Intelligent agricultural pest monitoring and spraying method and device based on YOLO v10

Abstract

The invention discloses an intelligent agricultural pest monitoring and spraying method and device based on YOLO v10, and belongs to the technical field of agricultural information. The method comprises the following steps: establishing a disease and pest comparison database by constructing a convolutional neural network model; by setting a nutrient supplement standard, soil humidity upper and lower limits and fertilizer concentration upper and lower limits, timely and appropriate supply of water and fertilizer is ensured; soil nutrient and moisture information is monitored in real time through a soil nutrient and moisture monitoring probe; shooting a monitoring area through a camera of the pest and disease monitoring device, and comparing the obtained picture with a pest and disease comparison database; by arranging a corresponding spraying device, real-time monitoring is achieved, and nutrients, water or chemicals are sprayed. The system has the advantages of high efficiency, accuracy, environmental protection and the like, can meet the requirements of modern agriculture on disease and pest control, and can realize intelligent agricultural disease and pest monitoring and control and integrated spraying of water, fertilizer and pesticide.

Description

Technical Field

[0001] The present invention belongs to the field of agricultural information technology, and in particular relates to an intelligent agricultural pest monitoring and spraying method and device based on YOLO v10. Background Art

[0002] Modern agriculture faces numerous challenges, with crop pests and diseases being particularly prominent. With global climate change and the diversification of cropping patterns, the frequency and severity of pests and diseases are increasing. Traditional pest and disease monitoring and control methods are no longer sufficient to meet the demands of modern agricultural production. In the field of smart agriculture, pest and disease control is crucial. Previous research has focused on responding to pests and diseases after they occur, such as using pattern recognition technology to diagnose and treat affected crops. However, in agricultural production, irrigation and fertilizer usage have significant impacts on pests and diseases. Improper irrigation can lead to excessive field humidity, fostering the growth of pests and diseases. Improper fertilizer usage, whether too little, resulting in crop malnutrition, or excessive, altering the soil environment, can also make crops susceptible to pests and diseases.

[0003] Traditional pest and disease monitoring relies primarily on manual observation. This method is extremely inefficient and difficult to guarantee accuracy. Observer experience and subjective factors significantly influence the results, often failing to detect early symptoms of pests and diseases, leading to their spread. Traditional pest and disease control primarily relies on pesticide spraying. While this method can control pests and diseases to a certain extent, long-term use presents a series of problems. For example, pesticide residues affect the quality and safety of agricultural products, pollute the environment, and promote pesticide resistance in pests and diseases. Furthermore, large-scale spraying of chemical pesticides can harm non-target organisms and disrupt the ecological balance.

[0004] As demand for agricultural product quality increases, the market demand for green and environmentally friendly pest control technologies and equipment is growing. Agricultural producers are also eager for more efficient and precise technologies to ensure crop yield and quality while reducing production costs. With increasing public awareness of environmental protection, traditional agricultural production methods are being questioned for their negative impacts on the environment. Therefore, the development of technologies that can effectively control pests and diseases while minimizing environmental impact has become a top priority. Summary of the Invention

[0005] In response to the above-mentioned technical problems, the present invention provides an intelligent agricultural pest and disease monitoring and spraying method and device based on YOLO v10. This application not only includes an advanced graphic recognition system for the identification and prevention of pests and diseases, but also integrates intelligent irrigation and fertilization control devices to comprehensively ensure the healthy growth of crops and reduce the occurrence of pests and diseases.

[0006] The object of the present invention is achieved through the following technical solutions:

[0007] The present invention provides an intelligent agricultural pest monitoring and spraying method based on YOLO v10, which is carried out in the following steps:

[0008] S1: Build a convolutional neural network model and establish a pest and disease comparison database;

[0009] S2: Set nutrient supplement standards, upper and lower limits for soil moisture, and upper and lower limits for fertilizer concentration;

[0010] S3: Real-time monitoring and prevention: Monitor soil nutrient and moisture information through the soil nutrient and moisture probes of the soil nutrient and moisture monitoring device; take pictures of the monitored area through the camera of the pest and disease monitoring device, and compare the obtained pictures with the pest and disease comparison database in step S1. If the camera detects that the leaf color is abnormal, and the nutrient and moisture contents in the monitored soil are normal at this time, it is determined that pests and diseases have occurred. According to the location of the pest and disease identified by the camera, the drug spraying device is started to spray the drug; if the camera detects that the leaf color is abnormal, and the nutrient and moisture contents in the monitored soil are less than the normal amount, the fertilizer pump or water pump is started to spray fertilizer or water containing the missing nutrients.

[0011] Furthermore, when spraying fertilizer, water and pesticides, the horizontal position of the nozzle is controlled by two mutually perpendicular synchronous belt slides, and the vertical position of the nozzle is controlled by a telescopic rod. The distance information between the crop and the nozzle is obtained by the laser ranging sensor module. The central control module controls the telescopic rod on which the nozzle is installed to move horizontally along the synchronous belt slide II, or to move horizontally along the synchronous belt slide I along the synchronous belt slide II, so that the nozzle is directly above the blade. The central control module controls the extension and retraction length of the telescopic rod to ensure that the nozzle and the blade always maintain a distance of 5 cm.

[0012] Furthermore, the upper and lower limits of soil moisture and fertilizer concentration in step S2 are set as follows:

[0013] (1) Upper and lower limits of soil moisture

[0014] The lower limit of soil moisture is set between 50% and 60% of field water holding capacity, and the upper limit of soil moisture is set between 80% and 90% of field water holding capacity;

[0015] (2) Upper and lower limits of fertilizer concentration

[0016] In the early stages of crop growth, the lower limit of nitrogen fertilizer concentration is set at 100-200 mg / L, and the lower limit of phosphorus fertilizer and potassium fertilizer concentration is set at 30-40 mg / L; in the vigorous growth period, the lower limit of nitrogen fertilizer concentration is set at 300-400 mg / L, and the lower limit of phosphorus fertilizer and potassium fertilizer concentration is set at 60-80 mg / L;

[0017] The upper limit of nitrogen fertilizer concentration is set at 500-600 mg / L, and the upper limit of phosphorus fertilizer and potassium fertilizer concentration is set at 100-120 mg / L;

[0018] (3) Nutrient supplementation standards are as follows:

[0019] When the soil nitrogen content is less than 30mg / kg, the nitrogen fertilizer pump is turned on; when the soil nitrogen content is greater than 60mg / kg, the nitrogen fertilizer pump is turned off;

[0020] When the soil phosphorus content is less than 10mg / kg, the phosphate fertilizer pump is turned on; when the soil phosphorus content is greater than 20mg / kg, the phosphate fertilizer pump is turned off;

[0021] When the soil potassium content is less than 60mg / kg, the potassium fertilizer pump is turned on; when the soil potassium content is greater than 100mg / kg, the potassium fertilizer pump is turned off;

[0022] When the soil moisture content is less than 50%, f ,θ f is the field water holding capacity, turn on the water pump, when the soil moisture content is greater than 90%θ f Turn off the water pump.

[0023] Furthermore, in step S1, a convolutional neural network model is constructed, specifically:

[0024] First, at least 3,000 photos of crops in normal and various pathological states were taken from multiple angles within the monitoring area. These photos were then fed into YOLO v10 for learning. When the red value (R) of the leaves was below 80-100, the green value (G) was below 120-140, and the blue value (B) was below 80-100, the leaves were identified as having pests and diseases, and the images were labeled accordingly, completing the construction of the convolutional neural network model.

[0025] Then, the pest and disease leaves learned by YOLO v10 are labeled, and the pest and disease comparison database obtained after learning is transferred to the Raspberry Pi unit;

[0026] The Raspberry Pi's built-in YOLO v10-based image recognition module analyzes images captured by the camera using its algorithm to identify the type, severity, and location of pests and diseases. This information is then transmitted to the central control module, where the corresponding database selects the appropriate pesticide type and dosage based on the type and severity of pests and diseases on the crop leaves.

[0027] The spraying device of the intelligent agricultural pest and disease monitoring and spraying method based on YOLO v10 described in the present invention is characterized in that it includes a support frame, a pest and disease monitoring device, a soil nutrient and moisture monitoring device, a fertilizer spraying device, a water spraying device, a drug spraying device and a control device. The support frame is a rectangular frame, and a synchronous belt slide bar I is symmetrically arranged between two columns on two opposite sides of the frame. A baffle is arranged outside the lower side of one side, and a drug pump, a water pump and a fertilizer pump are arranged on it. A synchronous belt slide bar II is vertically slidably connected to the two synchronous belt slide bars I, and a support beam parallel to the synchronous belt slide bar II is connected to the two opposite cross beams at the top. The support beam is provided with a control device and a pest and disease monitoring device. The camera and the soil nutrient moisture monitoring probe of the soil nutrient moisture monitoring device are placed on the bottom crossbeam on one side. The drug spraying device, the water spraying device, and the fertilizer spraying device are connected to different nozzles respectively, and the nozzles are connected to the synchronous belt slide bar II through a telescopic rod. The pest and disease monitoring device and the soil nutrient moisture monitoring device are respectively connected to the input end of the control device. The drug spraying device, the water spraying device, the fertilizer spraying device, the telescopic rod and the synchronous belt slide bars I and II are respectively connected to the output end of the control device. The crop status is monitored by the disease and pest monitoring device and the soil nutrient moisture monitoring device, and the fertilization, water spraying or pesticide application area and amount are identified and controlled through the database in the control device.

[0028] Furthermore, the control device includes a Raspberry Pi unit, a central control module, and a data storage unit. The central control module is respectively connected to the Raspberry Pi unit and the data storage unit for communication. The input end of the Raspberry Pi unit is respectively connected to the camera of the pest and disease monitoring device and the soil nutrient and moisture monitoring probe of the soil nutrient and moisture monitoring device; the output end of the central control module is also respectively connected to the pesticide pump relay V of the drug spraying device, the water pump relay IV of the water spraying device, the relays I to III of the fertilizer spraying device, the synchronous belt slide rods I and II and the telescopic rod to control the start and stop of each component.

[0029] Furthermore, the pest and disease monitoring device includes a camera and a Raspberry Pi unit of the control device. The camera is used to collect image information of the monitoring area and connect and transmit it to the Raspberry Pi unit of the control device via a data cable. The pest and disease monitoring device is identified and monitored through the built-in YOLO v10-based image recognition module in the control device.

[0030] Furthermore, the soil nutrient moisture monitoring device includes a soil nutrient moisture monitoring probe and a Raspberry Pi unit of the control device. The soil nutrient moisture monitoring probe is used to collect the current nitrogen, phosphorus, potassium content and soil moisture content information of the soil, and is connected and transmitted to the Raspberry Pi unit of the control device via a data cable for monitoring by the control device. When the measured value is lower than the set threshold, the central control module of the control device will automatically control the operation of the fertilizer spraying device and the water spraying device to replenish the required fertilizer and water.

[0031] Furthermore, the fertilizer spraying device includes a nitrogen pump, a phosphorus pump, a potassium pump, and relays I to III; the water spraying device includes a water pump and relay IV; the water pump is used to provide liquid delivery power; the nitrogen pump, phosphorus pump, and potassium pump are respectively connected to the central control module through relays I to IV, and the central control module controls each relay switch, thereby controlling the operation of various pumps.

[0032] Furthermore, the drug spraying device includes a pesticide pump and a relay V. The pesticide pump and relay V are connected to the central control module. The pesticide pump provides power for the medicine liquid. When the pest and disease monitoring device detects the occurrence of pests and diseases, the central control module controls the relay V to open and the medicine liquid is sprayed through the pesticide pump.

[0033] The beneficial effects of the present invention are:

[0034] 1. This invention utilizes RGB-based pattern recognition technology for pest and disease treatment, while also incorporating irrigation and fertilization control into the prevention and control system. Real-time soil nutrient and moisture monitoring probes can immediately replenish nutrients or water when the crop's growing environment is experiencing nutrient or water deficits. The system can automatically identify pest and disease outbreaks and automatically apply pesticides, eliminating the need for users to worry about determining the type and extent of pests due to lack of experience.

[0035] 2. There is a difference in priority between spraying and fertigation in the present invention, with fertigation taking precedence over spraying. If the camera detects an abnormal leaf color, and the nutrient and moisture content in the soil are both normal, it is determined that a pest has occurred, and the camera will automatically identify the location of the pest and apply spraying. If the camera detects an abnormal leaf color, and the nutrient and moisture content in the soil is less than normal, the fertilizer pump or water pump is activated to spray fertilizer or water containing specific nutrients. This combination provides a more effective solution for agricultural pest control.

[0036] 3. The present invention uses convolutional neural network model (CNN) training and previously established crop disease database and crop pesticide usage database to select appropriate pesticide types and dosages based on the crop type and the characteristics and degree of leaf pests and diseases, thereby generating a personalized pesticide spraying plan.

[0037] 4. The device of the present invention accurately sprays water, fertilizer and medicine on the surface of leaves through mutually perpendicular synchronous belt slide bars (I to II) and vertical telescopic rods, ensuring that the nozzles are facing the crop leaves and at a distance of 5 cm. While promoting the effective use of water, fertilizer and medicine, it also reduces pesticide residues on the surface of crop fruits, thereby promoting green food production.

[0038] 5. This invention can provide protection against accidental collisions, correcting manual errors and preventing excessive watering or fertilization, thus preventing pests and diseases from occurring in multiple ways. It is suitable for low-growing leafy crops such as melons and strawberries. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Flow chart of the method of the present invention.

[0040] Figure 2 This is a schematic diagram of the control device connection.

[0041] Figure 3 for Figure 2 Wiring diagram.

[0042] Figure 4 for Figure 2 Schematic diagram of the Raspberry Pi unit.

[0043] Figure 5 It is a schematic diagram of the three-dimensional structure of the device of the present invention.

[0044] Figure 6 It is a front view of the device of the present invention.

[0045] Figure 7 for Figure 6 Right view of .

[0046] Figure 8 for Figure 6 Top view of .

[0047] Figure 9 for Figure 6 A partial view of the middle synchronous belt slide.

[0048] Figure 10 for Figure 9 Partial diagram of the connection between the middle pulley and the track.

[0049] In the figure: 1-column, 2-baffle, 3-pump group, 4-relay group, 51-synchronous belt slide rod I, 52-synchronous belt slide rod II, 6-Raspberry Pi unit, 7-camera, 8-telescopic rod, 9-sprinkler, 10-power supply, 11-motor, 12-slide rod tail buckle, 13-pillar fixing part, 14-slide rod fixing part, 15-slider, 16-roller, 17-Raspberry Pi unit power interface, 18-GPIO interface, 19-soil nutrient and moisture monitoring probe. DETAILED DESCRIPTION

[0050] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0051] Example: Figure 1-Figure 4 As shown, the present invention provides an intelligent agricultural pest monitoring and spraying method based on YOLO v10, which is carried out in the following steps:

[0052] S1: Construct a convolutional neural network model and establish a pest and disease comparison database; the method for constructing the convolutional neural network model is:

[0053] First, at least 3,000 photos of crops in normal and various pathological states were taken from multiple angles in the monitoring area. These photos were then fed into YOLO v10 for learning. When the red value (R) of the leaves was below 80-100, the green value (G) was below 120-140, and the blue value (B) was below 80-100, the leaves were identified as having pests and diseases, and the images were labeled accordingly, completing the construction of the convolutional neural network model.

[0054] Obtaining a pest and disease comparison database: After labeling the pest and disease leaves learned by YOLO v10, the pest and disease comparison database obtained after learning is transferred to the Raspberry Pi unit 6;

[0055] The Raspberry Pi has a built-in image recognition module based on YOLO v10. It analyzes images captured by the camera using the image recognition module's (existing) algorithm to identify the type, severity, and location of pests and diseases, and transmits this information to the central control module. The corresponding database in the central control module selects the appropriate type and dosage of pesticide based on the type and severity of different pests and diseases on crop leaves.

[0056] In YOLO v10, learning is done using a database of photos that are initially fed into a Raspberry Pi unit.

[0057] Among them, the crops selected in this example are common crops such as melons and strawberries.

[0058] The model has an accuracy rate of over 85% in monitoring pests and diseases of melons, strawberries, etc.

[0059] S2: Set nutrient supplement standards, upper and lower limits for soil moisture, and upper and lower limits for fertilizer concentration;

[0060] (1) Upper and lower limits of soil moisture

[0061] The lower limit of soil moisture is set between 50% and 60% of the field water holding capacity. In order to prevent excessive soil moisture caused by over-irrigation, which affects crop growth and soil aeration, the upper limit of soil moisture is set between 80% and 90% of the field water holding capacity. The soil moisture is monitored in real time by the soil nutrient moisture monitoring probe 19. When the moisture is lower than the lower limit, the central control module of the control device automatically starts the water pump for irrigation; when the soil moisture reaches the upper limit during irrigation, the central control module of the control device automatically stops the water pump and stops irrigation.

[0062] (2) Set the upper and lower limits of fertilizer concentration

[0063] The lower limit of fertilizer concentration is determined according to the nutrient demand at different growth stages. In the early growth stages of crops, such as melons and strawberries, the nitrogen demand is relatively low, and the lower limit of nitrogen fertilizer concentration is set at 100-200 mg / L, and the lower limit of phosphorus fertilizer and potassium fertilizer concentration is set at 30-40 mg / L. During the vigorous growth period, the lower limit of nitrogen fertilizer concentration is set at 300-400 mg / L, and the lower limit of phosphorus fertilizer and potassium fertilizer concentration is set at 60-80 mg / L.

[0064] The upper limit of nitrogen fertilizer concentration is set at 500-600 mg / L, and the upper limit of phosphorus fertilizer and potassium fertilizer concentration is 100-120 mg / L; when the concentration is lower than the lower limit, the fertilization device is automatically adjusted to increase the fertilizer input; when the concentration reaches the upper limit, fertilization is stopped to ensure the rationality and safety of fertilization.

[0065] The fertilizer concentration is monitored by the soil nutrient moisture monitoring probe 19. When the concentration of nitrogen fertilizer, phosphorus fertilizer or potassium fertilizer is lower than the set lower limit, the central control module automatically controls the fertilization device to increase the fertilizer input, and at the same time controls the corresponding nitrogen pump, phosphorus pump or potassium pump to start fertilization; when the concentration reaches the upper limit, fertilization is stopped.

[0066] (3) The soil nutrient supplementation standards are as follows;

[0067] When the soil nitrogen content is less than 30mg / kg, relay I is turned on and the nitrogen fertilizer pump is started. When the soil nitrogen content is greater than 60mg / kg, relay I is turned off and the nitrogen fertilizer pump stops running.

[0068] When the soil phosphorus content is less than 10mg / kg, relay II is turned on and the phosphate fertilizer pump is started. When the soil phosphorus content is greater than 20mg / kg, relay II is turned off and the phosphate fertilizer pump stops running.

[0069] When the soil potassium content is less than 60mg / kg, relay III is turned on and the potassium fertilizer pump is started. When the soil potassium content is greater than 100mg / kg, relay III is turned off and the potassium fertilizer pump stops running.

[0070] When the soil moisture content is less than 50%, f ,θ f The field water holding rate is θ, open relay IV, start the water pump, when the soil moisture content is greater than 90% f When , relay IV is closed and the water pump stops running.

[0071] The nutrients and water required by crops are monitored in real time through the soil nutrient and moisture monitoring probe 19, and the monitoring data is transmitted to the Raspberry Pi unit 6. The central control module transmits the data of the Raspberry Pi unit 6 to the data storage unit for storage, analysis and comparison, and controls the start and stop of various pumps.

[0072] S3: Real-time monitoring and prevention: monitor soil nutrient information through the soil nutrient and moisture monitoring probe of the soil nutrient and moisture monitoring device; take pictures of the monitored area through the camera of the pest and disease monitoring device, and compare the obtained pictures with the pest and disease comparison database in step S1. If the camera detects that the leaf color is abnormal, and the nutrient and moisture contents in the monitored soil are normal at this time, it is judged that pests and diseases have occurred. According to the location of the pest and disease identified by the camera, the drug spraying device is started to spray the drug; if the camera detects that the leaf color is abnormal, and the nutrient and moisture contents in the monitored soil are less than the normal amount, the fertilizer pump or water pump is started to spray fertilizer or water containing the missing nutrients.

[0073] During spraying, the horizontal position of the nozzle 9 is controlled by mutually perpendicular synchronous belt slide bars I 51 and II 52, and the vertical position of the nozzle 9 is controlled by the telescopic rod 8. The distance information between the crop and the nozzle 9 is obtained by the GY-530VL53L0X laser ranging sensor module on the camera. The central control module controls the telescopic rod 8 on which the nozzle 9 is installed to move horizontally along the synchronous belt slide bar II 52, or along the synchronous belt slide bar I 51 along the synchronous belt slide bar II 52, so that the nozzle 9 is directly above the blades. The central control module controls the extension and retraction length of the telescopic rod 8 to ensure that the nozzle 9 always maintains a distance of 5 cm from the blades.

[0074] Specifically, when the distance between the crop leaves and the nozzle 9 is less than 5 cm, the telescopic rod 8 is turned on and raised until the distance between the crop leaves and the nozzle 9 reaches 5 cm; when the distance between the crop leaves and the nozzle 9 is greater than 5 cm, the telescopic rod 8 is turned on and lowered until the distance between the crop leaves and the nozzle reaches 5 cm for spraying.

[0075] The relay group is connected to the Raspberry Pi unit 6 through the GPIO interface. The telescopic rod 8 is equipped with a GT (gate turn-off thyristor) DC chopper step-down component. The telescopic module needs to frequently switch the current direction. To protect the telescopic rod, this component can provide power and stabilize the voltage for the Raspberry Pi unit 6, ensuring the stability of the device's operation and extending its service life.

[0076] The central control module analyzes and compares the database based on the types and severity of pests and diseases on crop leaves to select the appropriate type and dosage of pesticides.

[0077] The intelligent decision-making system employed in this invention is based on a convolutional neural network (CNN) model. This model achieves over 85% accuracy in monitoring pest and disease leaf patterns, enabling precise drug selection and timely identification and treatment of pest outbreaks. IoT technology connects sensors, controllers, and mobile terminals to enable remote monitoring and management. A cloud platform provides data storage, analysis, and visualization capabilities.

[0078] like Figure 5-10As shown, the spraying device for intelligent agricultural pest and disease monitoring based on YOLO v10 described in the present invention includes a support frame, a pest and disease monitoring device, a soil nutrient and moisture monitoring device, a drug spraying device, a water spraying device, a fertilizer spraying device and a control device. The support frame is a rectangular frame structure, with crossbeams connected at both ends between the four columns 1, and synchronous belt slide rods Ⅰ51 are symmetrically arranged between the two columns on the two opposite sides of the rectangular frame. A baffle 2 is arranged outside the lower side of one side, and a pump group 3 is arranged thereon, including a fertilizer pump, a water pump, and a drug pump. A synchronous belt slide rod Ⅱ52 is vertically slidably connected to the two synchronous belt slide rods Ⅰ51, and a support beam parallel to the synchronous belt slide rod Ⅱ52 is connected to the two opposite cross beams at the top. The control device Raspberry Pi unit 6 and the pest and disease control device are arranged on the support beam. The pest monitoring device camera 7 and the soil nutrient moisture monitoring probe 19 are placed on one side of the bottom crossbeam. The fertilizer, water, and pesticide spraying devices are all connected to the same nozzle 9, which is connected to the synchronous belt slider II 52 via a telescopic rod 8. The pest monitoring device and the soil nutrient moisture monitoring device are each connected to the input of the control device, namely the Raspberry Pi unit 6. The pesticide spraying device, water spraying device, fertilizer spraying device 9, telescopic rod 8, and synchronous belt sliders I 51 and II 52 are respectively connected to the output of the control device. The pest monitoring device and soil nutrient moisture monitoring device monitor crop conditions, and the control device identifies and controls the areas and amounts of fertilizer, water, or pesticide application. A motor 11 is connected to and drives the synchronous belt sliders I 51 and II 52.

[0079] In this embodiment, the power supply 10 provides power for the pump group 3, the relay group 4 (including relays I to VI), the synchronous belt slide bar I 51, the synchronous belt slide bar II 52, the Raspberry Pi unit 6, the camera 7, the telescopic rod 8, the nozzle 9, the motor 11, the soil nutrient and moisture monitoring probe 19, etc. For example, it can drive the spraying device to move on the slide rail to achieve all-round coverage.

[0080] The control device includes a Raspberry Pi unit 6, a central control module, and a data storage unit. The central control module is connected to the Raspberry Pi unit and the data storage unit for communication. The input end of the Raspberry Pi unit is connected to the camera 7 of the pest and disease monitoring device and the soil nutrient and moisture monitoring probe 19 of the soil nutrient and moisture monitoring device; the output end of the central control module is also connected to the relay V of the drug spraying device, the relay IV of the water spraying device, the relays I to III of the fertilizer spraying device, the synchronous belt slide bars I51, II52, the telescopic rod 8 and the nozzle 9 to control the start and stop of each component.

[0081] The pest and disease monitoring device includes a camera 7 and a Raspberry Pi unit 6 of the control device. The camera 7 is used to collect image information of the monitoring area and connect and transmit it to the Raspberry Pi unit 6 of the control device via a data cable. The Raspberry Pi unit 6 receives the leaf color image data taken by the camera, transmits the data to the data storage unit for storage, and performs rapid processing and analysis to identify the characteristics of pests and diseases, determine the type of pests and diseases, and then transmit the judgment results to the central control module, which uses the relevant database therein for analysis and comparison to obtain a solution.

[0082] The soil nutrient moisture monitoring device includes a soil nutrient moisture monitoring probe 19 and the Raspberry Pi unit 6 of the control device. The soil nutrient moisture monitoring probe 19 is used to collect the current nitrogen, phosphorus, potassium content and soil moisture content information of the soil, and connect and transmit it to the Raspberry Pi unit 6 of the control device via a data cable; when the measured value is lower than the set threshold, the central control module of the control device will automatically control the fertilizer spraying device and the water spraying device to replenish the required fertilizer and water.

[0083] The fertilizer spraying device includes a fertilizer pump (nitrogen pump, phosphorus pump, potassium pump), a water pump, and relays I to III. The water spraying device includes a water pump, relay IV, and a central control module of the control device. The nitrogen pump, phosphorus pump, potassium pump, and water pump are respectively connected to the central control module through relays I to IV, and the central control module controls the switches of each relay, thereby controlling the operation of various pumps.

[0084] In this embodiment, the relays I to V4 are used to control the on and off of the circuit to achieve precise control of the equipment. They can also withstand a certain degree of voltage fluctuation and current impact to ensure stable operation of the equipment.

[0085] The drug spraying device includes a drug pump and a relay V. The drug pump provides power for the drug solution. When the pest monitoring device detects the occurrence of pests, the central control module controls the relay V to open and the drug solution is delivered through the drug pump.

[0086] The soil nutrient and moisture monitoring probe 19 collects data at an interval of 15 minutes. It transmits information to the Raspberry Pi unit 6 simultaneously with the camera, and stores the data in the data storage unit. The information transmitted by the soil nutrient and moisture monitoring probe 19 takes precedence over the information transmitted by the camera. The soil is replenished with nutrients and moisture first, and then pesticides are sprayed on the area where pests and diseases occur.

[0087] In this embodiment, when the central control module detects that low leafy green crops such as melons and strawberries need to be treated with drugs, it adjusts the flow and pressure of the drug pump according to the severity of the pests and diseases, extracts the prepared drug from the storage container and pressurizes it, and transports it to the nozzle 9 through a pipeline for spraying, thereby achieving precise spraying.

[0088] The present invention uses a convolutional neural network (CNN) model for training, and an intelligent decision-making system retrieves data from a database to select appropriate pesticide types and dosages based on the crop type and the characteristics and degree of leaf pests and diseases, thereby generating a personalized pesticide spraying plan.

[0089] The soil nutrient and moisture monitoring probe 19 of the present invention can monitor the various nutrient and moisture contents in the soil in real time, and the monitored data is transmitted to the central control module. The central control module transmits the data from the Raspberry Pi unit 6 to the data storage unit for storage. The central control module will periodically obtain data from the soil nutrient and moisture monitoring probe 19 and compare it with the database to determine whether the soil is truly lacking the nutrients and moisture. If the nutrient and moisture content in the soil is greater than or equal to the upper limit, the fertilizer pump or water pump will not start, and the warning light connected to the central control module via relay VI will light up. If the nutrient and moisture content in the soil is less than the upper limit, the fertilizer pump or water pump will start to spray fertilizer and water into the soil, preventing human error from affecting the monitoring.

[0090] The control device used in the present invention is constructed based on the convolutional neural network (CNN), which can accurately realize the selection of applied drugs, thereby enabling timely identification and treatment of the occurrence of pests and diseases.

[0091] The components not described in detail in this application are all existing conventional technologies and will not be described in detail here.

[0092] It can be understood that the above specific description of the present invention is only used to illustrate the present invention and is not limited to the technical solutions described in the embodiments of the present invention. Those skilled in the art should understand that the present invention can still be modified or replaced by equivalents to achieve the same technical effects; as long as the use requirements are met, they are within the scope of protection of the present invention.

Claims

1. A smart agricultural pest monitoring and spraying method based on YOLO v10, characterized by: Proceed as follows: S1: Build a convolutional neural network model and establish a pest and disease comparison database; S2: Set nutrient supplement standards, upper and lower limits for soil moisture, and upper and lower limits for fertilizer concentration; S3: Real-time monitoring and prevention: soil nutrient and moisture information is monitored through the soil nutrient and moisture probes of the soil nutrient and moisture monitoring device; the camera of the pest and disease monitoring device takes a picture of the monitored area, and the obtained picture is compared with the pest and disease comparison database in step S1. If the camera detects abnormal leaf color, and the nutrient and moisture content in the monitored soil are both normal, it is determined that a pest and disease has occurred. According to the location of the pest and disease identified by the camera, the drug spraying device is activated to spray the pesticide; If the camera detects abnormal leaf color, and the nutrient and water content in the soil is less than normal, the fertilizer pump or water pump will be started to spray fertilizer or water containing the missing nutrients.

2. The intelligent agricultural pest monitoring and spraying method based on YOLO v10 according to claim 1 is characterized in that: When spraying fertilizer, water and pesticides, the horizontal position of the nozzle is controlled by two mutually perpendicular synchronous belt slides, and the vertical position of the nozzle is controlled by a telescopic rod. The distance information between the crop and the nozzle is obtained through the laser ranging sensor module. The central control module controls the telescopic rod on which the nozzle is installed to move horizontally along the synchronous belt slide II, or along the synchronous belt slide I along the synchronous belt slide II, so that the nozzle is directly above the blade. The central control module controls the extension and retraction length of the telescopic rod to ensure that the nozzle and the blade always maintain a distance of 5 cm.

3. The intelligent agricultural pest monitoring and spraying method based on YOLO v10 according to claim 1 is characterized in that: The upper and lower limits of soil moisture and fertilizer concentration in step S2 are set as follows: (1) Upper and lower limits of soil moisture The lower limit of soil moisture is set between 50% and 60% of field water holding capacity, and the upper limit of soil moisture is set between 80% and 90% of field water holding capacity; (2) Upper and lower limits of fertilizer concentration In the early stages of crop growth, the lower limit of nitrogen fertilizer concentration is set at 100-200 mg / L, and the lower limit of phosphorus fertilizer and potassium fertilizer concentration is set at 30-40 mg / L; in the vigorous growth period, the lower limit of nitrogen fertilizer concentration is set at 300-400 mg / L, and the lower limit of phosphorus fertilizer and potassium fertilizer concentration is set at 60-80 mg / L; The upper limit of nitrogen fertilizer concentration is set at 500-600 mg / L, and the upper limit of phosphorus fertilizer and potassium fertilizer concentration is set at 100-120 mg / L; (3) Nutrient supplementation standards are as follows: When the soil nitrogen content is less than 30mg / kg, the nitrogen fertilizer pump is turned on; when the soil nitrogen content is greater than 60mg / kg, the nitrogen fertilizer pump is turned off; When the soil phosphorus content is less than 10mg / kg, the phosphate fertilizer pump is turned on; when the soil phosphorus content is greater than 20mg / kg, the phosphate fertilizer pump is turned off; When the soil potassium content is less than 60mg / kg, the potassium fertilizer pump is turned on; when the soil potassium content is greater than 100mg / kg, the potassium fertilizer pump is turned off; When the soil moisture content is less than 50%, f ,θ f is the field water holding capacity, turn on the water pump, when the soil moisture content is greater than 90%θ f Turn off the water pump.

4. The intelligent agricultural pest monitoring and spraying method based on YOLO v10 according to claim 1, characterized in that: In step S1, a convolutional neural network model is constructed, specifically: First, at least 3,000 photos of crops in normal and various pathological states were taken from multiple angles within the monitoring area. These photos were then fed into YOLO v10 for learning. When the red value (R) of the leaves was below 80-100, the green value (G) was below 120-140, and the blue value (B) was below 80-100, the leaves were identified as having pests and diseases, and the images were labeled accordingly, completing the construction of the convolutional neural network model. Then, the pest and disease leaves learned by YOLO v10 are labeled, and the pest and disease comparison database obtained after learning is transferred to the Raspberry Pi unit; The Raspberry Pi's built-in YOLO v10-based image recognition module analyzes images captured by the camera using its algorithm to identify the type, severity, and location of pests and diseases. This information is then transmitted to the central control module, where the corresponding database selects the appropriate pesticide type and dosage based on the type and severity of pests and diseases on the crop leaves.

5. The spraying device of the intelligent agricultural pest monitoring and spraying method based on YOLO v10 according to any one of claims 1 to 4, characterized in that: The invention comprises a support frame, a pest and disease monitoring device, a soil nutrient and moisture monitoring device, a fertilizer spraying device, a water spraying device, a drug spraying device and a control device. The support frame is a rectangular frame, and a synchronous belt slide bar I is symmetrically arranged between two upright posts on two opposite sides of the frame. A baffle is arranged outside the lower side of one side, on which a drug pump, a water pump and a fertilizer pump are arranged. A synchronous belt slide bar II is vertically connected to the two synchronous belt slide bars I and the two opposite cross beams at the top are connected to a support beam parallel to the synchronous belt slide bar II. The support beam is provided with a control device and a camera of the pest and disease monitoring device, and the soil nutrient and moisture monitoring device. The soil nutrient and moisture monitoring probe is placed on the bottom crossbeam on one side. The drug spraying device, water spraying device, and fertilizer spraying device are connected to different nozzles respectively. The nozzles are connected to the synchronous belt slide bar II through a telescopic rod. The pest and disease monitoring device and the soil nutrient and moisture monitoring device are connected to the input end of the control device respectively. The drug spraying device, water spraying device, fertilizer spraying device, telescopic rod and synchronous belt slide bars I and II are connected to the output end of the control device respectively. The crop status is monitored by the disease and pest monitoring device and the soil nutrient and moisture monitoring device, and the fertilization, water spraying or pesticide application area and amount are identified and controlled through the database in the control device.

6. The YOLO v10-based intelligent agricultural pest monitoring spraying device according to claim 5, characterized in that: The control device includes a Raspberry Pi unit, a central control module, and a data storage unit. The central control module is connected to the Raspberry Pi unit and the data storage unit for communication. The input end of the Raspberry Pi unit is connected to the camera of the pest and disease monitoring device and the soil nutrient and moisture monitoring probe of the soil nutrient and moisture monitoring device; the output end of the central control module is also connected to the pesticide pump relay V of the drug spraying device, the water pump relay IV of the water spraying device, the relays I to III of the fertilizer spraying device, the synchronous belt slide bars I and II, and the telescopic rod to control the start and stop of each component.

7. The YOLO v10-based intelligent agricultural pest monitoring spraying device according to claim 5, characterized in that: The pest and disease monitoring device includes a camera and a Raspberry Pi unit of the control device. The camera is used to collect image information of the monitoring area and connect and transmit it to the Raspberry Pi unit of the control device via a data cable. The pest and disease monitoring device is identified and monitored through the built-in YOLO v10-based image recognition module of the control device.

8. The YOLO v10-based intelligent agricultural pest monitoring spraying device according to claim 5, characterized in that: The soil nutrient moisture monitoring device includes a soil nutrient moisture monitoring probe and a Raspberry Pi unit of the control device. The soil nutrient moisture monitoring probe is used to collect the current nitrogen, phosphorus, potassium content and soil moisture content information of the soil, and is connected and transmitted to the Raspberry Pi unit of the control device via a data cable for monitoring by the control device. When the measured value is lower than the set threshold, the central control module of the control device will automatically control the operation of the fertilizer spraying device and the water spraying device to replenish the required fertilizer and water.

9. The YOLO v10-based intelligent agricultural pest monitoring spraying device according to claim 5, characterized in that: The fertilizer spraying device includes a nitrogen pump, a phosphorus pump, a potassium pump, and relays I to III; the water spraying device includes a water pump and relay IV; the water pump is used to provide liquid delivery power; the nitrogen pump, phosphorus pump, and potassium pump are respectively connected to the central control module through relays I to IV, and the central control module controls the switches of each relay, thereby controlling the operation of various pumps.

10. The YOLO v10-based intelligent agricultural pest monitoring spraying device according to claim 5, characterized in that: The drug spraying device includes a pesticide pump and a relay V. The pesticide pump and the relay V are connected to the central control module. The pesticide pump provides power for the liquid medicine. When the pest monitoring device detects the occurrence of pests and diseases, the central control module controls the relay V to open and the liquid medicine is sprayed through the pesticide pump.

Images