Automatic monitoring device and method for rice planthoppers

Abstract

This invention relates to an automatic monitoring device and method for rice paddy planthoppers. The device includes a support frame and a whiteboard for collecting insects. A track is mounted on the support frame, and a trolley is mounted on the track. The trolley has a housing with wheels at the bottom. A drive motor and control system are located inside the housing. A solar panel is mounted on the top of the housing, and a support rod is located on the side of the housing, housing a camera and a blower. The control system includes a main controller and a mobile communication module. The method involves deploying the device and sampling points, automatically starting the trolley, blowing down planthoppers and taking pictures at the sampling points, automatically controlling the trolley to return, and the main controller processing the photos to obtain the species and quantity of planthoppers (including different instars, wing types, and sexes). This invention enables the device to operate automatically at set times daily to automatically survey the species and quantity of rice paddy planthoppers. Users can view data and set operating parameters through a server-side backend or an app, thereby greatly reducing the labor intensity of planthopper surveys.

Description

Technical Field

[0001] This invention relates to an automatic monitoring device and method for rice planthoppers, belonging to the field of digital agricultural plant protection. Background Technology

[0002] Rice planthoppers are a major pest of rice, and systematic monitoring of planthoppers is the foundation for accurate early warning of their occurrence trends, thus enabling scientific control. Currently, field surveys of planthoppers still rely on manual labor. Surveyors go to the field and tap the base of the rice stalks with a white porcelain basin. After the planthoppers fall into the basin, the species and number of planthoppers are visually counted. Although AR glasses have emerged in recent years to take pictures of the porcelain basin and then use models to intelligently identify the species and number of planthoppers, surveyors still need to go to the field to operate. Moreover, planthopper surveys are mostly conducted in the height of summer, which is not only time-consuming and laborious, but also difficult to see with the human eye due to strong light and obvious reflections. Changes in angle and distance affect the quality of the photos, making it difficult to improve the accuracy of the identification model. Smaller 1-2 instar planthoppers are basically undetectable, and only 3 instar and older planthoppers can be detected, which may lead to missing the optimal period for planthopper control.

[0003] In addition, existing technologies include methods for directly and automatically capturing images of the base of rice plants for identification. However, because planthoppers are very small, the leaves (including withered leaves and other leaves) at the base of the rice plants easily obstruct the lens's field of view, resulting in incomplete images. Furthermore, compared to rice plants, planthoppers are smaller, making it difficult for cameras to focus on the planthoppers during automatic shooting. In addition, the three-dimensional background of the rice plants results in insufficient depth of field, making it difficult to capture clear images. All of these factors make it difficult for such technologies to achieve the desired results. Summary of the Invention

[0004] The main objective of this invention is to provide an automatic monitoring device and method for rice paddy planthoppers, based on the problems existing in the prior art. This device can operate automatically at regular intervals every day to automatically investigate the types and quantities of rice paddy planthoppers. Users can view data or set operating parameters through the server backend, thereby greatly reducing the labor intensity of planthopper investigation and realizing automated and intelligent monitoring of rice paddy planthoppers.

[0005] The technical solution of this invention to solve its technical problem is as follows:

[0006] An automatic monitoring device for rice planthoppers includes a support frame and a whiteboard for collecting insects. The support frame is placed above the target rice plants in the paddy field, and the whiteboard is laid on a predetermined water surface or ground surface of the target rice plants. A track is installed on the support frame, and a trolley is mounted on the track. The trolley has a housing with wheels at the bottom. The housing contains a drive motor and a control system, and the top of the housing is equipped with a solar panel. Support rods are located on the sides of the housing. The wheels are matched with the track, and the output end of the drive motor is connected to the wheel drive. The support rods are located outside the track. The support rods are telescopic rods with telescopic motors. The fixed end of the support rods is fixedly connected to the housing, and the movable end of the support rods is equipped with a camera and a blower. The blower has an air outlet angled towards the rice plants. The control system has a main controller and a mobile communication module. The controlled ends of the drive motor, the telescopic motor, the camera, and the blower are respectively connected to the control end of the main controller. The solar panel is electrically connected to the main controller; the main controller's communication terminal is connected to the mobile communication module, which communicates with the server via a mobile network; a starting trigger is provided at the beginning of the track, and an ending trigger is provided at the end of the track; the trolley's body is equipped with a starting limit switch corresponding to the starting trigger and an ending limit switch corresponding to the ending trigger; the control system also has a photo processing module, a detection module, and a storage module, all connected to the main controller. The photo processing module processes photos taken by the camera; the detection module identifies and detects the planthoppers in the photos based on the processing results of the photo processing module, including different types (brown planthopper, white-backed planthopper, and gray planthopper), instars (1st, 2nd, 3rd, 4th, and 5th instars), wing types (long-winged and short-winged), and adult sexes; the storage module is used for the main controller to store and retrieve data.

[0007] The further improved technical solution for this device is as follows:

[0008] Preferably, the blower and the camera are located on the same pole; or, there are two poles, namely a first pole and a second pole, with the blower located on the first pole and the camera located on the second pole, and the blower and the camera located on opposite sides of the target rice clump.

[0009] More preferably, the air outlet of the blower is connected to an air duct, and the support rod has a retaining ring for securing the air duct outlet; or, the camera has a flash.

[0010] Preferably, the insect-collecting whiteboard is a white flat panel; the width of the insect-collecting whiteboard covers the predetermined water surface or ground of the target rice clump; when the insect-collecting whiteboard is placed on the water surface, it floats on the water surface; the track adopts a splicing structure.

[0011] The present invention also proposes:

[0012] The automatic monitoring method for rice planthoppers, using the automatic monitoring device for rice planthoppers described above, includes the following steps:

[0013] Step 1: Set up a support frame above the target rice clumps in the paddy field, install a track on the support frame, and place the trolley at the starting end of the track; lay insect-collecting whiteboards on the predetermined water surface or ground of the target rice clumps, and arrange several sampling points on the insect-collecting whiteboards along the length of the track.

[0014] The second step is to start the control system at a set time every day. After the control system automatically obtains the operating parameters from the server, the main controller of the control system starts the drive motor of the trolley and moves the trolley along the track by controlling the drive motor to run in the forward direction. Under the control of the main controller, the trolley moves to the sampling point and stops. The blowing device on the trolley support blows the planthoppers at the base of the rice stalks onto the insect collection board. The camera on the trolley support takes a picture of the insect collection board. Then the trolley moves to the next sampling point and stops. The above process is repeated.

[0015] Third, when the trolley moves to the end of the track, the end trigger of the track triggers the end limit switch of the trolley. The end limit switch sends a signal to the main controller. After the main controller controls the trolley to stop moving, it controls the drive motor to run in reverse so that the trolley moves in the opposite direction along the track. When the trolley moves to the beginning of the track, the start trigger of the track triggers the start limit switch of the trolley. The start limit switch sends a signal to the main controller, and the main controller controls the trolley to stop moving.

[0016] Fourth, the main controller first uses the photo processing module to process the photos taken by the camera, and then uses the detection module to identify and detect the planthoppers in the photos based on the processing results of the photo processing module. After that, the data is uploaded to the server via the mobile network.

[0017] Note: When implementing the above technical solution in perennial rice-growing areas, in the first step, the support frame can be fixed in the field after the initial deployment for subsequent multiple uses, without the need for re-deployment each time.

[0018] The technical solution for further improvement of this method is as follows:

[0019] Preferably, when the trolley adopts the first structure, the specific process of the second step is as follows: when the trolley is located at the sampling point, the insect-collecting whiteboard is located below the camera. The main controller controls the extension and retraction of the support rod to make the air outlet of the blowing device reach the first preset height. The blowing device blows air towards the base of the rice clump through its air outlet, blowing the planthoppers onto the insect-collecting whiteboard. If the camera is already at the second preset height, the camera is aimed at the insect-collecting whiteboard to take a picture. If the camera is not at the second preset height, the main controller controls the extension and retraction of the support rod to make the camera reach the second preset height, and then the camera is aimed at the insect-collecting whiteboard to take a picture. The first structure is: the blowing device and the camera are located on the same support rod.

[0020] When the trolley adopts the second structure, the specific process of the second step is as follows: When the trolley is located at the sampling point, the insect-collecting whiteboard is located below the camera. The main controller controls the extension and retraction of the first support rod to make the air outlet of the blowing device reach the first preset height. The blowing device blows air through its air outlet to the base of the rice clump, blowing the planthoppers down onto the insect-collecting whiteboard. Then, the main controller controls the extension and retraction of the second support rod to make the camera reach the second preset height. The camera is aimed at the insect-collecting whiteboard to take a picture. The second structure has two support rods, namely the first support rod and the second support rod. The blowing device is located on the first support rod, and the camera is located on the second support rod. The blowing device and the camera are located on both sides of the target rice clump.

[0021] Preferably, in the second step, the operating parameters include camera shooting parameters, drive motor control parameters, support rod control parameters, air blowing control parameters, and detection parameters; the control system starts up at a set time every day 1-2 hours before the local sunset time; the local sunset time is calculated by the server based on the local latitude and longitude and the date of the day, and the control system automatically obtains the local sunset time when it obtains the operating parameters from the server.

[0022] Preferably, in the fourth step, the photo processing module uses the auxiliary slicing Sahi algorithm to cut the photo; the specific process of cutting is as follows: the photo is translated and cut according to the preset slice translation parameters and slice size parameters to obtain multiple images for input into the detection module;

[0023] The detection module uses the YOLO model to first identify and detect each image obtained by the photo processing module and obtain the recognition results. Then, the recognition results are processed by non-maximum suppression to remove duplicates, and finally the species and number of planthoppers in the photo are obtained (including different types, instars, wing types, males and females, etc.).

[0024] Preferably, in the first step, the track laying length is determined according to actual needs; the number of sampling points is 20-50; the distance between adjacent sampling points is 20-30 cm; or, in the second step, the shooting area of ​​the camera is 20±2cm×15±1cm; the camera automatically focuses when taking pictures; or, in the third step, the starting limit switch and the ending limit switch are respectively contact-triggered mechanical limit switches, and the starting trigger and the ending trigger are respectively triggers that match the contact-triggered mechanical limit switches.

[0025] Preferably, in the fourth step, the data uploaded to the server includes the species and number of planthoppers (including different types, instars, wing shapes, males and females, etc.) in the photos obtained by the detection module; when the preset data to be uploaded to the server also includes photos taken by the camera and / or images processed by the photo processing module, these data are uploaded to the server together; the server uses the uploaded data to train, optimize, and iterate the detection model; or, in the fourth step, the YOLO model configuration file has a p2 detection head to improve the detection capability of small target objects.

[0026] Compared with existing technologies, the monitoring device of this invention can run automatically at set times every day, greatly reducing the need for manual operation in the field. Data can be viewed and the operating parameters of the main controller can be set through the server backend or APP, realizing the automation of planthopper system survey. The use of insect collection whiteboard can provide a simple background for taking pictures and make it easier for the camera to focus during automatic shooting. Taking pictures before sunset can avoid strong light reflection interference and use soft diffused light to make the pictures clear and obtain high-quality pictures, thereby training a more accurate detection model and making the recognition accuracy high enough. In addition, a sufficient number of sampling points can be set according to actual needs to ensure the sample volume. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the overall situation in Embodiment 1 of the present invention.

[0028] Figure 2 This is a front view schematic diagram of the vehicle in Embodiment 1 of the present invention.

[0029] Figure 3 This is a side view of the vehicle according to Embodiment 1 of the present invention. Figure A shows the first structure, and Figure B shows the second structure.

[0030] Figure 4 This is the electrical schematic diagram of Embodiment 1 of the present invention. Detailed Implementation

[0031] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. However, the present invention is not limited to the examples given.

[0032] Example 1

[0033] like Figures 1 to 4As shown, the automatic monitoring device for rice planthoppers in this embodiment includes a support 01 and a whiteboard for collecting insects 02; a track 03 is installed on the support 01, and a trolley 04 is mounted on the track 03; the trolley 04 has a housing 05, wheels 17 are located at the bottom of the housing 05, a drive motor and a control system are located inside the housing 05, a solar panel 06 is located on the top of the housing 05, and a support rod 07 is located on the side of the housing 05; the wheels 17 are matched with the track 03, the output end of the drive motor is connected to the wheels 17, and the support rod 07 is located outside the track 03; the support rod 07 is a telescopic rod with a telescopic motor, the fixed end of the support rod 07 is fixedly connected to the housing 05, and a camera 08 and a blower 09 are installed on the movable end of the support rod 07. (Note: The support rod 07 is located on the outside of the track 03; the support rod 07 is a telescopic rod with a telescopic motor, the fixed end of the support rod 07 is fixedly connected to the housing 05, and a camera 08 and a blower 09 are installed on the movable end of the support rod 07.) Figure 1 This is only a simplified representation; its specific structure can be found in [the original text]. Figure 3 。

[0034] The support frame 01 is positioned above the target rice clump 16 in the paddy field. The insect-collecting whiteboard 02 is a white, flat component that floats on the water surface; it is laid on a predetermined water or ground surface of the target rice clump 16, with its width covering the predetermined water or ground surface. A blowing device 09 has an outlet angled towards the rice clump, blowing planthoppers from the base of the rice clump onto the insect-collecting whiteboard 02, whereupon a camera 08 takes a picture. The insect-collecting whiteboard 02 provides a simple background for the picture and makes it easier for the camera 08 to focus during automatic shooting.

[0035] like Figure 4 As shown, the control system has a main controller and a mobile communication module; the controlled ends of the drive motor, the telescopic motor, the camera 08, and the blower 09 are respectively connected to the control end of the main controller; the solar power panel 06 is electrically connected to the main controller to provide power to the entire device; the communication end of the main controller is connected to the mobile communication module, and the mobile communication module communicates with the server via a mobile network.

[0036] Specifically, there are two possible layouts for the blower 09 and the camera 08, and one of them can be selected based on the actual situation during implementation.

[0037] The first structure: such as Figure 3 As shown in Figure A, the blower 09 and the camera 08 are located on the same support rod 07.

[0038] The second structure: such as Figure 3 As shown in Figure B, there are two support rods 07, namely the first support rod 071 and the second support rod 072; the blower 09 is located on the first support rod 071, and the camera 08 is located on the second support rod 072. The blower 09 and the camera 08 are located on both sides of the target rice clump 16.

[0039] To facilitate adjustment of the blowing direction, the air outlet of the blower 09 is connected to the air duct 10, and the support rod 07 has a retaining ring 11 to fix the outlet of the air duct 10; to facilitate supplemental lighting when the ambient light is insufficient, the camera 08 can be equipped with a flash.

[0040] Track 03 adopts a splicing structure, so the total length of track 03 can be extended by several meters to tens of meters as needed.

[0041] The starting end of track 03 is provided with a starting trigger 12, and the ending end of track 03 is provided with an ending trigger 13; the trolley 04 housing 05 is provided with a starting limit switch 14 corresponding to the starting trigger 12, and an ending limit switch 15 corresponding to the ending trigger 13.

[0042] like Figure 4 As shown, the control system also has a photo processing module, a detection module, and a storage module, which are respectively connected to the main controller. The photo processing module is used to process the photos taken by the camera 08. The detection module is used to identify and detect the planthoppers in the photos based on the processing results of the photo processing module, including different types (brown planthopper, white-backed planthopper, and gray planthopper), instars (1st, 2nd, 3rd, 4th, and 5th instars), wing types (long-winged and short-winged), and adult sexes. The storage module is used for the main controller to store and retrieve data.

[0043] Example 2

[0044] This embodiment uses the automatic monitoring device for rice planthoppers from Embodiment 1 to implement the automatic monitoring method for rice planthoppers.

[0045] The specific method of this embodiment is as follows.

[0046] The first step is to set up a support frame above the target rice clumps in the paddy field, install a track on the support frame, and place the pre-assembled trolley at the starting end of the track; lay insect-collecting whiteboards on the predetermined water surface or ground of the target rice clumps, and arrange several sampling points on the insect-collecting whiteboards along the length of the track.

[0047] The track length can be determined according to actual needs to ensure a sufficient number of sampling points to guarantee the sample volume. When the insect-collecting whiteboard is placed on the water surface, it floats. The number of sampling points is 20-50. The distance between adjacent sampling points is 20-30 cm, and the specific distance is matched with the running time of the drive motor. Note: The planthoppers on the insect-collecting whiteboard will return to the rice plants to feed on their own. The water flow formed by the frequent drainage of the rice paddy can clean the insect-collecting whiteboard, and rain will also clean it. Therefore, the insect-collecting whiteboard does not require manual cleaning.

[0048] The second step is to start the control system at a set time every day and automatically obtain operating parameters from the server. These parameters include camera shooting parameters (such as start shooting time, photo size, etc.), drive motor control parameters (such as drive motor running time corresponding to the distance between sampling points, etc.), support rod control parameters (such as telescopic motor running time, support rod extension amount, etc.), blowing control parameters (such as blowing force and duration, etc.), and detection parameters (such as the specific detection model and its parameters, whether the detection model needs to be updated and downloaded, etc.).

[0049] The main controller of the control system starts the drive motor of the trolley and controls the drive motor to run in the forward direction so that the trolley moves forward along the track. Under the control of the main controller, the trolley moves to the sampling point and stops. The blower on the trolley support rod blows the planthoppers at the base of the rice stalks onto the insect collection board. The camera on the trolley support rod takes a picture of the insect collection board. Then the trolley moves to the next sampling point and stops. The above process is repeated.

[0050] The vehicle repeatedly moves, blows down, and takes photos at each sampling point to obtain sampling photos for subsequent identification and detection of planthoppers.

[0051] Since there are two possible layouts for the blower and camera in Embodiment 1, the specific operation process differs slightly between the two scenarios, as detailed below.

[0052] If the layout of the blowing device and camera adopts the first scenario of Embodiment 1, then: when the trolley is located at the sampling point, the insect-collecting whiteboard is located below the camera. The main controller controls the extension and retraction of the support rod to position the air outlet of the blowing device at a first preset height. The blowing device blows air through its air outlet towards the base of the rice stalks, blowing the planthoppers onto the insect-collecting whiteboard. If the camera is already at a second preset height, the camera is aimed at the insect-collecting whiteboard to take a picture. If the camera is not at the second preset height, the main controller controls the extension and retraction of the support rod to position the camera at the second preset height, and then the camera is aimed at the insect-collecting whiteboard to take a picture. In this case, the support rod structure is relatively simple. During implementation, it is necessary to specifically adjust the orientation of the air outlet of the blowing device beforehand so that the airflow can blow the planthoppers at the base of the rice stalks onto the insect-collecting whiteboard.

[0053] If the layout of the blowing device and camera adopts the second scenario of Embodiment 1, then: when the trolley is located at the sampling point, the insect-collecting whiteboard is located below the camera. The main controller controls the extension and retraction of the first support rod to position the air outlet of the blowing device at a first preset height. The blowing device blows air through its air outlet towards the base of the rice stalk, blowing the planthoppers onto the insect-collecting whiteboard. Then, the main controller controls the extension and retraction of the second support rod to position the camera at a second preset height, and the camera takes a picture of the insect-collecting whiteboard. In this scenario, since the blowing device and camera are distributed on both sides of the rice stalk, it is easier to achieve the effect of blowing the planthoppers onto the insect-collecting whiteboard. Note: The first and second preset heights in the above two scenarios can be the same or different; when setting the first preset height, the main consideration is whether the blowing can effectively blow the planthoppers at the base of the rice stalk onto the insect-collecting whiteboard, and when setting the second preset height, the main consideration is whether the camera can take a clear picture through autofocus when the insect-collecting whiteboard fills the shooting area.

[0054] The camera's image capture area is 20±2cm × 15±1cm; it automatically focuses during image capture. Because the insect-collecting whiteboard is flat, the camera focuses easily; and by using a close-up mode, the small image capture area allows for more pixels per insect at the same image resolution. Furthermore, since the whiteboard is located on the water or ground of the target rice paddy, the camera's position is close to the base of the rice, resulting in diffused ambient light and a softer image, minimizing the impact of direct sunlight; the camera can also be equipped with a flash for supplemental lighting in low-light conditions.

[0055] The control system is scheduled to start 1-2 hours before the local sunset time each day. During this period, the base of the rice stalks is minimally affected by direct sunlight, avoiding strong light interference and facilitating camera photography. Specifically, the local sunset time is calculated by the server based on the local latitude and longitude and the date of the day. The control system automatically retrieves the local sunset time from the server along with its operating parameters.

[0056] Third, when the trolley moves along the track to the end of the track, the end trigger of the track triggers the end limit switch of the trolley. The end limit switch sends a signal to the main controller. After the main controller controls the trolley to stop moving, it controls the drive motor to run in reverse so that the trolley moves in the opposite direction along the track. When the trolley moves to the beginning of the track, the start trigger of the track triggers the start limit switch of the trolley. The start limit switch sends a signal to the main controller, and the main controller controls the trolley to stop moving.

[0057] Among them, the starting limit switch and the ending limit switch are contact-triggered mechanical limit switches; for example, plunger switch, roller switch, rocker switch, fork switch, micro switch, etc.; correspondingly, the starting trigger and the ending trigger are triggers that are matched with the contact-triggered mechanical limit switches.

[0058] The fourth step involves the main controller first processing the photos taken by the camera using the photo processing module, and then using the detection module to identify and detect the planthoppers in the photos based on the processing results, including different types (brown planthoppers, white-backed planthoppers, and gray planthoppers), instars (1st, 2nd, 3rd, 4th, and 5th instars), wing types (long-winged and short-winged), and adult sexes. This data is then uploaded to the server via a mobile network.

[0059] The data uploaded to the server includes the species and number of planthoppers in the photos obtained by the detection module. Depending on the actual needs, it can selectively include photos taken by the camera or the processing results of the photo processing module. The server uses this data to process it further and incorporate it into the label extended dataset to train and optimize the iterative detection model.

[0060] Because the photos captured by the camera are inherently large (e.g., 3000×4000 pixels, 6000×8000 pixels, or even larger), if the photos are directly detected by the detection module without being cropped, the YOLO model will first shrink the photos to a preset size (e.g., 640 pixels × 640 pixels) before detection. Since planthoppers are already small, shrinking the photos will make the insects even smaller, making them difficult to detect. Therefore, the method of first cropping the photos into multiple images by the photo processing module, and then having the detection module perform the detection, ensures that the insects in each image input to the detection module are at their original pixel size for identification and detection. This effectively guarantees the accuracy of insect detection and recognition.

[0061] The photo processing module employs the auxiliary slicing Sahi algorithm to slice photos, improving the recognition and detection of small targets like planthoppers. The specific slicing process involves translating and slicing the photo according to preset slicing translation and size parameters (e.g., 640 pixels × 640 pixels per slice) to obtain multiple images. This is more suitable for scenarios where the small size of planthoppers makes detection difficult.

[0062] The detection module uses the YOLO model. It first identifies and detects each processed image, obtaining the recognition results. Then, it performs non-maximum suppression (NMS) deduplication on the recognition results, ultimately obtaining the species and quantity of planthoppers in the photos (including different types, instars, wing types, and sexes). Experiments have verified that adding the p2 detection head (p2head) to the YOLO model's configuration file can improve the recognition rate of individual insects (especially 1st-2nd instar planthopper nymphs), thereby improving the detection effect for small target objects like planthoppers. Therefore, the p2 detection head can be added as needed in actual implementation.

[0063] In addition, regarding the settings of slice translation parameters and slice size parameters, if the parameter values ​​are small, more images will be generated. Although this can maintain a high recognition accuracy, it will increase the processing time. On the other hand, if the parameter values ​​are large, although the number of images will be reduced and the processing time can be shortened, it may reduce the recognition accuracy. Therefore, adjustments can be made according to the actual situation to try to meet both the requirements of recognition accuracy and processing time.

[0064] As can be seen from the above embodiments, the monitoring device of the present invention can operate automatically at set times every day, greatly reducing the need for manual operation in the field. Data can be viewed and the operating parameters of the main controller can be set on the server backend or APP, realizing the automation of planthopper survey. The use of insect-collecting whiteboards provides a simple background for photography and makes it easier for the camera to focus during automatic shooting, resulting in clear and high-quality images. This allows for the training of a more accurate detection model, ensuring a sufficiently high recognition accuracy. In addition, the monitoring device of the present invention can set a sufficient number of sampling points according to actual needs, thereby ensuring a sufficient sample size.

[0065] The applicant declares that the technical solution of this invention is illustrated by the above embodiments, but this invention is not limited to the above embodiments, that is, it does not mean that this invention must rely on the above embodiments to be implemented. Those skilled in the art should understand that any improvements to this invention, equivalent substitutions of the raw materials of the product, additions of auxiliary components, and selection of specific methods all fall within the protection and disclosure scope of this invention. Furthermore, based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort, as well as technical solutions formed by equivalent substitutions or equivalent transformations (such as shaking rice clumps with a motor to make planthoppers fall onto a whiteboard for collecting insects, etc.), also fall within the protection scope claimed by this invention.

[0066] Furthermore, the preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various simple modifications can be made to the technical solution of the present invention, and these simple modifications also fall within the protection scope of the present invention.

[0067] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not describe the various possible combinations separately.

Claims

1. An automatic monitoring device for rice planthoppers, including a support frame, characterized in that, It also includes an insect-collecting whiteboard; the support frame is installed above the target rice clumps in the paddy field, and the insect-collecting whiteboard is laid on the predetermined water surface or ground of the target rice clumps; a track is installed on the support frame, and a trolley is mounted on the track; the trolley has a box, wheels are located at the bottom of the box, a drive motor and a control system are located inside the box, a solar panel is located on the top of the box, and a support rod is located on the side of the box; the wheels are matched with the track, the output end of the drive motor is connected to the wheel drive, and the support rod is located outside the track; the support rod is a telescopic rod with a telescopic motor, the fixed end of the support rod is fixedly connected to the box, and the movable end of the support rod is equipped with a camera and a blowing device, the blowing device having an air outlet angled towards the rice clumps; the control system has a main controller and a mobile communication module; the controlled end of the drive motor, the controlled end of the telescopic motor, The controlled end of the camera and the controlled end of the blower are respectively connected to the control end of the main controller; the solar power panel is electrically connected to the main controller; the communication end of the main controller is connected to the mobile communication module, and the mobile communication module communicates with the server via a mobile network; the starting end of the track is equipped with a starting trigger and the ending end of the track is equipped with an ending trigger; the trolley's body is equipped with a starting limit switch corresponding to the starting trigger and an ending limit switch corresponding to the ending trigger; the control system also has a photo processing module, a detection module, and a storage module respectively connected to the main controller. The photo processing module is used to process the photos taken by the camera; the detection module is used to identify and detect the planthoppers in the photos based on the processing results of the photo processing module and obtain the species and quantity of planthoppers in the photos; the storage module is used for the main controller to store and retrieve data.

2. The automatic monitoring device for rice paddy planthoppers according to claim 1, characterized in that, The blowing device and the camera are located on the same pole; or, there are two poles, namely a first pole and a second pole, with the blowing device located on the first pole and the camera located on the second pole, and the blowing device and the camera located on opposite sides of the target rice clump.

3. The automatic monitoring device for rice paddy planthoppers according to claim 2, characterized in that, The air outlet of the blower is connected to an air duct, and the support rod has a retaining ring for fixing the air duct outlet; or, the camera has a flash.

4. The automatic monitoring device for rice paddy planthoppers according to any one of claims 1 to 3, characterized in that, The insect-collecting whiteboard is a white flat panel; the width of the insect-collecting whiteboard covers the predetermined water surface or ground of the target rice clump; when the insect-collecting whiteboard is placed on the water surface, it floats on the water surface; the track adopts a splicing structure.

5. An automatic monitoring method for rice paddy planthoppers, characterized by: The method using the automatic monitoring device for rice planthoppers according to any one of claims 1 to 4 includes the following steps: Step 1: Set up a support frame above the target rice clumps in the paddy field, install a track on the support frame, and place the trolley at the starting end of the track; lay insect-collecting whiteboards on the predetermined water surface or ground of the target rice clumps, and arrange several sampling points on the insect-collecting whiteboards along the length of the track. The second step is to start the control system at a set time every day. After the control system automatically obtains the operating parameters from the server, the main controller of the control system starts the drive motor of the trolley and moves the trolley along the track by controlling the drive motor to run in the forward direction. Under the control of the main controller, the trolley moves to the sampling point and stops. The blowing device on the trolley support blows the planthoppers at the base of the rice stalks onto the insect collection board. The camera on the trolley support takes a picture of the insect collection board. Then the trolley moves to the next sampling point and stops. The above process is repeated. Third, when the trolley moves to the end of the track, the end trigger of the track triggers the end limit switch of the trolley. The end limit switch sends a signal to the main controller. After the main controller controls the trolley to stop moving, it controls the drive motor to run in reverse so that the trolley moves in the opposite direction along the track. When the trolley moves to the beginning of the track, the start trigger of the track triggers the start limit switch of the trolley. The start limit switch sends a signal to the main controller, and the main controller controls the trolley to stop moving. Fourth, the main controller first uses the photo processing module to process the photos taken by the camera, and then uses the detection module to identify and detect the planthoppers in the photos based on the processing results of the photo processing module. After that, the data is uploaded to the server via the mobile network.

6. The automatic monitoring method for rice paddy planthoppers according to claim 5, characterized in that, When the trolley adopts the first structure, the specific process of the second step is as follows: When the trolley is located at the sampling point, the insect-collecting whiteboard is located below the camera. The main controller controls the extension and retraction of the support rod to bring the air outlet of the blowing device to a first preset height. The blowing device blows air towards the base of the rice clump through its air outlet, blowing the planthoppers onto the insect-collecting whiteboard. If the camera is already at the second preset height, the camera is aimed at the insect-collecting whiteboard to take a picture. If the camera is not at the second preset height, the main controller controls the extension and retraction of the support rod to bring the camera to the second preset height, and then the camera is aimed at the insect-collecting whiteboard to take a picture. The first structure is: the blowing device and the camera are located on the same support rod. When the trolley adopts the second structure, the specific process of the second step is as follows: When the trolley is located at the sampling point, the insect-collecting whiteboard is located below the camera. The main controller controls the extension and retraction of the first support rod to position the air outlet of the blowing device at a first preset height. The blowing device blows air towards the base of the rice clump through its air outlet, blowing the planthoppers onto the insect-collecting whiteboard. Then, the main controller controls the extension and retraction of the second support rod to position the camera at a second preset height. The camera is aimed at the insect-collecting whiteboard to take a picture. The second structure is as follows: There are two support rods, namely the first support rod and the second support rod. The blowing device is located on the first support rod, and the camera is located on the second support rod. The blowing device and the camera are located on both sides of the target rice clump.

7. The automatic monitoring method for rice paddy planthoppers according to claim 5, characterized in that, In the second step, the operating parameters include camera shooting parameters, drive motor control parameters, support rod control parameters, air blowing control parameters, and detection parameters; the control system starts up at a set time every day 1-2 hours before the local sunset time; the local sunset time is calculated by the server based on the local latitude and longitude and the date of the day, and the control system automatically obtains the local sunset time when it obtains the operating parameters from the server.

8. The automatic monitoring method for rice paddy planthoppers according to claim 5, characterized in that, in the fourth step, the photo processing module uses the auxiliary slicing Sahi algorithm to cut the photo; the specific process of the cutting is as follows: the photo is translated and cut according to preset slicing translation parameters and slicing size parameters to obtain multiple images for input into the detection module; The detection module uses the YOLO model to first identify and detect each image obtained by the photo processing module and obtain the identification results. Then, the identification results are processed by non-maximum suppression to remove duplicates, and finally the species and number of planthoppers in the photo are obtained.

9. The automatic monitoring method for rice paddy planthoppers according to any one of claims 5 to 8, characterized in that, in the first step, the laying length of the track is determined according to actual needs; the number of sampling points is 20-50; the distance between adjacent sampling points is 20-30 cm; or, in the second step, the shooting area of ​​the camera is 20±2cm×15±1cm; the camera automatically focuses when taking pictures; or, in the third step, the starting limit switch and the ending limit switch are respectively contact-triggered mechanical limit switches, and the starting trigger and the ending trigger are respectively triggers that match the contact-triggered mechanical limit switches.

10. The automatic monitoring method for rice paddy planthoppers according to any one of claims 5 to 8, characterized in that, In the fourth step, the data uploaded to the server includes the species and number of planthoppers in the photos obtained by the detection module; if the data to be uploaded to the server also includes photos taken by the camera and / or images processed by the photo processing module, then these data are uploaded to the server together; the server uses the uploaded data to train, optimize, and iterate the detection model; or, in the fourth step, the configuration file of the YOLO model has a p2 detection head.

Images