Field pest quantity estimation method based on moth expelling operation

Abstract

The embodiment of the invention discloses a field pest quantity estimation method based on moth expelling operation. The method comprises the steps that after a number estimation instruction is detected, video data corresponding to one-time complete moth expelling operation collected by a camera is acquired, and a moth area, a gradienter area and a compass area in the video data are determined; identifying bubble positions of gradienters in the gradienter area, and verifying the validity of the video data based on the bubble positions; and after the video data is verified to be valid, identifying the number of regional moths in the moth region, determining a sweeping area corresponding to the complete moth driving operation according to the angle change of the compass in the compass region, and determining an estimated number of field pests according to the number of regional moths and the sweeping area. According to the embodiment of the invention, only the video data subjected to validity verification can be used for quantity estimation, the moth quantity and the sweeping area are determined through image recognition instead of being judged by artificial experience, and the finally estimated pest quantity is more accurate.

Description

Technical Field

[0001] One or more embodiments of the present specification relate to image recognition technology, and more particularly to a method for estimating the number of field pests based on moth-chasing operations. Background Art

[0002] Pests that harm rice, such as the rice leaf roller, have the habit of long-distance migration. They migrate from Southeast Asia and other places to domestic rice-growing areas with seasonal changes. Conducting surveys on the occurrence of pests in the field is the basis for prediction and forecasting, as well as field prevention and control. The current survey of the occurrence of field pests still uses a purely manual moth-driving estimation method, that is, manually holding a bamboo pole and slowly moving the upper and middle parts of the rice bushes along the ridge, and using a counter to record the number of moths that fly up during the shaking process, and then combining this with the area of ​​shaking to estimate the overall moth quantity in the field. However, in this process, when manually moving the rice bushes, it is not easy for the moth-driving pole to remain completely horizontal, and when using fixed-point moth-driving, the rotation angle may be different each time, resulting in a large gap between the actual shaking area and the visually estimated area, and the movement state of the moth-driving pole is usually not recorded. Combined with the fact that the number of moths that fly up during the shaking process is also manually visually observed, the final estimated number of moths has a large error, which affects the judgment of the actual occurrence of pests in the field. Summary of the invention

[0003] In order to solve the above problems, one or more embodiments of this specification describe a method for estimating the number of field pests based on moth-chasing operations.

[0004] According to a first aspect, a method for estimating the number of field pests based on a moth-chasing operation is provided, which is applied to a moth-chasing device, the device comprising a camera component and a moth-chasing rod component, the camera component comprising a camera support rod and a camera for shooting from a top-down angle, the moth-chasing rod component comprising a moth-chasing rod, a clamping module for clamping the moth-chasing rod, a compass and a level respectively arranged on the moth-chasing rod, the method comprising: After detecting the quantity estimation instruction, obtaining video data corresponding to a complete moth-chasing operation collected by the camera, and determining the moth area, the level area, and the compass area in the video data; identifying a bubble position of the level in the level area, and verifying the validity of the video data based on the bubble position; After the video data is verified to be valid, the number of regional moths in the moth area is identified, and the swept area corresponding to the complete moth-chasing operation is determined according to the angle change between the compass in the compass area and the moth-chasing rod, and the estimated number of field pests is determined according to the number of regional moths and the swept area.

[0005] Preferably, after the quantity estimation instruction is detected, video data corresponding to a complete moth-chasing operation collected by a camera is obtained, including: After detecting the survey information released by the target institution, determining the field area to be surveyed based on the rice growth stage, rice variety and rice growth, and determining the regional coordinates and survey time of each field area to be surveyed, wherein the survey time includes the survey date and survey period; When any area coordinate that meets the survey time is reached and a quantity estimation instruction is detected, video data corresponding to a complete moth-chasing operation captured by the camera is obtained.

[0006] Preferably, the step of obtaining video data corresponding to a complete moth-chasing operation collected by a camera includes: Identify the moth-driving rod in the real-time image of the camera, and when the moth-driving rod does not move within a preset time, control the camera to start collecting video data; When the moth-repelling rod does not move again within a preset time period after moving, the camera is controlled to stop collecting the video data.

[0007] Preferably, the determining of the moth area, the level area and the compass area in the video data comprises: Divide the video data into frames according to the time sequence to obtain images of each frame; Based on image recognition, the moth area, the level area and the compass area in each of the frame images are respectively determined and marked.

[0008] Preferably, the identifying the bubble position of the level in the level area and verifying the validity of the video data based on the bubble position includes: Based on a preset first image recognition model, the level area of ​​each of the framed images is respectively identified to obtain each first recognition result, wherein the first image recognition model is used to identify the bubble position of the level in the image, and the first recognition result includes the bubble being centered, the bubble being biased toward the palm direction, or the bubble being away from the palm direction; Calculate the result ratio of the target first recognition result among all the first recognition results, and obtain the validity verification result of the video data based on the result ratio, wherein the target first recognition result is the first recognition result characterized by the bubble being centered, and the validity verification result is characterized as valid when the result ratio is higher than a preset ratio.

[0009] Preferably, the identifying the number of moths in the moth area includes: Based on a preset second image recognition model, respectively identify the moth region of each of the frame images to obtain each second recognition result, wherein the second image recognition model is used to identify the moth in the image; Adjacent second recognition results are matched based on a multi-target tracking algorithm, and the number of moths in the moth area is determined according to the number of matched moth tracks.

[0010] Preferably, the method of matching adjacent second recognition results based on a multi-target tracking algorithm and determining the number of moths in the moth area according to the number of matched moth tracks includes: After enlarging the target frame of each second recognition result based on a preset magnification, generating at least one prediction frame of each target frame in the next adjacent second recognition result based on a Kalman filter; Matching the predicted box and the target box in the next second recognition result based on the Hungarian algorithm to determine the best matching result; Repeat the step of generating at least one prediction frame for each target frame in the next adjacent second recognition result based on the Kalman filter until all optimal matching results are obtained, generate the moth trajectory of each moth object according to the optimal matching results, and determine the number of moths in the moth area according to the number of moth trajectories.

[0011] Preferably, the step of determining the swept area corresponding to the complete moth-chasing operation according to the angle change between the compass in the compass area and the moth-chasing rod comprises: Based on a preset third image recognition model, the compass area of ​​each frame image is respectively identified to obtain each third recognition result, and the compass straight line parameters and the moth-catching rod straight line parameters in each third recognition result are determined according to Hough transform; The compass direction vector and the moth-chasing rod direction vector are determined according to the compass straight line parameters and the moth-chasing rod straight line parameters respectively, and the moth-chasing angle is calculated according to the compass direction vector and the moth-chasing rod direction vector, and the swept area corresponding to the complete moth-chasing operation is calculated according to the moth-chasing angle and the rod length.

[0012] Preferably, determining the estimated number of field pests based on the number of moths in the area and the swept area includes: The number of moths and the swept area in the area of ​​each video data with the same classification are summarized, the estimated number of field pests per unit area under the classification is calculated, and the estimated number of field pests in the classification is calculated based on the estimated number per unit area and the total area of ​​the field of the classification, and the classification is obtained based on at least one of the shooting time period, rice type and moth-chasing location.

[0013] Preferably, the method further comprises: A moth occurrence level is set for each of the classified fields based on the estimated number per unit area.

[0014] The method provided in the embodiment of this specification can collect video data of the complete moth-chasing operation of the moth-chasing device, and after verifying the validity of the video data according to the level area, determine the number of moths and the swept area according to the moth area and the compass area respectively, and then determine the estimated number of field pests. In the above process, only the video data with verified validity will be used for quantity estimation, and the number of moths and the swept area do not rely on manual experience judgment, but are determined through image recognition, and the final estimated number of pests is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a flow chart of a method for estimating the number of field pests based on moth-chasing operations in one embodiment of the present specification.

[0017] Figure 2 The present invention is a schematic diagram of the working principle of a moth-repelling device in one embodiment of the present invention. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0019] In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application, and different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments recorded. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, although the embodiment may not be clearly recorded in the following text.

[0020] 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 functions and arrangements of the elements described without departing from the scope of the present application. Various processes or components may be appropriately omitted, substituted or added to each example. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted or combined. In addition, features described in some examples may be combined in other examples.

[0021] See also Figure 1 , Figure 1 It is a flow chart of a method for estimating the number of field pests based on moth-chasing operation provided in an embodiment of the present application. In an embodiment of the present application, the method is applied to a moth-chasing device, the device comprising a camera component and a moth-chasing rod component, the camera component comprising a camera support rod and a camera for shooting from a top-down angle, the moth-chasing rod component comprising a moth-chasing rod, a clamping module for clamping the moth-chasing rod, a compass and a level respectively arranged on the moth-chasing rod, and the method comprises: S101. After detecting a quantity estimation instruction, obtain video data corresponding to a complete moth-chasing operation collected by a camera, and determine a moth area, a level area, and a compass area in the video data.

[0022] The execution subject of the present application may be a controller disposed in the moth-chasing device, and the controller may be communicatively connected with a camera.

[0023] In the embodiments of this specification, the moth-repelling device of the present application will be first introduced. Figure 2 As shown, the moth-driving device is composed of a camera component and a moth-driving rod component. The camera component and the moth-driving rod component can be connected together through a clamping module and a camera support rod, or they can be separately installed on the staff. The moth-driving rod can be an ordinary bamboo pole or a PP or PVC plastic pipe. The clamping module can include a fixing fixture and a holding arm to assist the proper combination of the moth-driving rod and the operator's arm in the process of driving moths, and the length of the fixing fixture can be adjusted as needed, thereby adjusting the height between the moth-driving rod and the ground. With the assistance of the clamping module, the staff can keep the moth-driving rod horizontal by holding the clamping module without bending over, so that the moth-driving rod can be kept horizontal as much as possible during the subsequent sweeping of the moth-driving rod. The shooting angle of the camera is a top-down angle, so that the shooting picture is just facing the rice bush, and the moth-driving rod and the compass and level meter arranged on the moth-driving rod can be photographed, which is convenient for subsequent image recognition. In addition, the moth-chasing device may also include an indication module, which may be a display connected to the controller for displaying images captured by the camera, or may be directly integrated with the controller to generate corresponding instructions to the controller through touch operations on the display. Figure 2 The example shown is a situation where moths are driven away from a field on a ridge. In other embodiments, the moths can also be driven away from a field.

[0024] In the embodiment of the present specification, after detecting the quantity estimation instruction sent by the staff, the controller will control the camera to start shooting, and shoot the video data corresponding to the staff performing a complete moth-chasing operation process, and send the video data to the controller. The controller will identify the video data through a pre-trained image recognition model to determine the moth area, level area and compass area therein, wherein the level area and the compass area can be the target frame area of ​​the level and the compass after the level and the compass are identified as the level area and the compass area, and the moth area can be the target frame area corresponding to each identified moth, or it can be an area other than the level area and the compass area, and then the position of each identified moth is separately marked in the area.

[0025] The quantity estimation instruction can be a button electrically connected to the controller set on the moth-chasing device. When the staff presses the button, the quantity estimation instruction will be generated to indicate the need to start shooting images. The quantity estimation instruction can also be generated by the staff by clicking and selecting on the touch screen. The user can directly send instructions to the controller using a mobile phone terminal or directly perform voice control. The camera's judgment method for a complete moth-chasing operation can also be set according to specific needs. For example, it can detect the position of the moth-chasing rod through image recognition. When the moth-chasing rod remains motionless for 5-10 seconds, it is considered that the staff is about to start sweeping the rod and start collecting video data. After the moth-chasing rod moves to a certain position and stops again, it is considered that the sweeping of the rod is completed and the video collection stops. For example, after detecting the quantity estimation instruction or the video shooting instruction, it can also start collecting data of a preset time as video data.

[0026] In one possible implementation, after the quantity estimation instruction is detected, obtaining video data corresponding to a complete moth-chasing operation collected by a camera includes: After detecting the survey information released by the target institution, determining the field area to be surveyed based on the rice growth stage, rice variety and rice growth, and determining the regional coordinates and survey time of each field area to be surveyed, wherein the survey time includes the survey date and survey period; When any area coordinate that meets the survey time is reached and a quantity estimation instruction is detected, video data corresponding to a complete moth-chasing operation captured by the camera is obtained.

[0027] In the embodiment of this specification, target institutions such as plant protection stations and planting industry promotion centers will release survey information on pest coordination. After detecting the survey information, the controller will begin to determine the field area to be investigated. In order to ensure that the estimated number of pests is more real and effective, it is necessary to select fields with different rice growth stages, rice varieties and rice growth trends to carry out moth-driving pole sweeping tests respectively. Therefore, the controller will determine a number of field areas to be investigated according to the rice growth stage, rice variety and rice growth trend, and determine the regional coordinates of the field to be investigated. The moth-driving device can also be set with GPS, and the controller can determine the current coordinates of the moth-driving device according to GPS. Only when the current coordinates reach the regional coordinates, the arrival time meets the survey event and the quantity estimation instruction is detected, the controller will obtain video data. Among them, the specific method of determining the field area to be investigated can be to divide the field into field areas in advance, and mark the corresponding rice growth stage, rice variety and rice growth trend for each field area according to the actual situation. Next, the total number of field areas that need to be investigated is determined based on the total area to be investigated, and based on the total number, the number of field areas to be investigated is evenly distributed for each rice growth stage, rice variety and rice growth combination, and the corresponding number of field areas to be investigated are randomly selected from each field that meets the combination conditions.

[0028] In one possible implementation, the step of obtaining video data corresponding to a complete moth-chasing operation collected by a camera includes: Identify the moth-driving rod in the real-time image of the camera, and when the moth-driving rod does not move within a preset time, control the camera to start collecting video data; When the moth-repelling rod does not move again within a preset time period after moving, the camera is controlled to stop collecting the video data.

[0029] In the embodiment of the present specification, the controller first obtains the real-time image captured by the camera, and determines the position of the moth-chasing rod in the image by image recognition of the real-time image. Then, the controller monitors the change of the position of the moth-chasing rod, and if it does not move within a preset time, the controller starts to collect video data. When the moth-chasing rod stops again after moving, and does not move within a preset time, the controller stops collecting video data to obtain video data corresponding to a complete moth-chasing operation. This method can also simultaneously reduce the influence of the magnetic needle damping time.

[0030] In one possible implementation, the determining of the moth area, the level area, and the compass area in the video data includes: Divide the video data into frames according to the time sequence to obtain images of each frame; Based on image recognition, the moth area, the level area and the compass area in each of the frame images are respectively determined and marked.

[0031] In the embodiment of this specification, the video data is essentially a continuous image data. The image data will be recognized separately during the recognition process. Therefore, the video data will be framed in time sequence to obtain multiple framed images. Then, each framed image will be processed separately according to the pre-trained image recognition model to determine the moth area, level area and compass area in each framed image, and mark the areas.

[0032] S102: Identify the bubble position of the level in the level area, and verify the validity of the video data based on the bubble position.

[0033] In the embodiment of the present specification, when the camera is capturing images at a top-down angle, only when the moth-chasing rod is swept horizontally, can the swept area detected in the image match the actual swept area of ​​the rod, thereby ensuring the accuracy of the estimated number. Therefore, the controller will first determine whether the moth-chasing rod has been kept horizontal according to the bubble position of the level in the level area. When the moth-chasing rod is in a horizontal state, the bubble should be in the center of the level. Only when the bubbles in the video data are basically in the center, can the video data be considered valid. Video data judged to be invalid will not be subject to subsequent recognition processes.

[0034] In one possible implementation, the identifying the bubble position of the level in the level area and verifying the validity of the video data based on the bubble position includes: Based on a preset first image recognition model, the level area of ​​each of the framed images is respectively identified to obtain each first recognition result, wherein the first image recognition model is used to identify the bubble position of the level in the image, and the first recognition result includes the bubble being centered, the bubble being biased toward the palm direction, or the bubble being away from the palm direction; Calculate the result ratio of the target first recognition result among all the first recognition results, and obtain the validity verification result of the video data based on the result ratio, wherein the target first recognition result is the first recognition result characterized by the bubble being centered, and the validity verification result is characterized as valid when the result ratio is higher than a preset ratio.

[0035] In the embodiment of this specification, a first image recognition model for identifying the position of bubbles is pre-trained. The model can adopt the improved YOLOv5 model. The model structure consists of multiple important components, including a backbone network, a feature fusion network and a detection head. The backbone network is responsible for extracting low-level features of the input image. Common structures include CSPNet, which introduces more information during feature extraction and improves the expressiveness of the model. The feature fusion network integrates features from different levels and can better capture the information of multi-scale targets. The detection head is responsible for the final output of the bounding box and category probability to ensure that the target can be accurately judged. In order to enhance the detection performance, the CBAM (Convolutional Block Attention Module) attention mechanism is introduced. The work of CBAM is divided into two stages: channel attention and spatial attention. Channel attention extracts global information of feature maps through global average pooling and global maximum pooling, thereby generating importance weights for each channel. These weights will affect the sensitivity of the network to information from different channels, so that important features receive more attention during the training process. Spatial attention processes the feature map after channel weighting to generate a weight map that focuses on a specific area. The combination of these two modules can effectively improve the overall detection performance of the network, making the recognition of moth targets more accurate. In addition, during the model training process, NMS is used for each target box in the training results to eliminate overlapping boxes and retain the most accurate detection results. The training images should be taken under different ambient lighting, angles, and backgrounds to enhance the robustness of the model.

[0036] After obtaining the first image recognition model, each frame image will be processed separately to obtain the first recognition result. Each recognition result can characterize whether the bubble in the image is centered, biased toward the palm, or away from the palm. Finally, the proportion of the target first recognition result characterized as the bubble centered in all the first recognition results will be calculated. If the result proportion is higher than the preset proportion, the video data is considered valid, otherwise the video data is considered invalid.

[0037] S103. After the video data is verified to be valid, the number of moths in the moth area is identified, and the swept area corresponding to the complete moth-chasing operation is determined according to the angle change between the compass in the compass area and the moth-chasing rod, and the estimated number of field pests is determined according to the number of moths in the area and the swept area.

[0038] In the embodiments of the present specification, after the video data is verified to be valid, the number of regional moths and the swept area in the video data will be identified separately according to another image recognition model. The number of moths per unit area can be determined based on the number of moths and the swept area, and the estimated number of field pests can be estimated by combining this with the total field area. Among them, the number of regional moths can be obtained by directly identifying the number of moths in each frame of the video data and then calculating the average value, or it can be determined based on the number of moving trajectories after determining the moving trajectory of each moth through multi-target matching of continuous images. The swept area is determined by identifying the compass and the moth-driving rod through the image recognition model, determining the angle change between the two, and then determining the actual sweep angle, and calculating it based on the sweep angle combined with the rod length of the moth-driving rod.

[0039] In one embodiment, identifying the number of moths in the moth area includes: Based on a preset second image recognition model, respectively identify the moth region of each of the frame images to obtain each second recognition result, wherein the second image recognition model is used to identify the moth in the image; Adjacent second recognition results are matched based on a multi-target tracking algorithm, and the number of moths in the moth area is determined according to the number of matched moth tracks.

[0040] In the embodiment of the present specification, a second image recognition model will be pre-trained, and the second image recognition model will be specifically used to identify moths in the image, which can also be obtained by the improved YOLOv5 model training mentioned above. Since moths will be startled and fly during the sweeping process, the position of the same moth in different frame images will change. In order to ensure the accuracy of the moth number statistics, the controller will match the second recognition results of adjacent frame images according to the multi-target tracking algorithm to determine the position of the same recognition object in different images, and then construct the moth trajectory of the recognition object, and finally the number of moth trajectories can be used as the number of moths in the area. Among them, the multi-target tracking algorithm can be selected from SORT (Simple Online and Realtime Tracking), FairMOT (Fair Multi-Object Tracking), JDE (Joint Detection and Embedding), etc.

[0041] In one possible implementation, the matching of adjacent second recognition results based on a multi-target tracking algorithm and determining the number of moths in the moth area according to the number of matched moth tracks includes: After enlarging the target frame of each second recognition result based on a preset magnification, generating at least one prediction frame of each target frame in the next adjacent second recognition result based on a Kalman filter; Matching the predicted box and the target box in the next second recognition result based on the Hungarian algorithm to determine the best matching result; Repeat the step of generating at least one prediction frame for each target frame in the next adjacent second recognition result based on the Kalman filter until all optimal matching results are obtained, generate the moth trajectory of each moth object according to the optimal matching results, and determine the number of moths in the moth area according to the number of moth trajectories.

[0042] In the embodiment of the present specification, since the moth target is small and flies fast, the interval between adjacent frames is large, so the target frame in the second recognition result is first enlarged according to a preset magnification (for example, 3 times) to greatly improve the success rate of trajectory matching in the subsequent Kalman filter prediction and matching process. Then, the motion of each moth object is predicted using a Kalman filter, which is a recursive algorithm that can estimate the target position at the next moment through the current measurement data and the previous state. Therefore, the Kalman filter can predict the possible position (i.e., the prediction frame) of each target frame in the next frame image, and then match the target frame detected in the next frame image with the prediction frame. In the matching process, the Hungarian algorithm will be used, which finds the optimal matching result by calculating the intersection-and-union ratio of each pair of target frames. The intersection-and-union ratio threshold is set to 0.5, and it will only be regarded as the same target if the match is successful. If no match is found, it will be identified as a new target and a new Kalman filter will be initialized for it. After repeating the above process, each moth object in each second recognition result can be matched to generate a moth track for each moth object. One moth track corresponds to one moth, and the number of moths in the area is finally determined.

[0043] In one possible implementation, determining the swept area corresponding to the complete moth-chasing operation according to the angle change between the compass in the compass area and the moth-chasing rod includes: Based on a preset third image recognition model, the compass area of ​​each frame image is respectively identified to obtain each third recognition result, and the compass straight line parameters and the moth-catching rod straight line parameters in each third recognition result are determined according to Hough transform; The compass direction vector and the moth-chasing rod direction vector are determined according to the compass straight line parameters and the moth-chasing rod straight line parameters respectively, and the moth-chasing angle is calculated according to the compass direction vector and the moth-chasing rod direction vector, and the swept area corresponding to the complete moth-chasing operation is calculated according to the moth-chasing angle and the rod length.

[0044] In the embodiment of this specification, the compass and the moth-catching rod are first identified according to the pre-trained third image recognition model, which can also be obtained by training the improved YOLOv5 model mentioned above. For each third recognition result, the straight line of the compass and the moth-catching rod is identified according to the Hough transform, and then the corresponding straight line parameters are extracted to obtain the endpoint coordinates of the straight line. Assume that the starting point of the compass pointer is determined to be , the end point of the pointer is The starting point of the moth-chasing rod is The end point of the rod is , then the compass direction vector is: , the direction vector of the moth-chasing rod is: .

[0045] The dot product formula of two vectors can be used to calculate the final moth-catching angle of the moth-catching rod: The corresponding swept area can be calculated based on the moth-chasing angle and the known length of the moth-chasing rod.

[0046] In one embodiment, the method of determining the estimated number of field pests based on the number of moths in the area and the swept area includes: The number of moths and the swept area in the area of ​​each video data with the same classification are summarized, the estimated number of field pests per unit area under the classification is calculated, and the estimated number of field pests in the classification is calculated based on the estimated number per unit area and the total area of ​​the field of the classification, and the classification is obtained based on at least one of the shooting time period, rice type and moth-chasing location.

[0047] In the embodiment of this specification, in order to more accurately reflect the situation of field pests, it is necessary to classify the fields according to different situations to evaluate the pest occurrence in the fields under different classifications. Therefore, the controller will summarize the number of moths and the swept area of ​​the video data of the same classification to calculate the estimated number per unit area, and calculate the total estimated number of field pests based on the total area of ​​the fields of the classification.

[0048] In one embodiment, the method further comprises: A moth occurrence level is set for each of the classified fields based on the estimated number per unit area.

[0049] In the embodiments of the present specification, different moth occurrence levels can be pre-divided for different unit area estimated quantity intervals. After determining the unit area estimated quantity of each category, the corresponding moth occurrence level can be determined according to the interval to which it belongs. In this way, the relevant personnel can intuitively understand the severity of moth occurrence in the field based on the moth occurrence level.

[0050] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0051] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0052] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0053] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by entering a program to instruct related hardware. The program may be stored in a computer-readable memory, and the memory may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0054] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for estimating the number of field pests based on moth-chasing operations, characterized in that: The invention is applied to a moth-chasing device, the device comprising a camera component and a moth-chasing rod component, the camera component comprising a camera support rod and a camera for shooting from a top-down angle, the moth-chasing rod component comprising a moth-chasing rod, a clamping module for clamping the moth-chasing rod, a compass and a level respectively arranged on the moth-chasing rod, and the method comprising: After detecting the quantity estimation instruction, obtaining video data corresponding to a complete moth-chasing operation collected by the camera, and determining the moth area, the level area, and the compass area in the video data; identifying a bubble position of the level in the level area, and verifying the validity of the video data based on the bubble position; After the video data is verified to be valid, the number of regional moths in the moth area is identified, and the swept area corresponding to the complete moth-chasing operation is determined according to the angle change between the compass in the compass area and the moth-chasing rod, and the estimated number of field pests is determined according to the number of regional moths and the swept area.

2. The method according to claim 1, characterized in that After the quantity estimation instruction is detected, video data corresponding to a complete moth-chasing operation collected by the camera is obtained, including: After detecting the survey information released by the target institution, determining the field area to be surveyed based on the rice growth stage, rice variety and rice growth, and determining the regional coordinates and survey time of each field area to be surveyed, wherein the survey time includes the survey date and survey period; When any area coordinate that meets the survey time is reached and a quantity estimation instruction is detected, video data corresponding to a complete moth-chasing operation captured by the camera is obtained.

3. The method according to claim 1, characterized in that The step of obtaining video data corresponding to a complete moth-chasing operation collected by a camera includes: Identify the moth-driving rod in the real-time image of the camera, and when the moth-driving rod does not move within a preset time, control the camera to start collecting video data; When the moth-repelling rod does not move again within a preset time period after moving, the camera is controlled to stop collecting the video data.

4. The method according to claim 1, characterized in that: The determining of the moth area, the level area and the compass area in the video data comprises: Divide the video data into frames according to the time sequence to obtain images of each frame; Based on image recognition, the moth area, the level area and the compass area in each of the frame images are respectively determined and marked.

5. The method according to claim 4, characterized in that The step of identifying the bubble position of the level in the level area and verifying the validity of the video data based on the bubble position includes: Based on a preset first image recognition model, the level area of ​​each of the framed images is respectively identified to obtain each first recognition result, wherein the first image recognition model is used to identify the bubble position of the level in the image, and the first recognition result includes the bubble being centered, the bubble being biased toward the palm direction, or the bubble being away from the palm direction; Calculate the result ratio of the target first recognition result among all the first recognition results, and obtain the validity verification result of the video data based on the result ratio, wherein the target first recognition result is the first recognition result characterized by the bubble being centered, and the validity verification result is characterized as valid when the result ratio is higher than a preset ratio.

6. The method according to claim 4, characterized in that The identifying the number of moths in the moth area comprises: Based on a preset second image recognition model, respectively identify the moth region of each of the frame images to obtain each second recognition result, wherein the second image recognition model is used to identify the moth in the image; Adjacent second recognition results are matched based on a multi-target tracking algorithm, and the number of moths in the moth area is determined according to the number of matched moth tracks.

7. The method according to claim 6, characterized in that The method of matching adjacent second recognition results based on a multi-target tracking algorithm and determining the number of moths in the moth area according to the number of matched moth tracks includes: After enlarging the target frame of each second recognition result based on a preset magnification, generating at least one prediction frame of each target frame in the next adjacent second recognition result based on a Kalman filter; Matching the predicted box and the target box in the next second recognition result based on the Hungarian algorithm to determine the best matching result; Repeat the step of generating at least one prediction frame for each target frame in the next adjacent second recognition result based on the Kalman filter until all optimal matching results are obtained, generate the moth trajectory of each moth object according to the optimal matching results, and determine the number of moths in the moth area according to the number of moth trajectories.

8. The method according to claim 4, characterized in that The step of determining the swept area corresponding to the complete moth-chasing operation according to the angle change between the compass in the compass area and the moth-chasing rod comprises: Based on a preset third image recognition model, the compass area of ​​each frame image is respectively identified to obtain each third recognition result, and the compass straight line parameters and the moth-catching rod straight line parameters in each third recognition result are determined according to Hough transform; The compass direction vector and the moth-chasing rod direction vector are determined according to the compass straight line parameters and the moth-chasing rod straight line parameters respectively, and the moth-chasing angle is calculated according to the compass direction vector and the moth-chasing rod direction vector, and the swept area corresponding to the complete moth-chasing operation is calculated according to the moth-chasing angle and the rod length.

9. The method according to claim 1, characterized in that: The estimated number of field pests determined based on the number of moths in the area and the area swept includes: The number of moths and the swept area in the area of ​​each video data with the same classification are summarized, the estimated number of field pests per unit area under the classification is calculated, and the estimated number of field pests in the classification is calculated based on the estimated number per unit area and the total area of ​​the field of the classification, and the classification is obtained based on at least one of the shooting time period, rice type and moth-chasing location.

10. The method according to claim 9, characterized in that The method further comprises: A moth occurrence level is set for each of the classified fields based on the estimated number per unit area.

Images