corresponding to the current location, and the number of each type of rice planthopper in the white disk image corresponding to each historical time. The change curve is sent to the client so that the user can view the change curve on the client.

6. A field rice planthopper survey device based on AR glasses, characterized in that, include: The image acquisition module is used to acquire the white disk image captured by the AR glasses; An image processing module is used to transmit the white disk image to a convolutional neural network model to obtain the white disk image labeled with rice planthopper types; wherein, the convolutional neural network model is trained from multiple sample images labeled with rice planthopper types; the rice planthopper types in the sample images labeled with rice planthopper types include any at least one of the following: white-backed planthopper long-winged adult, white-backed planthopper short-winged adult, white-backed planthopper late-stage nymph, brown planthopper long-winged adult, brown planthopper short-winged adult, brown planthopper late-stage nymph, gray planthopper long-winged adult, gray planthopper short-winged adult, gray planthopper late-stage nymph, rice planthopper early-stage nymph, spider, stink bug, rove beetle, leafhopper adult, and leafhopper nymph; the predicted target and the real target of the convolutional neural network model are separated by Gaussian Wasserstein distance; The image display module is used to perform statistical processing on the white plate image marked with rice planthopper types, obtain the quantity of each rice planthopper type, and send the white plate image, each rice planthopper type and the corresponding quantity to the client, so that the client can display each rice planthopper type and the corresponding quantity in the white plate image; The white disc image is transmitted to a convolutional neural network model to obtain a white disc image labeled with the rice planthopper type, including: The white disk image is divided into blocks to obtain m sub-images with the same area; where m is a positive integer greater than or equal to 4; the block size of each sub-image is 1120 pixels × 868 pixels, and there is a 50-pixel overlap between any two adjacent sub-images; Each of the sub-images is transmitted to a convolutional neural network model to obtain the sub-images labeled with the rice planthopper type; Image synthesis processing is performed on all the sub-images labeled with the rice planthopper type to obtain the white disc image labeled with the rice planthopper type.

7. A field rice planthopper survey device based on AR glasses, characterized in that, Including the processor and memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code stored in the memory to perform the steps of the method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer-readable storage medium stores instructions that, when executed on a computer or processor, cause the computer or processor to perform the steps of the method as described in any one of claims 1-5.

Images