Plant growth monitoring system

Abstract

The invention discloses a plant growth monitoring system, and relates to the technical field of plant growth monitoring, and multi-parameter monitoring equipment in the system comprises a main body, a supporting rod, a data processing module, a fixed sensor group and a sensor expansion module interface, atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, illumination intensity, rainfall, wind speed, wind direction, soil temperature, soil humidity, soil conductivity, short-wave radiation flux, long-wave radiation flux, photosynthetically active radiation flux, multiband solar spectrum, plant reflection spectrum, dynamic crop coefficient, canopy temperature, latent heat flux and net radiation flux can be obtained; according to the method, various plant growth monitoring parameters such as the vegetation index, the chlorophyll index and the effective accumulated temperature are normalized, so that more plant growth parameters can be obtained, and more accurate plant growth information is provided.

Description

Technical Field

[0001] The present application relates to the technical field of plant growth monitoring, and in particular to a plant growth monitoring system. Background Art

[0002] With the development of modern agriculture, the demand for precision in agricultural production is increasing. Although existing intelligent agricultural monitoring systems can collect and process a large amount of data, they mainly focus on the collection of basic data such as meteorological parameters and soil parameters. This limitation makes the existing intelligent agricultural monitoring systems have certain inaccuracies in crop growth monitoring and cannot meet the needs of modern agriculture for precision management. Summary of the invention

[0003] The purpose of this application is to provide a plant growth monitoring system that can obtain more plant growth parameters and thus provide more accurate plant growth information.

[0004] To achieve the above objectives, this application provides the following solutions: The present application provides a plant growth monitoring system, comprising: Multi-parameter monitoring equipment is used to obtain plant growth monitoring parameters; The multi-parameter monitoring device comprises: a main body, a support rod, a data processing module, a fixed sensor group and a sensor expansion module interface; The main body is connected to the support rod; The fixed sensor group, the data processing module and the sensor expansion module interface are all installed on the main body; The sensor expansion module interface is used to connect the sensor expansion module, wherein the sensor expansion module is used to connect the external sensor, and the external sensor is a sensor of multiple different types of communication protocols and communication interfaces; The fixed sensor group includes a photosynthetically active radiation sensor, a spectrum sensor, a shortwave radiation sensor, a longwave radiation sensor, a light sensor, a rainfall sensor, a temperature sensor, a humidity sensor, an air pressure sensor and a carbon dioxide sensor; The fixed sensor group and the sensor expansion module are used to collect first plant growth monitoring parameters, which include atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, light intensity, rainfall, wind speed, wind direction, soil temperature, soil moisture, soil conductivity, shortwave radiation flux, longwave radiation flux, photosynthetically active radiation flux, multi-band solar spectrum and plant reflectance spectrum; The data processing module is used to calculate the second plant growth monitoring parameter based on the first plant growth monitoring parameter, wherein the second plant growth monitoring parameter includes a dynamic crop coefficient, a canopy temperature, a latent heat flux, a net radiation flux, a normalized vegetation index, a chlorophyll index and an effective accumulated temperature.

[0005] According to the specific embodiments provided in this application, this application discloses the following technical effects: The present application provides a plant growth monitoring system, in which the multi-parameter monitoring equipment includes: a main body, a support rod, a data processing module, a fixed sensor group and a sensor expansion module interface, which can obtain atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, light intensity, rainfall, wind speed, wind direction, soil temperature, soil moisture, soil conductivity, shortwave radiation flux, longwave radiation flux, photosynthetically active radiation flux, multi-band solar spectrum, plant reflectance spectrum, dynamic crop coefficient, canopy temperature, latent heat flux, net radiation flux, normalized vegetation index, chlorophyll index and effective accumulated temperature and other plant growth monitoring parameters. Therefore, the present application can obtain more plant growth parameters, thereby providing more accurate plant growth information. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] In order to more clearly illustrate the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0007] Figure 1 This is a schematic diagram of the structure of the plant growth monitoring system in the embodiment of the present application; Figure 2 This is a schematic diagram of the structure of a multi-parameter monitoring device in an embodiment of the present application; Figure 3 A top view of a multi-parameter monitoring device in an embodiment of the present application; Figure 4 A bottom view of the multi-parameter monitoring device in an embodiment of the present application; Figure 5 This is a schematic diagram of the structure of the power module in the embodiment of the present application; Figure 6 This is a schematic diagram of the structure of the sensor expansion module in the embodiment of the present application; Figure 7 This is a functional schematic diagram of the intelligent control interface in the embodiment of the present application; Figure 8 This is a flow chart of a regional collaborative plant growth status assessment algorithm in an embodiment of the present application; Fig. 9Schematic diagram of the structure of the CAE-RNN neural network model in the embodiment of the present application.

[0008] Reference numerals: 1-data processing module; 2-fixed sensor group; 3-wireless transmission module; 4-cloud storage server; 5-power module; 6-sensor expansion module; 101-main body; 102-power indicator light; 103-charging status indicator light; 104-network status indicator light; 105-warning indicator light; 106-compass; 107-power button; 108-support rod; 109-support rod base; 110 external power supply interface; 111-sensor expansion module interface; 201-photosynthetically active radiation sensor; 202-downlink spectrum sensor; 203 -Downlink shortwave radiation sensor; 204-Downlink longwave radiation sensor; 205-Light sensor; 206-Rainfall sensor; 207-Temperature, humidity and pressure sensor; 208-Carbon dioxide sensor; 209-Uplink shortwave radiation sensor; 210-Uplink spectrum sensor; 211-Uplink longwave radiation sensor; 301-4G antenna; 302-4G antenna socket; 601-Power interface; 602-UART interface; 603-RS485 interface; 604-Analog signal interface; 605-Serial digital component interface; 606-Host interface. DETAILED DESCRIPTION

[0009] Research has found that there are still many shortcomings in the existing intelligent agricultural monitoring system: First, although the existing intelligent agricultural monitoring system can collect and process a large amount of data, it is limited to basic parameters such as meteorological parameters and soil parameters, that is, the parameter features are relatively few, and it is difficult to directly extract the key growth parameters of crops and valuable information for crop growth. Therefore, the diversity of data collection needs to be improved. In addition, the types of sensors mounted on the existing intelligent agricultural monitoring system are fixed. If the sensor categories are expanded, a new monitoring system needs to be customized, which undoubtedly increases the cost of users.

[0010] Secondly, the existing intelligent agricultural monitoring system has a limited monitoring range, and the monitoring equipment at multiple points is isolated from each other. This will result in the inability to share data between devices, forming information islands, which will lead to a waste of data resources and the inability to exert the synergy between data. In addition, due to the existence of information islands, it will affect the monitoring system's assessment of the health status of crops and the guidance provided to users on crop optimization decisions, thereby affecting the growth of crops.

[0011] In response to the shortcomings in the prior art, the present embodiment provides a plant growth monitoring system, which solves the problems in the existing intelligent agricultural monitoring system, such as few measurement parameter features, limited monitoring range, isolated data, and the monitoring of plant growth status information is highly targeted but less universal, resulting in the need for customization based on demand.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0013] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0014] See also Figure 1 , this embodiment provides a plant growth monitoring system, including: Multi-parameter monitoring equipment, wireless transmission module 3 and cloud storage server 4.

[0015] The multi-parameter monitoring device is connected to the cloud storage server 4 via the wireless transmission module 3 .

[0016] The multi-parameter monitoring device includes a device for obtaining plant growth monitoring parameters.

[0017] The multi-parameter monitoring device comprises: a data processing module 1, a fixed sensor group 2 and a sensor expansion module interface.

[0018] The sensor expansion module interface is used to connect to the sensor expansion module 6, wherein the sensor expansion module 6 is used to connect to an external sensor, and the external sensor includes sensors of various types of communication protocols and communication interfaces.

[0019] The first plant growth monitoring parameters can be collected by fixing the sensor group 2 and the sensor expansion module 6, and the first plant growth monitoring parameters include atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, light intensity, rainfall, wind speed, wind direction, soil temperature, soil moisture, soil conductivity, shortwave radiation flux, longwave radiation flux, photosynthetically active radiation flux, multi-band solar spectrum and plant reflectance spectrum.

[0020] The fixed sensor group 2 includes a photosynthetically active radiation sensor, a spectrum sensor, a shortwave radiation sensor, a longwave radiation sensor, a light sensor, a rainfall sensor, a temperature sensor, a humidity sensor, an air pressure sensor and a carbon dioxide sensor.

[0021] As an example, the sensor expansion module 6 is used to connect an external sensor with an RS485 interface. For example, the sensor expansion module 6 is connected to an external sensor with an RS485 interface such as a wind speed sensor, a wind direction sensor, a soil temperature sensor, a soil moisture sensor, and a soil conductivity sensor.

[0022] The data processing module 1 is used to calculate the second plant growth monitoring parameter based on the first plant growth monitoring parameter, wherein the second plant growth monitoring parameter includes a dynamic crop coefficient, a canopy temperature, a latent heat flux, a net radiation flux, a normalized vegetation index, a chlorophyll index and an effective accumulated temperature.

[0023] The fixed sensor group 2 and the sensor expansion module 6 send the collected plant growth monitoring parameters to the data processing module 1 through RS485, I2C and other interfaces for calculation and analysis, including data preprocessing, edge computing, etc. The data processing module 1 will calibrate and process the 15 plant growth monitoring parameters read, and derive another 7 parameters (i.e., the second plant growth monitoring parameters) in combination with the corresponding algorithms.

[0024] Specifically, the normalized vegetation index is calculated by using the red wave component and the near infrared band component in the multi-band solar spectrum and the red wave component and the near infrared band component in the plant reflectance spectrum.

[0025] The dynamic crop coefficient is calculated from atmospheric temperature, atmospheric humidity, wind speed, soil temperature, soil moisture, rainfall and normalized difference vegetation index.

[0026] The net radiation flux is calculated by the upward longwave radiation flux, the downward longwave radiation flux, the upward shortwave radiation flux and the downward shortwave radiation flux.

[0027] The canopy temperature is calculated through atmospheric temperature, atmospheric humidity and net radiation flux.

[0028] The latent heat flux is calculated by atmospheric temperature, atmospheric humidity, atmospheric pressure, wind speed, upward longwave radiation flux, downward longwave radiation flux, upward shortwave radiation flux, downward shortwave radiation flux and normalized difference vegetation index.

[0029] The chlorophyll index is calculated using the green wave component and near-infrared band component in the multi-band solar spectrum and the green wave component and near-infrared band component in the plant reflectance spectrum.

[0030] The effective accumulated temperature is calculated by atmospheric temperature and soil temperature.

[0031] Therefore, the plant growth monitoring parameters obtained in this embodiment include eight meteorological environmental parameters, namely, atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, light intensity, rainfall, wind speed and wind direction; six soil environmental parameters, namely, soil temperature, soil moisture, soil conductivity, dynamic crop coefficient, canopy temperature and latent heat flux; four atmospheric radiation parameters, namely, shortwave radiation flux, longwave radiation flux, photosynthetically active radiation flux and net radiation flux; and five plant spectral parameters, namely, multi-band solar spectrum, plant reflectance spectrum, normalized vegetation index, chlorophyll index and effective accumulated temperature.

[0032] The data processing module 1 is also used to pre-process the raw data collected by the fixed sensor group 2 and the sensor expansion module 6, and select different correction models according to different plant growth monitoring parameters to correct the data; and package the pre-processed plant growth monitoring parameters in JSON (JavaScript Object Notation) format, and send the data to the cloud storage server 4 through the wireless transmission module 3 to facilitate subsequent data analysis and processing.

[0033] For example, the data to be packaged is: [Temperature: 10.51℃, Humidity: 37.35, Air pressure: 1000HPa, Light intensity: 100,000.12, Carbon dioxide content: 800.05PPM, Wind speed: 3.6m / s, Wind direction: 12.45°], The data packaged in JSON format is: {"command":"setValue","devType":"4","devId":"20230001","channelNo":"1","dataType":"6","data":["10.51","37.35","1000","100,000.12","800.05","3.6","12.45"]}, and ends with the hexadecimal data 0x0D 0x0A 0x0D 0x0A.

[0034] Among them, command is the request frame format, devType is the device type, devId is the device serial number, channelNo is the device data channel number, dataType is the data type, and data is the data collected by the device.

[0035] The cloud storage server 4 is also used for data storage, and through machine learning and big data analysis technology, a plant growth health assessment model and a yield prediction model are constructed to assess and analyze the health status of plants and crop yields. In addition, a corresponding service interface can be deployed on the cloud storage server 4 to simplify the user's operation settings.

[0036] The CAE-RNN combined neural network is used to construct a plant growth health assessment model and a yield prediction model.

[0037] Different plant growth monitoring parameters are used as feature data for different detection indicators. Therefore, when conducting health assessment, atmospheric temperature, atmospheric humidity, soil temperature, soil moisture, soil conductivity, multi-band solar spectrum, plant reflectance spectrum, canopy temperature, latent heat flux, net radiation flux, normalized vegetation index, chlorophyll index and effective accumulated temperature are selected as feature data, and the historical data on the classification and grading of plant leaf lesion types and lesion degrees are used as label data to train the CAE-RNN neural network model.

[0038] When making yield predictions, atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, light intensity, rainfall, wind speed, wind direction, soil moisture, soil temperature, soil conductivity, shortwave radiation flux, longwave radiation flux, photosynthetically active radiation flux, multi-band solar spectrum, plant reflectance spectrum, dynamic crop coefficient, normalized vegetation index and effective accumulated temperature are selected as feature data, and historical environmental monitoring data as well as the current year's yield and plant growth trend assessment indicators at various stages are used as label data to train the CAE-RNN neural network model.

[0039] The following combination Figure 2-Figure 4 The structure of the multi-parameter monitoring device provided in this embodiment is introduced.

[0040] The multi-parameter monitoring device includes a main body 101, an indicator light group (including a power indicator light 102, a charging status indicator light 103, a networking status indicator light 104, and a warning indicator light 105), a compass 106, a power button 107, a support rod 108, a support rod base 109, a power module 5, an external power supply interface 110, a sensor expansion module interface 111, a fixed sensor group 2 (photosynthetically active radiation sensor 201, a downstream spectrum sensor 202, a downstream shortwave radiation sensor 203, a downstream longwave radiation sensor 204, a light sensor 205, a rainfall sensor 206, a temperature, humidity and pressure sensor 207, a carbon dioxide sensor 208, an upstream shortwave radiation sensor 209, an upstream spectrum sensor 210, and an upstream longwave radiation sensor 211), a 4G antenna 301, a 4G antenna socket 302, and a data processing module 1.

[0041] The indicator light group, the compass 106 , the power button 107 , the power module 5 , the sensor expansion module interface 111 , the fixed sensor group 2 , the 4G antenna 301 and the data processing module 1 are all installed on the main body 101 .

[0042] The main body 101 is connected to the support rod 108 , and the support rod 108 is connected to the support rod base 109 .

[0043] The 4G antenna 301 is connected to the 4G antenna socket 302 .

[0044] The power module 5 is connected to an external power supply via the external power supply interface 110 .

[0045] The sensor expansion module interface 111 is connected to the sensor expansion module 6 .

[0046] The power indicator light 102 is used to display the on / off status of the multi-parameter monitoring device.

[0047] The charging status indicator light 103 is used to display the charging status of the multi-parameter monitoring device. When the device is fully charged, the blue light is on; when the device is charging, the red light is on.

[0048] The network status indicator light 104 is used to display the network status of the multi-parameter monitoring device. When the device is in the network search state, the indicator light will flash at a slower frequency, that is, it is in a slow flashing state, lighting up for 200 milliseconds and then extinguishing for 1800 milliseconds; when the device is in the standby state, the indicator light is in a slow flashing state, that is, it is lit up for 1800 milliseconds and then extinguished for 200 milliseconds; when the device is in the data transmission state, the indicator light will flash at a faster frequency, that is, it is in a fast flashing state, and the time of lighting up and extinguishing is 125 milliseconds.

[0049] The warning indicator light 105 is used to display abnormal conditions, and the lighting conditions can be set through software according to actual needs, such as excessively high temperature, abnormal soil conductivity, etc.

[0050] The compass 106 is used to mark the installation direction of the multi-parameter monitoring device.

[0051] The external power supply interface 110 is used to connect to an external solar cell to provide additional power supply.

[0052] The photosynthetically active radiation sensor 201 is responsible for collecting visible light within the wavelength range of 400-700nm. The light within this range directly affects the life activities of plants and is the energy source for the synthesis of organic matter in plants. In this embodiment, a filter is used in combination with a photoelectric sensor to design the photosynthetically active radiation sensor 201. The filter filters out the light radiation within the range of 400~700nm, and then the photoelectric sensor measures it. In terms of distribution position, the photosynthetically active radiation sensor 201 is located at a position where it can fully receive direct sunlight to accurately measure the radiation intensity within the spectral range required for plants to photosynthesize.

[0053] The downlink spectral sensor 202 and the uplink spectral sensor 210 are responsible for collecting spectral data of multiple bands within the range of 380-1100nm for evaluating the growth status of plants. In order to better meet the monitoring needs of users for different plants and different application scenarios, the spectral sensor provides universal spectra of multiple specific bands, such as 405nm, 425nm, 450nm, 515nm, 555nm, 600nm, 640nm, 670nm, 745nm, 855nm, etc. In terms of distribution position, the spectral sensor should be placed in a position that can fully reflect the spectral distribution in the plant growth environment and can fully receive the spectral distribution reflected by ground plants.

[0054] The downlink shortwave radiation sensor 203 and the uplink shortwave radiation sensor 209 are used to measure the energy of shortwave radiation in the 300-1100nm band and calculate the reflectivity of shortwave radiation. For shortwave radiation sensors, the measurement principle is usually based on the photoelectric effect, and the radiation intensity is indirectly reflected by measuring the current generated by photons hitting photoelectric elements. This embodiment uses a high-sensitivity silicon photodiode with an integrated 300-1100nm filter as the sensing element of the shortwave radiation sensor. The element has the advantages of small size, fast response speed, and wide spectral response range. In terms of distribution position, the shortwave radiation sensor should be placed in a position where it can fully receive direct solar radiation and scattered radiation to accurately measure the shortwave radiation intensity in the environment where the plant is located.

[0055] The downlink long-wave radiation sensor 204 and the uplink long-wave radiation sensor 211 are used to measure the energy of long-wave radiation in the 8μm-14μ band, and to calculate the reflectivity of long-wave radiation. The net radiation flux is calculated by combining the uplink short-wave radiation flux and the downlink short-wave radiation flux. For the long-wave radiation sensor, its measurement principle is mainly based on the thermal effect, and the radiation intensity is indirectly measured by sensing the temperature change caused by the long-wave radiation on the surface of the object. In order to reduce the volume, this embodiment uses a micro-thermistor or a thermopile sensor as the sensing element of the long-wave radiation sensor. In order to measure the long-wave radiation in both the uplink and downlink directions, and at the same time avoid the influence of the thermal effect on each other, they are staggered and distributed to improve the measurement accuracy.

[0056] The light sensor 205 is used to measure the light intensity in the planting area.

[0057] The rainfall sensor 206 is used to measure the rainfall in the planting area.

[0058] The temperature, humidity and pressure sensor 207 is used to measure the temperature, air humidity and atmospheric pressure in the planting area.

[0059] The carbon dioxide sensor 208 is used to measure the carbon dioxide content in the planting area.

[0060] External sensors such as a wind speed sensor, a wind direction sensor, a soil temperature sensor, a soil moisture sensor, and a soil conductivity sensor are connected to the sensor expansion module 6 in a cascade form.

[0061] The 4G antenna 301 is used to send the data collected by the multi-parameter monitoring device to the cloud storage server 4.

[0062] The power module 5 is used to provide a stable power supply for the multi-parameter monitoring device. The module adopts a variety of power supply solutions, including a high-capacity battery built into the device, a built-in solar panel, an optional external backup battery, and an environmentally friendly and efficient solar panel. In addition, the power module 5 is also equipped with an intelligent energy consumption management system, which can automatically adjust the power distribution strategy according to the actual operating status of the device and the external environmental conditions to ensure the effective use of electricity. For example, during the day when there is sufficient light, the system will give priority to solar power supply and charge the built-in and external batteries at the same time; at night or on cloudy days, it will switch to battery power supply mode. In addition, under working conditions where no external sensors are required, the internal battery and the internal solar panel can be used alone for power supply, which greatly extends the working time of the equipment and reduces the dependence on external power, improving the overall efficiency and sustainability of the system.

[0063] The hardware structure of the power module 5 is as follows Figure 5 shown.

[0064] In order to ensure that the multi-parameter monitoring equipment can operate stably in outdoor environments and manage power efficiently, its power module 5 is carefully designed as a three-level architecture, including external power supply, internal power supply and emergency charging and debugging functions.

[0065] External power supply mode: When the number of external sensors of the multi-parameter monitoring device needs to be increased, an external power supply is used. The external power supply system consists of a 12V lithium battery and a 60Wh solar panel to ensure that the external sensors can operate normally. While powering the external sensors, the solar panel can not only charge the 12V lithium battery, but also charge the internal 8.4V lithium battery to maintain the continuous operation of the device.

[0066] Internal battery power supply mode: When there is no need to expand external sensors, only the internal 8.4V lithium battery can be used to power the device. The internal lithium battery can be charged by the built-in solar panel, thereby reducing the size of the device and reducing unnecessary energy consumption.

[0067] Emergency charging debugging mode: When the device is in a bad working environment and the internal battery is exhausted, you can use the Type-C port on the device to charge the internal battery. When you need to obtain real-time operation data of the device, you can connect the Type-C port to the USB port of the computer and obtain real-time operation data through the serial port assistant.

[0068] In addition, the three power supply modes of external battery power, internal battery power and Type-C can all be converted into 3.3V1A power for MCU and some sensors through the power distribution system; converted into 5V1A power for AD converter; converted into 3.3V5A power for 4G module.

[0069] In order to meet the needs of different users, the plant growth monitoring system provided in this embodiment also includes a sensor expansion module 6. The sensor expansion module 6 is compatible with external sensors of various types of protocols, and can expand sensors of different protocols and unify the protocols, such as analog signals, RS485, serial digital components, etc.

[0070] By expanding external sensors, the meteorological environmental parameters collected by this application can also include PM2.5 concentration, PM10 concentration, oxygen concentration and other indicators monitoring; soil environmental parameters can include soil pH monitoring, nitrogen, phosphorus and potassium content monitoring, etc.; plant spectral parameters can be expanded to different bands, such as ultraviolet band, near infrared band, mid-infrared band, far infrared band, etc.

[0071] The specific structure of the sensor expansion module 6 is as follows Figure 6 shown.

[0072] The sensor expansion module 6 includes: a power supply interface 601 , a UART interface 602 , an RS485 interface 603 , an analog signal interface 604 , a serial digital component interface 605 and a host interface 606 .

[0073] The power interface 601 is connected to the power module 5 .

[0074] The UART interface 602 is used to connect sensors for serial communication.

[0075] The RS485 interface 603 is used to connect sensors of the MODBUS protocol.

[0076] The analog signal interface 604 is used to connect analog sensors, including 4-20mA current output type and 0-10V voltage output type sensors.

[0077] The serial digital component interface 605 is used to connect a sensor of a digital component signal.

[0078] The host interface 606 is used to connect to the sensor expansion module interface 111. When an external sensor is needed to collect other plant growth monitoring parameters, the data collected by the external sensor is sent to the data processing module 1 of the multi-parameter monitoring device by connecting the host interface 606 to the sensor expansion module interface 111, and then transmitted to the cloud storage server 4.

[0079] In order to break through the bottleneck of limited sensor types in traditional smart agricultural monitoring systems, this embodiment innovatively embeds an intelligent control interface in the cloud storage server 4. This interface can send protocol instructions to the data processing module 1 and the sensor expansion module 6 through the highly flexible wireless transmission module 3, realizing the seamless integration and dynamic configuration of a variety of different protocol sensors. The configuration information includes protocol type, data type and data format. Users only need to purchase the corresponding sensors according to actual needs and simply connect them to the sensor expansion module 6. Then, by accessing the service interface in the cloud, the user can easily complete the configuration of the sensor without complex hardware adjustments or additional programming knowledge. This process not only greatly improves the scalability and flexibility of the system, but also reduces the user's operating difficulty and cost, making the deployment and maintenance of the smart agricultural monitoring system more efficient and convenient. The functions of the intelligent control interface are as follows: Figure 7 shown.

[0080] In order to solve the problems of limited monitoring range and isolation of monitoring equipment between multiple points in the existing intelligent agricultural monitoring system, this embodiment innovatively deploys a regional collaborative monitoring method and a plant growth status assessment model (i.e., a plant growth health assessment model and a yield prediction model for assessing and analyzing plant growth conditions) in the cloud storage server 4 to improve data sharing between multiple monitoring points and realize plant growth status assessment and analysis.

[0081] Therefore, the plant growth monitoring parameters and the trained plant growth status assessment model are used to assess the growth status of the plant. The specific implementation process of this step is as follows: Figure 8 As shown, including: Step S1: Read plant growth monitoring parameters of multiple monitoring points within a planting area monitoring period T from a cloud storage database to obtain a plant growth monitoring data matrix.

[0082] ; in, represents the plant growth monitoring data matrix; represents the plant growth monitoring data of the jth monitoring point at the i-th moment, , ; n is the number of monitoring points; m is the number of monitoring time points within period T.

[0083] Step S2: Use principal component analysis to analyze the data Perform dimensionality reduction processing to screen out monitoring points with high correlation, that is, target monitoring points, and obtain a p-dimensional eigenvalue matrix, which is expressed as: ; in, is the eigenvalue of each dimension and satisfies .

[0084] The eigenvector matrix is: ; The matrix after dimensionality reduction is: ; in, Represents the plant growth monitoring data matrix The matrix after dimensionality reduction; represents the data of the jth monitoring point at the i-th time after dimensionality reduction, , ; p is the number of monitoring points with high correlation screened out after dimensionality reduction, ; m is the number of monitoring time points within the period T; where , then the selection of p point data from n monitoring points is completed.

[0085] Step S3: According to the monitoring points selected in step S2 (i.e., target monitoring points), record the normalized vegetation index of each target monitoring point. and chlorophyll index per unit area , calculate the weight information of each point among p points .

[0086] ; in," "represent Elements and Multiply the elements at corresponding positions in .

[0087] Step S4: The original data of the selected points Assign corresponding weight information As the weighted data, that is: .

[0088] Step S5: Select different plant growth parameters for different growth evaluation indicators, and perform dimensionality reduction through steps S1 to S5 to obtain a series of two-dimensional matrices. , .in, is the number of plant growth parameter types selected. The number of monitoring points is , the number of plant growth parameter types is The data is sliced, that is, Convert to , , and use the sliced ​​data as the input of the neural network model.

[0089] Step S6: Slice the data processed in S5 As the input of the convolutional autoencoder, it is encoded. Assume that convolution kernel, after the The features after convolution are: ; in, is the feature after the kth convolution, is the convolution kernel of the cth convolution operation, To encode part of the multiplication factor, is the coding part additive factor, is the ReLU activation function.

[0090] Step S7: Decode the data encoded in S6. The features after convolution are: ; in, For the The features after deconvolution are For the The convolution kernel of the deconvolution operation, To decode some of the multiplication factors, To decode the partial additive factors, is the ReLU activation function.

[0091] Step S8: The last decoded feature information After passing through the fully connected layer, the sequence features are extracted as the input of the RNN neural network. .

[0092] ; in, , Represent the weight parameters between the RNN input layer and the hidden layer, and between the hidden layers; , Represent the bias between the input layer and the hidden layer, and between the hidden layers respectively; Represents the output of the previous time step; for Activation function.

[0093] Step S9: Extract the sequence features The data is mapped into a one-dimensional vector through a fully connected layer and classified by the Softmax classifier to complete the plant growth status assessment.

[0094] For example, in the plant growth health assessment model, by collecting multi-dimensional plant growth multispectral time series and plant growth environment time series in the planting area, the Convolutional Autoencoder (CAE) is used to extract the appearance change features such as plant leaf color, which intuitively reflects the growth status and health of plants in the planting area. Then, the Recurrent Neural Network (RNN) is used to deeply explore the periodic influence of plant growth rate and environmental factors from the time dimension, providing technical support for plant growth prediction. The specific structure is as follows: Fig. 9 shown.

[0095] The plant growth monitoring system provided in this embodiment can obtain more plant growth parameters, thereby providing more accurate plant growth information. Furthermore, the sensor expansion module 6 and the intelligent control interface can expand the sensor category according to user needs, thereby reducing the cost of re-customizing a new monitoring system.

[0096] This embodiment adopts a regional collaborative monitoring method, which gives full play to the synergistic effect between data and can more accurately evaluate the growth status of plants.

[0097] This embodiment integrates the Internet of Things technology with multispectral sensing technology to achieve non-contact, non-destructive monitoring. By measuring the reflection, transmission or radiation characteristics of plants at different wavelengths, rich plant physiological information can be obtained, which helps to improve agricultural production efficiency, reduce production costs, reduce resource waste, and promote the sustainable development of agricultural production.

[0098] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, 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 this application.

Claims

1. A plant growth monitoring system, characterized in that: The plant growth monitoring system comprises: Multi-parameter monitoring equipment is used to obtain plant growth monitoring parameters; The multi-parameter monitoring device comprises: a main body, a support rod, a data processing module, a fixed sensor group and a sensor expansion module interface; The main body is connected to the support rod; The fixed sensor group, the data processing module and the sensor expansion module interface are all installed on the main body; The sensor expansion module interface is used to connect the sensor expansion module, wherein the sensor expansion module is used to connect the external sensor, and the external sensor is a sensor of multiple different types of communication protocols and communication interfaces; The fixed sensor group includes a photosynthetically active radiation sensor, a spectrum sensor, a shortwave radiation sensor, a longwave radiation sensor, a light sensor, a rainfall sensor, a temperature sensor, a humidity sensor, an air pressure sensor and a carbon dioxide sensor; The fixed sensor group and the sensor expansion module are used to collect first plant growth monitoring parameters, which include atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, light intensity, rainfall, wind speed, wind direction, soil temperature, soil moisture, soil conductivity, shortwave radiation flux, longwave radiation flux, photosynthetically active radiation flux, multi-band solar spectrum and plant reflectance spectrum; The data processing module is used to calculate the second plant growth monitoring parameter based on the first plant growth monitoring parameter, wherein the second plant growth monitoring parameter includes a dynamic crop coefficient, a canopy temperature, a latent heat flux, a net radiation flux, a normalized vegetation index, a chlorophyll index and an effective accumulated temperature.

2. The plant growth monitoring system according to claim 1, characterized in that: The plant growth monitoring system also includes: Cloud storage server; The cloud storage server is connected to the multi-parameter monitoring device; The cloud storage server is used to: According to the plant growth monitoring parameters and the trained plant growth health assessment model, the health status of the plant is assessed, wherein the trained plant growth health assessment model is a model obtained by training a CAE-RNN neural network model using atmospheric temperature, atmospheric humidity, soil temperature, soil humidity, soil conductivity, multi-band solar spectrum, plant reflectance spectrum, canopy temperature, latent heat flux, net radiation flux, normalized vegetation index, chlorophyll index and effective accumulated temperature as feature data, and using historical data for classifying and grading plant leaf lesion types and lesion degrees as label data; The crop yield is predicted according to the plant growth monitoring parameters and the trained yield prediction model, wherein the trained yield prediction model is obtained by training the CAE-RNN neural network model using atmospheric temperature, atmospheric humidity, atmospheric pressure, carbon dioxide content, light intensity, rainfall, wind speed, wind direction, soil moisture, soil temperature, soil conductivity, shortwave radiation flux, longwave radiation flux, photosynthetically active radiation flux, multi-band solar spectrum, plant reflectance spectrum, dynamic crop coefficient, normalized vegetation index and effective accumulated temperature as feature data, and using historical environmental monitoring data and the current year's yield and plant growth stage growth assessment indicators as labels.

3. The plant growth monitoring system according to claim 2, characterized in that: The cloud storage server also includes: an intelligent control interface; The intelligent control interface is used to integrate and configure information of the external sensor after the sensor expansion module is connected to the sensor expansion module interface.

4. The plant growth monitoring system according to claim 1, characterized in that: The multi-parameter monitoring device also includes: a power module; The power module is used to provide power and adjust the power distribution strategy according to the operating status of the multi-parameter monitoring device and external environmental conditions, wherein the power distribution strategy includes an external power supply mode, an internal battery power supply mode and an emergency charging debugging mode.

5. The plant growth monitoring system according to claim 1, characterized in that: The multi-parameter monitoring device also includes: an indicator light group; The indicator light group includes a power indicator light, a charging status indicator light, an Internet of Things indicator light and a warning indicator light; The power indicator light is used to display the switch status of the multi-parameter monitoring device; The charging status indicator light is used to display the charging status of the multi-parameter monitoring device; The Internet of Things indicator light is used to display the networking status of the multi-parameter monitoring device; The warning indicator light is used to display an abnormal state.

6. The plant growth monitoring system according to claim 2, characterized in that: In terms of evaluating the health status of the plant according to the plant growth monitoring parameters and the trained plant growth health assessment model, the cloud storage server is used to: According to the plant growth monitoring parameters of multiple monitoring points within the monitoring period, a plant growth monitoring data matrix is ​​obtained; Performing dimensionality reduction processing on the plant growth monitoring data matrix to obtain a dimensionality-reduced matrix; The monitoring points are screened according to the dimension-reduced matrix to obtain target monitoring points; Calculating the weight of each target monitoring point according to the normalized vegetation index and chlorophyll index of the target monitoring point; Obtaining a two-dimensional matrix according to the reduced-dimensional matrix and the weights; The two-dimensional matrix is ​​used as input, and the trained plant growth health assessment model is used to output the health status of the plant.

7. The plant growth monitoring system according to claim 1, characterized in that: The multi-parameter monitoring device further includes: a compass; the compass is installed on the main body, and the compass is used to mark the installation direction of the multi-parameter monitoring device.

8. The plant growth monitoring system according to claim 4, characterized in that: The sensor expansion module includes: a power interface, a UART interface, an RS485 interface, an analog signal interface, a serial digital component interface and a host interface; The power interface is used to connect the power module; The UART interface is used to connect a sensor for serial communication; The RS485 interface is used to connect the sensor of MODBUS protocol; The analog signal interface is used to connect an analog sensor; The serial digital component interface is used to connect a sensor of a digital component signal; The host interface is used to connect to the sensor expansion module interface.

9. The plant growth monitoring system according to claim 2, characterized in that: The plant growth monitoring system further includes: a wireless transmission module; the wireless transmission module is connected to the multi-parameter monitoring device and the cloud storage server respectively.

Images