Automatic online precision pesticide mixing system, method and device

Abstract

The invention provides an automatic online precision pesticide mixing system, method and device, and relates to the technical field of pesticide treatment, the system comprises an image recognition analysis module, a pesticide mixing module and a control module which are connected in sequence; and the image recognition and analysis module is used for acquiring an original image data set, obtaining target detection data according to the constructed crop disaster recognition model, encoding the detected and analyzed data, and converting the data into a format required by a two-dimensional code. The medicine mixing module generates a prescription two-dimensional code according to the target detection data and obtains medicine mixing instruction data according to the prescription two-dimensional code, and the control module is used for controlling the concentration of the liquid medicine according to the medicine mixing instruction data. The problems that in the prior art, the pesticide application matching precision is low, the online pesticide mixing control flow precision is low, and the pesticide mixing amount is not uniform enough are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of pesticide treatment, and in particular to an automated online precision pesticide mixing system, method and device. Background Art

[0002] my country has a large population, and its land area and pesticide usage are among the highest in the world. However, the comprehensive utilization rate of pesticides is far lower than that of developed countries. The vast majority of plant protection machinery in my country uses premixed pesticides, which can easily lead to air or skin contact for operators and endanger human health. The application of pesticides based on traditional experience can easily lead to insufficient or excessive application of pesticides. Excessive pesticides reserved in the device will evaporate and remain, which will also cause environmental pollution and corrode the reserved pesticide device.

[0003] Many existing domestic online drug mixing technology control systems have a long detection and identification processing time and cannot quickly identify all problems, which are prone to problems such as low accuracy of drug application ratio, low accuracy of online drug mixing control flow, and uneven mixing amount. Summary of the invention

[0004] In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide an automated online precision drug mixing system, method and device, which solves the problems of low accuracy in drug application ratio, low accuracy in online drug mixing flow control and uneven mixing amount in the prior art.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] An automated online precision drug mixing system, comprising:

[0007] An image recognition and analysis module, a drug mixing module and a control module connected in sequence;

[0008] The image recognition and analysis module is used to obtain the original image data set, and obtain target detection data based on the constructed crop disaster recognition model. The medicine mixing module generates a prescription QR code based on the target detection data, and obtains medicine mixing instruction data based on the prescription QR code. The control module is used to control the concentration of the medicine solution according to the medicine mixing instruction data.

[0009] Preferably, the image recognition and analysis module includes:

[0010] An image collection submodule, a feature extraction submodule and a recognition submodule connected in sequence;

[0011] The image collection submodule is used to prepare a sample set based on images of field crops and common pests to obtain original image data; the feature extraction submodule is used to perform image preprocessing on the original image data and perform feature extraction on the preprocessed original image data to obtain feature image data; the recognition submodule is used to obtain target detection data based on the feature image data and a convolutional neural network algorithm.

[0012] An automated online precision drug mixing method, comprising:

[0013] Obtain the original image data set, and obtain the target detection data based on the constructed crop disaster recognition model;

[0014] generating a prescription QR code according to the target detection data;

[0015] Obtaining drug mixing instruction data according to the prescription QR code;

[0016] The concentration of the drug solution is controlled according to the drug mixing instruction data.

[0017] An automated online precision medicine mixing device, comprising:

[0018] Water pump, turbine flowmeter, check valve, medicine box, water tank, medicine mixing cabin, nozzle, medicine liquid pump, pharmaceutical concentration flow controller, infusion tube, metering pump, online medicine mixing intelligent control processor, flow control valve;

[0019] The water pump is used to pump water from the water tank; the turbine flowmeter is used to measure the real-time flow of water and pesticides respectively; the check valve is used to prevent liquid backflow; the medicine box is used to store liquid medicine; the water tank is used to store water; the medicine mixing cabin is used to mix water and liquid medicine, and preliminarily mix water and liquid medicine; the liquid medicine pump is used to evenly mix the pesticide and water under the stirring of the high-speed impeller of the liquid medicine pump; the pharmaceutical concentration flow controller is used to detect the concentration of the mixed liquid medicine; the metering pump is used to extract liquid medicine from the medicine box; the online medicine mixing intelligent control processor is used to mainly complete the download and analysis of the two-dimensional code medicine mixing prescription map, vehicle speed and pressure collection, flow calculation and collection, flow control valve and metering pump drive, LCD display and control algorithm implementation. The control module controls the concentration of the liquid medicine according to the medicine mixing instruction data. Store the medicine mixing ratio relationship model and manage the medicine mixing prescription data, control and optimize according to these data to ensure that the device can accurately mix the required concentration of liquid medicine; the flow control valve is used to control the flow size of pesticides and water.

[0020] The water in the water tank is pumped out by a pump, and after the flow rate is measured by a turbine flowmeter, it enters the mixing chamber through a check valve. The liquid medicine in the medicine box is pumped out by a metering pump, and after the flow rate is measured by a turbine flowmeter, it enters the mixing chamber through a flow control valve. The online mixing intelligent control processor is based on the STM32F4 chip, including a minimum system unit, a power conversion unit, a signal acquisition unit, an execution drive unit, and a communication unit. The controller mainly completes the download and analysis of the QR code mixing prescription map, the speed and pressure acquisition, the calculation and acquisition of the flow rate, the flow control valve and the metering pump drive, and the implementation of the LCD display and control algorithm. The control module is used to control the concentration of the liquid medicine according to the mixing instruction data. The mixing ratio relationship model is stored and the mixing prescription data is managed. Control and optimization are performed based on these data to ensure that the device can accurately mix the liquid medicine of the required concentration. In the mixing chamber, water and liquid medicine will be preliminarily mixed according to the mixing prescription. The mixed liquid medicine is stirred by the high-speed impeller of the liquid medicine pump to evenly mix the pesticide and water. After being extracted by the drug liquid pump, it enters the drug liquid concentration detection device through the infusion tube to detect the concentration, and then is sprayed out through the nozzle.

[0021] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0022] The present invention provides an automated online precision drug mixing system, method and device, the system comprising: an image recognition and analysis module, a drug mixing module and a control module connected in sequence; the image recognition and analysis module is used to obtain the original image data set, and obtain target detection data according to the constructed crop disaster recognition model, the drug mixing module generates a prescription QR code according to the target detection data, and obtains drug mixing instruction data according to the prescription QR code, and the control module is used to control the concentration of the drug solution according to the drug mixing instruction data. The present invention can simultaneously use image recognition technology to prevent seedling crushing, realize the digital composite operation of intelligent plant protection machinery and tools, and improve the utilization rate of machinery and tools. In the process of plant protection operations, the online dynamic drug mixing is based on the prescription, the QR code information is quickly generated through the unit prescription, and the drug mixing mode is started by scanning the code, so as to realize the full digital composite operation of uniform and precise drug mixing. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0024] Figure 1 A schematic diagram of an automated online precision drug mixing system provided by an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of an automated online precision medicine mixing device provided in an embodiment of the present invention.

[0026] Description of reference numerals:

[0027] 1. Water pump; 2. Turbine flowmeter; 3. Check valve; 4. Medicine box; 5. Water tank;

[0028] 6. Drug mixing cabin; 7. Nozzle; 8. Drug liquid pump; 9. Drug liquid concentration flow controller; 10. Infusion tube; 11. Metering pump; 12. Online drug mixing intelligent control processor; 13. Flow control valve. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] The purpose of the present invention is to provide an automated online precision drug mixing system, method and device, which solves the problems of low drug application ratio accuracy, low online drug mixing flow control accuracy and uneven drug mixing amount in the prior art.

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] like Figure 1 As shown, the present invention provides an automated online precision drug mixing system, comprising:

[0033] An image recognition and analysis module, a drug mixing module and a control module connected in sequence;

[0034] The image recognition and analysis module is used to obtain the original image data set, and obtain the target detection data based on the constructed crop disaster recognition model, encode the data after detection and analysis, and convert it into the format required for the two-dimensional code. Based on the encoded data, a matrix composed of black and white squares is generated. The size of this matrix is ​​determined by the required information capacity and the size of the two-dimensional code. Locators and correction symbols are added to the matrix so that the scanning device can accurately identify and decode the two-dimensional code. Therefore, the drug mixing module generates a prescription two-dimensional code based on the target detection data, and obtains drug mixing instruction data based on the prescription two-dimensional code. The control module is used to control the concentration of the drug solution according to the drug mixing instruction data.

[0035] Specifically, based on information fusion technology, intelligent mixing of pesticides is carried out after scanning the QR code, and the whole digital composite operation of fast, accurate, and uniform online dynamic mixing of pesticides is realized according to the unit prescription. Based on the deep learning method, the image recognition in the agricultural field is studied and reasonably designed. The sample set is made based on the images of field crops and common pests. The neural network model is trained by making full use of the advantages of deep learning methods in autonomous mining and learning complex data feature expressions. The convolutional neural network image recognition and analysis model with high performance of field crops and common pests is constructed and optimized in a targeted manner to achieve end-to-end rapid detection and identification of targets. In the process of autonomous tracking plant protection operations, the equipment adopts variable precision application of pesticides, and accurately targets crops according to the running speed and plant protection equipment. The flow is controlled by the flow controller in the mixing cabin: the water flow, drug flow, and mixed liquid concentration are monitored by the flow controller, and the pesticide liquid flow is controlled in real time according to the unprocessed information instructions after scanning the QR code, and the concentration of the overall liquid is controlled by the size of the pesticide flow.

[0036] Furthermore, the image recognition and analysis module includes:

[0037] An image collection submodule, a feature extraction submodule and a recognition submodule connected in sequence;

[0038] The image collection submodule is used to prepare a sample set based on images of field crops and common pests to obtain original image data. The feature extraction submodule is used to perform image preprocessing on the original image data, and to specifically construct and optimize a convolutional neural network image recognition and analysis model with high performance for field crops and common pests to achieve end-to-end rapid target detection and recognition, and to perform feature extraction on the original image data to obtain feature image data. The recognition submodule is used to obtain target detection data based on the feature image data and the convolutional neural network algorithm.

[0039] Specifically, we fully utilize the advantages of deep learning methods in autonomously mining and learning complex data feature expressions to train neural network models, and perform image preprocessing and feature extraction on the collected images. We also use deep learning methods to identify images in the agricultural field, and use images of field crops and common pests to make sample sets. We fully utilize the advantages of deep learning methods in autonomously mining and learning complex data feature expressions to train neural network models, and specifically build and optimize high-performance convolutional neural network image recognition and analysis models for field crops and common pests to achieve end-to-end rapid target detection and recognition.

[0040] Specifically, the real-time dynamic positioning technology based on carrier phase observations can provide real-time three-dimensional positioning results of the station in the specified coordinate system with centimeter-level accuracy. In the RTK operation mode, the base station transmits its observation values ​​and station coordinate information to the mobile station through the data link. The mobile station not only receives data from the base station through the data link, but also collects GPS observation data, and forms differential observations in the system for real-time processing, and gives centimeter-level positioning results at the same time, which takes less than one second.

[0041] Specifically, a modular design is adopted, and an image recognition and analysis module, a drug mixing module, a transmission module, a control module, and an execution module are set up respectively. A convolutional neural network image recognition and analysis model with high performance for field crops and common pests is constructed and optimized in a targeted manner to achieve end-to-end rapid target detection and recognition. By transmitting the recognized image data, the drug is dynamically mixed online according to the data unit prescription, and the target detection data is obtained according to the constructed crop disaster identification model. The data after detection and analysis is encoded and converted into the format required for the QR code. According to the encoded data, a matrix composed of black and white squares is generated. The size of this matrix is ​​determined by the required information capacity and the size of the QR code. Locators and correction symbols are added to the matrix so that the scanning device can accurately identify and decode the QR code. The drug mixing module generates a prescription QR code based on the target detection data, and obtains drug mixing instruction data based on the prescription QR code. The execution module is used to control the concentration of the drug solution according to the drug mixing instruction data.

[0042] Specifically, we use python to build a neural network model, and specifically build and optimize a high-performance convolutional neural network image recognition and analysis model for field crops and common pests to achieve end-to-end rapid target detection and identification. We use the elimination method to distinguish between grass and insect conditions. We digitize information transmission and perform streaming operations to reduce the processing time of the control system. We optimize the execution equipment structure to realize the streaming plant protection machinery operation process, so that the machinery can travel, mix pesticides, analyze targeted targets, and simultaneously perform intelligent operations to complete the operation requirements.

[0043] The inertial navigation module, visual yaw module, complementary filter, heading data fusion meter, execution trajectory operation and vehicle speed sensor modules are connected in sequence;

[0044] The above module intends to adopt a navigation method that combines visual yaw monitoring and inertial navigation. Self-propelled unmanned tracking precision operation control includes speed control and direction control. Speed ​​control mainly realizes precise control of the equipment's travel speed by sending precise synchronous pulse signals to the servo motor. Each motor is equipped with an independent tachometer to provide real-time feedback on the motor's accurate speed. The linear speed of the equipment is calculated in real time through the reduction ratio, ensuring the accuracy of the unmanned tracking operation's travel speed.

[0045] Specifically, direction control is the core of unmanned tracking operation. Video yaw monitoring and inertial navigation modules composed of GPS, electronic compass, accelerometer and gyroscope are used, which are divided into high-frequency devices and low-frequency devices in the frequency domain. The fast performance of gyroscope and accelerometer is used to make up for the insufficient dynamic response of low-frequency devices such as electronic compass. Through complementary filtering method and image recognition technology, the deviation between the actual and the target caused by multiple factors such as uneven ground and motor control error is monitored, and the accurate deviation is input to the control system for correction, so as to achieve precise adjustment of the direction of unmanned tracking operation.

[0046] This embodiment also provides an automated online precision drug mixing method, comprising:

[0047] Step 100: Obtain the original image data set, and obtain target detection data based on the constructed crop disaster recognition model;

[0048] Step 200: Generate a prescription QR code according to the target detection data;

[0049] Step 300: Obtaining drug mixing instruction data according to the prescription QR code;

[0050] Step 400: Control the concentration of the drug solution according to the drug mixing instruction data.

[0051] like Figure 2 As shown, this embodiment provides an automated online precision medicine mixing device, wherein the water pump 1 is used to pump water from the water tank 5; the turbine flowmeter 2 is used to measure the real-time flow of water and pesticides respectively; the check valve 3 is used to prevent liquid backflow; the medicine box 4 is used to store liquid medicine; the water tank 5 is used to store water; the medicine mixing chamber 6 is used to mix water and liquid medicine, and preliminarily mix water and liquid medicine; the liquid medicine pump 8 is used to evenly mix the pesticide and water under the stirring of the high-speed impeller of the liquid medicine pump 8; the liquid medicine concentration flow controller 9 is used to detect the concentration of the mixed liquid medicine; the metering pump 11 is used to extract liquid medicine from the medicine box 4; the online medicine mixing intelligent control processor 12 is used to mainly complete the download and analysis of the two-dimensional code medicine mixing prescription map, the speed and pressure collection, the calculation and collection of the flow, the flow control valve 13 and the metering pump 11 drive, the LCD display and the implementation of the control algorithm. The control module controls the concentration of the liquid medicine according to the medicine mixing instruction data. The mixing ratio relationship model is stored and the mixing prescription data is managed. Control and optimization are performed based on these data to ensure that the device can accurately mix the required concentration of liquid medicine; the flow control valve 13 is used to control the flow of pesticides and water.

[0052] The water in the water tank 5 is pumped out by the pump 1, and after the flow rate is measured by the turbine flowmeter 2, it enters the medicine mixing cabin 6 through the check valve. The liquid medicine in the medicine box 4 is pumped out by the metering pump 11, and after the flow rate is measured by the turbine flowmeter 2, it enters the medicine mixing cabin 6 through the flow control valve 13. The online medicine mixing intelligent control processor 12 is based on the STM32F4 chip, including a minimum system unit, a power conversion unit, a signal acquisition unit, an execution drive unit and a communication unit. The controller mainly completes the download and analysis of the two-dimensional code medicine mixing prescription map, the speed and pressure collection, the calculation and collection of the flow rate, the flow control valve 13 and the metering pump 11 drive, and the LCD display and control algorithm implementation. The control module is used to control the concentration of the liquid medicine according to the medicine mixing instruction data. The medicine mixing ratio relationship model is stored and the medicine mixing prescription data is managed. Control and optimization are performed based on these data to ensure that the device can accurately mix the medicine liquid of the required concentration. In the medicine mixing cabin 6, water and liquid medicine will be preliminarily mixed according to the medicine mixing prescription. The mixed liquid medicine is stirred by the high-speed impeller of the liquid medicine pump 8 to uniformly mix the pesticide and water. After being extracted by the liquid medicine pump 8, it enters the liquid medicine concentration detection device through the infusion tube 10 to detect the concentration, and then is sprayed out through the nozzle 7.

[0053] The beneficial effects of the present invention are as follows:

[0054] (1) The utilization rate of pesticides and the accuracy of the pesticide application ratio of online mixing technology have been improved. The range of the mixing ratio that can be adjusted has been expanded. The use of QR code information technology and information transmission technology based on the Internet of Things can identify crops through images, infrared thermal imaging can assist laser insect identification, and infrared remote sensing with infrared thermal imaging technology as the core can provide reliable and timely information on the growth of large-scale crops, the presence of pests and diseases, and yield estimation. Various crops have different absorption, reflection and radiation spectral characteristics. This spectral feature is not only reflected differently by various crops in the same spectral region, but also by the reflection of the same crop in different spectral regions. By combining the Internet of Things with field remote sensing observations, using crop-related image information and spectral analysis technology, the wireless sensor network soil moisture measurement and irrigation research can be carried out to extract crop growth information and estimate crop water and fertilizer requirements in real time.

[0055] (2) Variable adjustable speed control to achieve intelligent targeted precision plant protection. End-to-end rapid target detection and identification is achieved. During the autonomous tracking plant protection operation, the equipment uses variable precision spraying to accurately target crops based on the operating speed and plant protection equipment.

[0056] (2) The speed and accuracy of automated pesticide mixing have been improved. Python has been used to establish a neural network model and the elimination method has been adopted to distinguish between weed and insect conditions. Information transmission has been digitized and stream operations have been carried out to reduce the processing time of the control system.

[0057] (3) The control flow accuracy and uniformity of online drug mixing are improved, and the application (drug) concentration and travel speed are accurately controlled according to the prescription, so as to achieve precise control of the operation. A modular design is adopted, and an image recognition and analysis module, a drug mixing module, a transmission module, a control module, and an execution module are set up respectively. Intelligent control plant protection equipment that can realize automatic metering, automatic drug mixing, and automatic drug application functions can quickly, accurately, and evenly mix drugs, thereby smoothly realizing the automatic metering, transmission, and fertilization of water and fertilizer.

[0058] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. As for the methods and devices disclosed in the embodiments, since they correspond to the systems disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method part.

[0059] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. An automated online precision drug mixing system, characterized in that: include: An image recognition and analysis module, a drug mixing module and a control module connected in sequence; The image recognition and analysis module is used to acquire the original image data set, and obtain target detection data based on the constructed crop disaster recognition model, encode the data after detection and analysis, and convert it into the format required by the QR code. The medicine mixing module generates a prescription QR code based on the target detection data, and obtains medicine mixing instruction data based on the prescription QR code. The control module is used to control the concentration of the medicine solution according to the medicine mixing instruction data.

2. The automated online precision drug mixing system according to claim 1, characterized in that: The image recognition and analysis module comprises: An image collection submodule, a feature extraction submodule and a recognition submodule connected in sequence; The image collection submodule is used to prepare a sample set based on images of field crops and common pests to obtain original image data; the feature extraction submodule is used to perform image preprocessing on the original image data and perform feature extraction on the original image data to obtain feature image data; the recognition submodule is used to obtain target detection data based on the feature image data and a convolutional neural network algorithm.

3. An automated online precision drug mixing method, characterized in that: include: Obtain the original image data set, and obtain the target detection data based on the constructed crop disaster recognition model; generating a prescription QR code according to the target detection data; Obtaining drug mixing instruction data according to the prescription QR code; The concentration of the drug solution is controlled according to the drug mixing instruction data.

4. An automated online precision drug mixing device, characterized in that: include: Water pump, turbine flowmeter, check valve, medicine box, water tank, medicine mixing cabin, nozzle, medicine liquid pump, pharmaceutical concentration flow controller, infusion tube, metering pump, online medicine mixing intelligent control processor, flow control valve; The water pump is used to pump water from the water tank; the turbine flowmeter is used to measure the real-time flow of water and pesticides respectively; the check valve is used to prevent liquid backflow; the medicine box is used to store liquid medicine; the water tank is used to store water; the medicine mixing cabin is used to mix water and liquid medicine, and preliminarily mix water and liquid medicine; the liquid medicine pump is used to evenly mix the pesticide and water under the stirring of the high-speed impeller of the liquid medicine pump; the pharmaceutical concentration flow controller is used to detect the concentration of the mixed liquid medicine; the metering pump is used to extract liquid medicine from the medicine box; The online drug mixing intelligent control processor is mainly used to complete the download and analysis of the QR code drug mixing prescription map, vehicle speed and pressure collection, flow calculation and collection, flow control valve and metering pump drive, LCD display and control algorithm implementation. The control module controls the concentration of the drug solution according to the drug mixing instruction data. Store the drug mixing ratio relationship model and manage the drug mixing prescription data, and control and optimize according to these data to ensure that the device can accurately mix the drug solution of the required concentration; The flow control valve is used to control the flow of pesticides and water. The water in the water tank is pumped out by a pump, and after the flow rate is measured by a turbine flowmeter, it enters the mixing chamber through a check valve. The liquid medicine in the medicine box is pumped out by a metering pump, and after the flow rate is measured by a turbine flowmeter, it enters the mixing chamber through a flow control valve. The online mixing intelligent control processor is based on the STM32F4 chip, including a minimum system unit, a power conversion unit, a signal acquisition unit, an execution drive unit, and a communication unit. The controller mainly completes the download and analysis of the QR code mixing prescription map, the speed and pressure acquisition, the calculation and acquisition of the flow rate, the flow control valve and the metering pump drive, and the implementation of the LCD display and control algorithm. The control module is used to control the concentration of the liquid medicine according to the mixing instruction data. The mixing ratio relationship model is stored and the mixing prescription data is managed. Control and optimization are performed based on these data to ensure that the device can accurately mix the liquid medicine of the required concentration. In the mixing chamber, water and liquid medicine will be preliminarily mixed according to the mixing prescription. The mixed liquid medicine is stirred by the high-speed impeller of the liquid medicine pump to evenly mix the pesticide and water. After being extracted by the drug liquid pump, it enters the drug liquid concentration detection device through the infusion tube to detect the concentration, and then is sprayed out through the nozzle.

Images