Unmanned aerial vehicle-mounted fruit tree pest and disease damage intelligent monitoring device

Abstract

The invention discloses an unmanned aerial vehicle-mounted fruit tree disease and insect pest intelligent monitoring device, and relates to the technical field of fruit tree disease and insect pest control, the unmanned aerial vehicle-mounted fruit tree disease and insect pest intelligent monitoring device comprises an unmanned aerial vehicle body platform, an intelligent monitoring device carried on an unmanned aerial vehicle body, and a marking assembly fixedly mounted on an undercarriage of the unmanned aerial vehicle body; the multi-sensor module synchronously collects images and spatial position information, the airborne edge calculation unit performs real-time analysis and quickly outputs disease and insect pest types, severity and geographic coordinates, and after the airborne edge calculation unit analyzes and outputs results, the marking assembly responds to a marking control instruction and performs automatic physical marking on recognized target fruit trees. The analysis result is instantly converted into the on-site visual geographic identification, the data transmission delay is eliminated, the timeliness requirement of dynamic monitoring is met, the problem that it is difficult for workers to quickly and accurately position the pain point of a specific infected fruit tree in a dense orchard is greatly solved, and the efficiency and accuracy of the whole pest control process are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of fruit tree disease and pest control technology, specifically to an unmanned aerial vehicle (UAV)-borne intelligent monitoring device for fruit tree diseases and pests. Background Technology

[0002] As a crucial component of modern agriculture, the fruit tree planting industry's yield and quality are directly affected by pests and diseases. Early and accurate monitoring of these pests and diseases is a key prerequisite for scientific prevention and control, minimizing losses. Traditional methods of monitoring fruit tree pests and diseases primarily rely on manual inspections, depending on the experience and judgment of the personnel. This approach suffers from drawbacks such as low monitoring efficiency, limited coverage, and strong subjectivity. Especially in complex settings like large-scale orchards and mountain orchards, comprehensive and high-frequency systematic monitoring is difficult to achieve, often resulting in pests and diseases being discovered only after they have spread, missing the optimal window for control.

[0003] Existing technologies sometimes combine drones with single sensors for image acquisition, but due to limitations in data processing methods, the acquired data often needs to be transmitted to a ground platform for offline analysis. This results in data transmission delays and poor real-time performance, failing to meet the timeliness requirements of dynamic pest and disease monitoring. Furthermore, even if the general area of ​​pest and disease occurrence is determined through analysis, it is extremely difficult for staff to accurately locate specific infected trees in dense orchards. Often, they need to bring analysis maps back into the field for comparison and searching, a time-consuming and labor-intensive process that significantly reduces the efficiency of initial monitoring. Therefore, this invention proposes a drone-borne intelligent monitoring device for fruit tree pests and diseases. Summary of the Invention

[0004] The purpose of this invention is to provide an unmanned aerial vehicle (UAV)-borne intelligent monitoring device for fruit tree diseases and pests. This addresses the shortcomings of some monitoring devices mentioned in the background art, which attempt to combine UAVs with single sensors for image acquisition. However, due to limitations in data processing methods, the acquired data often needs to be transmitted to a ground platform for offline analysis, resulting in data transmission delays and poor real-time performance, failing to meet the timeliness requirements of dynamic disease and pest monitoring. Furthermore, even if the approximate area of ​​disease and pest occurrence is determined through analysis, it is extremely difficult for staff to accurately locate specific infected fruit trees in dense orchards. Often, they need to bring analysis maps back into the field for comparison and searching, a time-consuming and labor-intensive process that significantly reduces the efficiency of initial monitoring.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A drone-borne intelligent monitoring device for fruit tree diseases and pests includes a drone platform, an intelligent monitoring device mounted on the drone body, and a marking component fixedly installed on the drone body's landing gear.

[0007] The intelligent monitoring device integrates a multi-sensor module, an airborne edge computing unit, and a communication module. The multi-sensor module is used to synchronously collect image information and spatial location information of the target fruit tree. The airborne edge computing unit is communicatively connected to the multi-sensor module and is used to receive and process the image information and spatial location information. The communication module is communicatively connected to the airborne edge computing unit and is used to transmit the processing results to external devices.

[0008] The airborne edge computing unit has a built-in pest and disease identification algorithm model, which can analyze the image information collected by the multi-sensor module in real time and output identification results including the type, severity and geographic coordinates of pests and diseases. The airborne edge computing unit is communicatively connected to the marking component and generates marking control commands based on the identification results. The marking component responds to the marking control commands and performs automatic physical marking on the identified target fruit trees.

[0009] Optionally, the multi-sensor module includes a visible light camera, a multispectral camera, and a cooperative positioning unit. The visible light camera is used to acquire RGB morphological images of fruit trees, the multispectral camera is used to acquire multispectral images including red-edge and near-infrared bands, and the cooperative positioning unit integrates a high-precision GNSS receiver and an inertial measurement unit to provide synchronized spatiotemporal position and attitude data for the image information.

[0010] Optionally, the communication module adopts a redundant design combining G / G mobile communication technology and integrated image / data transmission wireless communication technology.

[0011] Optionally, the airborne edge computing unit further includes a data processing unit, which is configured to preprocess, register and fuse the heterogeneous data collected by the multi-sensor module, and input the processed standardized data into the pest and disease identification algorithm model.

[0012] Optionally, it also includes a ground control station and software platform, which are connected to the communication module to receive and display the identification results and generate the flight path of the UAV body.

[0013] Optionally, the airborne edge computing unit is configured to execute a hierarchical decision-making strategy. When the identification result indicates that the severity of the pest or disease is below a first threshold, only the geographic coordinates are recorded. When the severity reaches or exceeds the first threshold but is below a higher second threshold, the tagging control command is generated. When the severity reaches or exceeds the second threshold, in addition to generating the tagging command, an emergency alarm message is sent to the ground control station and software platform through the communication module.

[0014] Optionally, the marking assembly includes a connecting steel frame, a support plate, a mounting plate, a rotating rod, a mounting component, a marking gun, and a drive motor. The connecting steel frame is welded to the landing gear of the UAV body. The support plate is fixedly connected to the bottom of the connecting steel frame. The mounting plate is symmetrically fixed to the bottom of the support plate. The rotating rod passes through the mounting plate laterally and rotates with the mounting plate through a bearing. One end of the rotating rod is fixedly connected to the output shaft of the drive motor. The marking gun is fixedly mounted on the rotating rod through an externally provided mounting component.

[0015] Optionally, the marking assembly further includes a pigment tank, a delivery pump, and a filling pipe. The pigment tank is located on top of the support plate. The input end of the delivery pump is connected to the pigment tank, and its output end is fixedly connected to the filling pipe. The other end of the filling pipe is connected to the feed port of the marking gun.

[0016] The beneficial effects of this invention are:

[0017] In this invention, images and spatial location information are collected synchronously by a multi-sensor module, and the airborne edge computing unit analyzes them in real time, quickly outputting the type, severity, and geographic coordinates of pests and diseases. After the airborne edge computing unit analyzes and outputs the results, the marking component responds to the marking control command and performs automatic physical marking on the identified target fruit trees, instantly converting the analysis results into on-site visualized geographic markers. This eliminates data transmission delays, meets the timeliness requirements of dynamic monitoring, and greatly solves the pain point of staff having difficulty quickly and accurately locating specific diseased fruit trees in dense orchards, significantly improving the efficiency and accuracy of the entire pest and disease control process. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a schematic diagram of the structure of an unmanned aerial vehicle (UAV)-borne intelligent monitoring device for fruit tree diseases and pests according to the present invention;

[0020] Figure 2 This is a front view of the present invention;

[0021] Figure 3 This is a schematic diagram of the marking component in this invention;

[0022] Figure 4 This is a schematic diagram of the marking component from another perspective in this invention;

[0023] Figure 5 This is a general block diagram of the present invention;

[0024] Figure 6 This is a system block diagram of the intelligent monitoring device in this invention.

[0025] The numbers on the map are:

[0026] 1. The drone itself; 2. Intelligent monitoring device;

[0027] 3. Marking assembly; 301. Connecting steel frame; 302. Bearing plate; 303. Mounting plate; 304. Rotating rod; 305. Mounting component; 306. Marking gun; 307. Drive motor; 308. Pigment box; 309. Conveying pump; 310. Filling pipe. Detailed Implementation

[0028] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0029] Example 1:

[0030] As attached Figure 1 To be continued Figure 6 As shown, the present invention provides an unmanned aerial vehicle (UAV)-borne intelligent monitoring device for fruit tree diseases and pests, including a UAV body 1 platform, an intelligent monitoring device 2 mounted on the UAV body 1, and a marking component 3 fixedly installed on the landing gear of the UAV body 1. By mounting the intelligent monitoring device 2 on the UAV body 1, the UAV's flexible flight capability can be used to quickly and efficiently cover a large area of ​​orchard and monitor fruit trees in different locations. The marking component 3 is installed on the landing gear to facilitate accurate marking after identifying diseased and pest-infested fruit trees, providing clear instructions for subsequent treatment.

[0031] The intelligent monitoring device 2 integrates a multi-sensor module, an airborne edge computing unit, and a communication module. The multi-sensor module is used to synchronously collect image information and spatial location information of the target fruit tree. The airborne edge computing unit is connected to the multi-sensor module and is used to receive and process image information and spatial location information. The communication module is connected to the airborne edge computing unit and is used to transmit the processing results to external devices.

[0032] The airborne edge computing unit incorporates a pest and disease identification algorithm model. This model analyzes image information collected by multiple sensor modules in real time and outputs identification results including pest and disease type, severity, and geographic coordinates. The airborne edge computing unit communicates with the marking component 3 and generates marking control commands based on the identification results. The marking component 3 responds to these commands by automatically physically marking the identified target fruit trees. Specifically, the pest and disease identification algorithm model is trained on a large amount of fruit tree pest and disease image data, enabling accurate identification of pest and disease types through image feature analysis. Furthermore, by combining spatial location information, the algorithm model can also assess the severity of pests and diseases. The airborne edge computing unit generates marking control commands based on these identification results. Upon receiving the commands, the marking component 3 physically marks the target fruit trees through spraying, facilitating subsequent manual processing or further monitoring. Through the linkage between the algorithm model and the marking component 3, automatic identification and marking of fruit tree pests and diseases are achieved, improving the timeliness and targeted nature of pest and disease control.

[0033] In one embodiment of the present invention, a ground control station and a software platform are also included, which are connected to a communication module to receive and display the identification results and generate the flight path of the UAV body 1. Specifically, the ground control station and software platform adopt a dual-function design of "data visualization + task planning". Through data analysis algorithms, the received identification results are transformed into intuitive heat maps of pest and disease distribution and statistical lists of types, making it easy for staff to quickly grasp the orchard's disease situation. The flight path generation is based on a digital map of the orchard (including fruit tree location, row spacing, and terrain data), and a path optimization algorithm (such as a greedy algorithm or A* algorithm) is used to plan a full-coverage, non-repeating flight path. At the same time, combined with the UAV's endurance and sensor monitoring range, the flight area and take-off and landing points are automatically divided to improve monitoring efficiency. The path optimization algorithm (such as a greedy algorithm or A* algorithm) is a prior art and will not be described in detail here.

[0034] In one embodiment of the present invention, the multi-sensor module includes a visible light camera, a multispectral camera, and a cooperative positioning unit. The visible light camera is used to acquire RGB morphological images of the fruit tree, the multispectral camera is used to acquire multispectral images including the red-edge and near-infrared bands, and the cooperative positioning unit integrates a high-precision GNSS receiver and an inertial measurement unit to provide synchronized spatiotemporal position and attitude data for the image information. Specifically, the RGB morphological images acquired by the visible light camera can clearly present the appearance, color, and other characteristics of the fruit tree, helping to detect obvious pest and disease symptoms on the fruit tree surface. The multispectral images acquired by the multispectral camera contain optical information in different bands. Different bands have different responses to the physiological state and pest and disease conditions of the fruit tree. By analyzing these multispectral images, a deeper understanding of the health status of the fruit tree can be obtained. The high-precision GNSS receiver in the cooperative positioning unit can accurately determine the geographical location of the UAV, while the inertial measurement unit can measure the attitude information of the UAV, such as pitch angle, roll angle, and yaw angle. Synchronizing these spatiotemporal position and attitude data with the image information can provide accurate geographic reference and attitude correction for subsequent image analysis and processing. The multi-sensor module acquires fruit tree information from multiple angles through a combination of different types of sensors and provides accurate spatiotemporal positioning, providing a data foundation for comprehensive and accurate monitoring of fruit tree diseases and pests.

[0035] It should be further described that the high-precision GNSS receiver in the cooperative positioning unit is a positioning module that supports real-time dynamic differential technology.

[0036] In one embodiment of the present invention, the communication module employs a redundant design combining 4G / 5G mobile communication technology and integrated image / data transmission wireless communication technology. By adopting this redundant design, when one communication technology fails or has a poor signal, another communication technology can immediately take over, ensuring continuous data transmission and improving the reliability and stability of communication.

[0037] In one embodiment of the present invention, the airborne edge computing unit further includes a data processing unit configured to preprocess, register, and fuse heterogeneous data collected by the multi-sensor modules, and input the processed standardized data into the pest and disease identification algorithm model. Specifically, the data collected by the multi-sensor modules may differ in format, accuracy, etc., and are considered heterogeneous data. The data processing unit first preprocesses this heterogeneous data, such as denoising and normalization, to improve data quality and consistency. Then, it performs registration to align the data acquired by different sensors spatially and temporally, ensuring accurate data correspondence. Finally, it performs fusion to comprehensively analyze the data from multiple sensors and extract more valuable information. The processed standardized data can better meet the requirements of the pest and disease identification algorithm model, improving the recognition accuracy and efficiency of the algorithm model.

[0038] In one embodiment of the present invention, the airborne edge computing unit is configured to execute a hierarchical decision-making strategy. When the identification result indicates that the severity of pests and diseases is below a first threshold, only the geographic coordinates are recorded for subsequent tracking and statistics. When the severity reaches or exceeds the first threshold but is below a higher second threshold, a marking control command is generated. When the severity reaches or exceeds the second threshold, in addition to generating the marking command, an emergency alarm message is sent to the ground control station and software platform via a communication module, enabling ground operators to understand the situation in a timely manner and take emergency measures to prevent further spread of pests and diseases. The hierarchical decision-making strategy differentiates treatment based on the severity of pests and diseases, improving the efficiency and targeting of pest and disease control, and enabling timely response to different levels of pest and disease threats.

[0039] Specifically, the hierarchical decision-making strategy executed by the airborne edge computing unit is based on a comprehensive analysis and scientific demonstration of a large amount of experimental data on fruit tree diseases and pests, the experience of agricultural disease and pest control experts, and the actual needs of the fruit tree planting industry.

[0040] During the research and development phase, researchers collected detailed data on various fruit trees from different regions encountering different pests and diseases under various growing conditions. This data included information on multiple dimensions such as the type of pests and diseases, the extent of infection, the degree of impact on the physiological functions of fruit trees (e.g., the proportion of leaf damage, fruit rot rate, branch wilting, etc.), and the potential damage to fruit yield and quality. Through in-depth mining and analysis of massive amounts of experimental data, using statistical methods and models, and combining expert advice, a grading standard was determined that accurately reflects the severity of pests and diseases and their impact on fruit trees.

[0041] Example 2:

[0042] like Figure 1-4 As shown, in one embodiment of the present invention, the marking assembly 3 includes a connecting steel frame 301, a support plate 302, a mounting plate 303, a rotating rod 304, a mounting component 305, a marking gun 306, and a drive motor 307. The connecting steel frame 301 is welded to the landing gear of the UAV body 1. The support plate 302 is fixedly connected to the bottom of the connecting steel frame 301. The mounting plate 303 is symmetrically fixed to the bottom of the support plate 302. The rotating rod 304 passes through the mounting plate 303 laterally and rotates with the mounting plate 303 through a bearing. One end of the rotating rod 304 is fixedly connected to the output shaft of the drive motor 307. The marking gun 306 is fixedly mounted on the rotating rod 304 through the externally provided mounting component 305.

[0043] The marking assembly 3 also includes a pigment tank 308, a delivery pump 309, and a filling pipe 310. The pigment tank 308 is located on the top of the support plate 302. The input end of the delivery pump 309 is connected to the pigment tank 308, and its output end is fixedly connected to the filling pipe 310. The other end of the filling pipe 310 is connected to the feed port of the marking gun 306.

[0044] Specifically, after the output results are analyzed by the onboard edge computing unit, an electrical signal is transmitted to the marking component 3. The marking component 3 responds to the marking control command and performs automatic physical marking on the identified target fruit trees. Upon receiving the electrical signal command, the drive motor 307 starts, and its output shaft begins to rotate. Since the output shaft of the drive motor 307 is fixedly connected to one end of the rotating rod 304, the rotation of the output shaft will drive the rotating rod 304 to rotate synchronously. The rotating rod 304 is rotatably engaged with the mounting plates 303, which are symmetrically fixed to the bottom of the support plate 302, via bearings. The bearings reduce rotational resistance, ensuring smooth rotation of the rotating rod 304. The rotation of the rotating rod 304 is transmitted to the marking gun 306 through the mounting component 305, causing the marking gun 306 to adjust its angle around the axis of the rotating rod 304 until the spray direction of the marking gun 306 accurately points to the target fruit tree, thus meeting the angle requirements for marking target fruit trees in different locations and complex terrains. Simultaneously, the delivery pump 309 starts working after receiving an electrical signal command. Its input end is connected to the pigment tank 308 located on top of the support plate 302. The delivery pump 309 generates negative pressure in the pigment tank 308, drawing the pigment out of the pigment tank 308. The drawn pigment enters the filling pipe 310 through the output end of the delivery pump 309. The filling pipe 310 accurately delivers the pigment to the feed port of the marking gun 306, ensuring that the marking gun 306 can continuously receive a stable supply of pigment. After receiving the pigment, the marking gun 306 sprays the pigment with an appropriate spraying method and force according to its internal preset nozzle design and pressure control parameters. The sprayed pigment accurately lands on the target fruit tree, forming a clear and durable physical mark, which facilitates subsequent targeted management and treatment of the target fruit tree.

[0045] During use, the drone flies along a pre-set flight path from the ground control station and software platform, relying on its onboard intelligent monitoring device 2 to monitor fruit tree diseases and pests. During flight, the multi-sensor modules within the intelligent monitoring device 2 operate synchronously, acquiring RGB morphological images of the target fruit tree, multispectral images containing red edges and near-infrared bands, and synchronous spatiotemporal position and attitude data through a visible light camera, a multispectral camera, and a collaborative positioning unit. The onboard edge computing unit's data processing unit then preprocesses, registers, and fuses this heterogeneous data. Subsequently, the built-in disease and pest identification algorithm model performs real-time analysis based on the processed standardized data, outputting identification results including disease and pest types, severity, and geographic coordinates. This achieves real-time analysis and monitoring of fruit tree diseases and pests, effectively solving the problems of low monitoring efficiency, limited coverage, and strong subjectivity inherent in traditional manual inspection methods. This addresses the issue of data transmission delays preventing the timely fulfillment of dynamic monitoring requirements. Simultaneously, the airborne edge computing unit executes a tiered decision-making strategy based on the identification results. When the severity of pests and diseases is below the first threshold, only geographical coordinates are recorded. When the severity reaches or exceeds the first threshold but is below the second threshold, a marking control command is generated. When the severity reaches or exceeds the second threshold, in addition to generating a marking command, an emergency alarm message is sent, and the marking control command is transmitted to the marking component 3. Upon receiving the command, the marking component 3 drives the motor 307 to rotate the rotating rod 304, causing the marking gun 306 to adjust its angle and point towards the target fruit tree. Simultaneously, the delivery pump 309 delivers pigment from the pigment box 308 to the marking gun 306 for spraying, forming a clear and durable physical mark. This solves the problem of time-consuming and labor-intensive work for staff to accurately locate diseased fruit trees in dense orchards, greatly improving the efficiency and accuracy of fruit tree pest and disease monitoring and control.

[0046] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A drone-borne intelligent monitoring device for fruit tree diseases and pests, comprising a drone body (1) platform, an intelligent monitoring device (2) mounted on the drone body (1), and a marking assembly (3) fixedly installed on the landing gear of the drone body (1), characterized in that: The intelligent monitoring device (2) integrates a multi-sensor module, an airborne edge computing unit, and a communication module. The multi-sensor module is used to synchronously collect image information and spatial location information of the target fruit tree. The airborne edge computing unit is communicatively connected to the multi-sensor module and is used to receive and process the image information and spatial location information. The communication module is communicatively connected to the airborne edge computing unit and is used to transmit the processing results to external devices. The airborne edge computing unit has a built-in pest and disease identification algorithm model, which can analyze the image information collected by the multi-sensor module in real time and output the identification results including the type, severity and geographical coordinates of pests and diseases. The airborne edge computing unit is connected to the marking component (3) and generates marking control instructions based on the identification results. The marking component (3) responds to the marking control instructions and performs automatic physical marking on the identified target fruit trees.

2. The unmanned aerial vehicle-borne intelligent monitoring device for fruit tree diseases and pests according to claim 1, characterized in that: The multi-sensor module includes a visible light camera, a multispectral camera, and a cooperative positioning unit. The visible light camera is used to acquire RGB morphological images of fruit trees, the multispectral camera is used to acquire multispectral images including red-edge and near-infrared bands, and the cooperative positioning unit integrates a high-precision GNSS receiver and an inertial measurement unit to provide synchronized spatiotemporal position and attitude data for the image information.

3. The unmanned aerial vehicle-borne intelligent monitoring device for fruit tree diseases and pests according to claim 2, characterized in that: The communication module adopts a redundant design that combines 4G / 5G mobile communication technology with integrated wireless communication technology for image transmission / data transmission.

4. The unmanned aerial vehicle-borne intelligent monitoring device for fruit tree diseases and pests according to claim 3, characterized in that: The airborne edge computing unit also includes a data processing unit, which is configured to preprocess, register and fuse the heterogeneous data collected by the multi-sensor module, and input the processed standardized data into the pest and disease identification algorithm model.

5. The unmanned aerial vehicle-borne intelligent monitoring device for fruit tree diseases and pests according to claim 1, characterized in that: It also includes a ground control station and a software platform, which are connected to the communication module to receive and display the identification results and generate the flight path of the UAV body (1).

6. The unmanned aerial vehicle-borne intelligent monitoring device for fruit tree diseases and pests according to claim 5, characterized in that: The airborne edge computing unit is configured to execute a hierarchical decision-making strategy. When the identification result indicates that the severity of the pests and diseases is below a first threshold, only the geographic coordinates are recorded. When the severity reaches or exceeds the first threshold but is below a higher second threshold, the tagging control command is generated. When the severity reaches or exceeds the second threshold, in addition to generating the tagging command, an emergency alarm message is sent to the ground control station and software platform through the communication module.

7. The unmanned aerial vehicle-borne intelligent monitoring device for fruit tree diseases and pests according to claim 1, characterized in that: The marking assembly (3) includes a connecting steel frame (301), a bearing plate (302), a mounting plate (303), a rotating rod (304), a mounting component (305), a marking gun (306), and a drive motor (307). The connecting steel frame (301) is welded to the landing gear of the UAV body (1). The bottom of the connecting steel frame (301) is fixedly connected to the bearing plate (302). The mounting plate (303) is symmetrically fixed to the bottom of the bearing plate (302). The rotating rod (304) passes through the mounting plate (303) laterally and rotates with the mounting plate (303) through a bearing. One end of the rotating rod (304) is fixedly connected to the output shaft of the drive motor (307). The marking gun (306) is fixedly mounted on the rotating rod (304) through the external mounting component (305).

8. The unmanned aerial vehicle-borne intelligent monitoring device for fruit tree diseases and pests according to claim 7, characterized in that: The marking assembly (3) also includes a pigment tank (308), a delivery pump (309), and a filling pipe (310). The pigment tank (308) is located on the top of the support plate (302). The input end of the delivery pump (309) is connected to the pigment tank (308), and its output end is fixedly connected to the filling pipe (310). The other end of the filling pipe (310) is connected to the feed port of the marking gun (306).

Images