Orchard irrigation system, method, electronic device, and storage medium

Abstract

The application relates to the technical field of intelligent irrigation, in particular to a fruit tree irrigation system and method, an electronic device and a storage medium. The system comprises an infrared thermal imager, which is in communication connection with an image acquisition device to pre-process the image of a target plant and send the pre-processed image to a processing unit; the processing unit is configured to receive the image of the target plant; the crown layer in the image of the target plant is identified based on a pre-trained neural network model to determine the crown layer temperature data of the target plant; wherein the crown layer temperature data further comprises micro-environment information for representing the local climate and environmental conditions of the growth of the target plant; according to the micro-environment information, a target strategy for calculating the water stress index of the target plant is determined; and according to the crown layer temperature data and the target strategy, the water stress index of the target plant is calculated to irrigate the target plant. A portable monitoring system for the field of fruit tree irrigation is provided.

Description

Technical Field

[0001] This application relates to the field of smart irrigation technology, and in particular to an irrigation system, method, electronic device and storage medium for fruit trees. Background Technology

[0002] Current smart irrigation typically relies on soil moisture (such as the water status of plant roots or soil) and meteorological data to irrigate crops. Specifically, these schemes mainly depend on soil moisture sensors and weather station data. By monitoring soil moisture content and combining it with reference to crop evapotranspiration and the characteristics of the crop itself, the water requirement of the crop is estimated, thereby making irrigation decisions. Although this type of method achieves on-demand water supply to a certain extent, it is essentially an indirect monitoring method. Soil moisture reflects the water status of the root zone environment, and meteorological data reflects atmospheric evaporation demand; neither can directly and in real time reflect the water physiological status of the crop plant itself.

[0003] For fruit trees, especially tall trees with complex canopy structures such as walnuts, water is absorbed from the roots and eventually lost through transpiration from the leaves (canopy). Currently, there is no corresponding smart irrigation solution for monitoring and irrigating such canopy-related plants.

[0004] Therefore, how to design an independent monitoring and irrigation system for targets with irregular shapes and three-dimensional structures, such as fruit tree canopies, has become an urgent problem to be solved. Summary of the Invention

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, this application provides a fruit tree irrigation system, method, electronic device and storage medium. The main purpose is to solve the problem that current smart irrigation lacks corresponding supporting facilities and has low monitoring accuracy when monitoring and irrigating canopy plants such as fruit trees.

[0006] To achieve the above objectives, the main technical solutions adopted in this application include:

[0007] In a first aspect, embodiments of this application provide a fruit tree irrigation system, comprising:

[0008] An infrared thermal imager is connected to an image acquisition device to preprocess images of the target plant and then sends the preprocessed images to the processing unit.

[0009] The processing unit is configured as follows:

[0010] Receive an image of the target plant;

[0011] The canopy of the target plant is identified in the image based on a pre-trained neural network model to determine the canopy temperature data of the target plant; wherein, the canopy temperature data also includes microenvironment information used to characterize the local climate and environmental conditions for the growth of the target plant;

[0012] Based on the microenvironment information, a target strategy for calculating the water stress index of the target plant is determined;

[0013] Based on the canopy temperature data and the target strategy, the water stress index of the target plant is calculated in order to irrigate the target plant.

[0014] Optionally, the image of the target plant includes a visible light image and an infrared thermal image;

[0015] The processing unit identifies the canopy in the image of the target plant based on a pre-trained neural network model to determine the canopy temperature data of the target plant. This includes: performing semantic segmentation on the visible light image to extract the canopy region mask matrix of the target plant; mapping the canopy region mask matrix to the infrared thermal image to determine the pixel set corresponding to the canopy region in the infrared thermal image; and based on the original radiation values ​​of the pixel set corresponding to the canopy region, calling a plant surface emissivity model and an atmospheric correction model for processing to determine the canopy temperature data of the target plant.

[0016] Optionally, the processing unit determines a target strategy for calculating the water stress index of the target plant based on the microenvironment information, including: calculating the water stress index of the target plant using a theoretical model when equivalent environmental parameters are available; wherein the local climate and environmental conditions represented by the equivalent environmental parameters are the same as those represented by the microenvironment information; calculating the water stress index of the target plant using an empirical model when the equivalent environmental parameters cannot be used and historical canopy temperature data matching the microenvironment information is pre-stored in the local database; and calculating the water stress index of the target plant using a simplified model when the equivalent environmental parameters cannot be used and historical canopy temperature data matching the microenvironment information does not exist in the local database.

[0017] Optionally, when the target strategy adopts the theoretical model, the processing unit calculates the water stress index of the target plant based on the canopy temperature data and the target strategy, including: calculating the first canopy temperature difference and the second canopy temperature difference using the Penman-Monteith equation based on the equivalent environmental parameters; and calculating the water stress index based on the first canopy temperature difference, the second canopy temperature difference, and the canopy temperature data; wherein the first canopy temperature difference is the difference between the canopy temperature and the ambient temperature under no water stress conditions, and the second canopy temperature difference is the difference between the canopy temperature and the ambient temperature under no transpiration conditions.

[0018] Optionally, when the target strategy employs the empirical model, the processing unit calculates the water stress index of the target plant based on the canopy temperature data and the target strategy, including: calculating the saturated vapor pressure difference based on the current air temperature and humidity; retrieving a historical non-water stress baseline from a local database that matches the microenvironment information; the baseline characterizing the relationship between the saturated vapor pressure difference and the canopy temperature difference; determining a first canopy temperature difference and a second canopy temperature difference based on the saturated vapor pressure difference and the historical non-water stress baseline; and calculating the water stress index based on the first canopy temperature difference, the second canopy temperature difference, and the canopy temperature data; wherein the first canopy temperature difference is the difference between the canopy temperature and the ambient temperature under no water stress conditions, and the second canopy temperature difference is the difference between the canopy temperature and the ambient temperature under no transpiration conditions.

[0019] Optionally, when the target strategy adopts the simplified model, the processing unit calculates the water stress index of the target plant based on the canopy temperature data and the target strategy, including: obtaining the reference temperature without water stress and the reference temperature without transpiration calibrated in the current microenvironment; and calculating the water stress index based on the canopy temperature data, the reference temperature without water stress, and the reference temperature without transpiration.

[0020] Optionally, the processing unit is further configured to: determine the corresponding water stress level based on the calculated water stress index; generate and output visualization information, the visualization information including at least: the canopy outline of the target plant, a pseudo-color temperature distribution map superimposed on the canopy outline, the value of the water stress index, and the water stress level; wherein the water stress level includes a level used to indicate irrigation demand.

[0021] Secondly, embodiments of this application provide a fruit tree irrigation method, applied to any of the fruit tree irrigation systems described in the first aspect above, the method comprising:

[0022] Preprocess the image of the target plant;

[0023] The canopy of the target plant is identified in the image based on a pre-trained neural network model to determine the canopy temperature data of the target plant; the canopy temperature data also includes the microenvironment information of the target plant.

[0024] Based on the microenvironment information, a target strategy for calculating the water stress index of the target plant is determined;

[0025] Based on the canopy temperature data and the target strategy, the water stress index of the target plant is calculated in order to irrigate the target plant.

[0026] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the fruit tree irrigation method described in the second aspect.

[0027] Fourthly, this application provides an electronic device including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the computer program to implement the fruit tree irrigation method described in the second aspect.

[0028] By employing the above technical solution, this application provides a fruit tree irrigation system, including an infrared thermal imager, which is communicatively connected to an image acquisition device to preprocess images of target plants and send the preprocessed images to a processing unit. The processing unit is configured to: receive images of target plants; identify the canopy in the images of target plants based on a pre-trained neural network model to determine the canopy temperature data of the target plants; wherein the canopy temperature data also includes microenvironmental information used to characterize the local climate and environmental conditions for the growth of the target plants; determine a target strategy for calculating the water stress index of the target plants based on the microenvironmental information; and calculate the water stress index of the target plants based on the canopy temperature data and the target strategy for irrigating the target plants. Compared with related technologies, this application first proposes a portable monitoring facility for fruit tree irrigation. Users only need to communicate with the image acquisition device via an infrared camera to take pictures of the target plants to obtain relevant temperature information for subsequent irrigation. Furthermore, regarding the treatment of canopy plants, image recognition of the target plant's canopy was performed. In addition to canopy temperature data, microenvironmental information characterizing the local climate and environmental conditions of the target plant's growth was determined. This considered the local climate conditions implicit in the image of the target plant, and thus determined the optimal calculation strategy for the target plant's water stress index based on the microenvironmental information. This ensured the scientific validity and reliability of the output results under different data conditions, and solved the technical problem of poor universality in calculating the water stress index using only a single strategy. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the architecture of an orchard irrigation system provided in an embodiment of this application;

[0030] Figure 2 This is a schematic flowchart of a fruit tree irrigation method provided in an embodiment of this application. Detailed Implementation

[0031] To better understand the above technical solutions, exemplary embodiments of this application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application can be understood more clearly and thoroughly, and that the scope of this application can be fully conveyed to those skilled in the art.

[0032] To address the shortcomings of current smart irrigation systems, such as the lack of supporting facilities and low monitoring accuracy when monitoring and irrigating canopy plants like fruit trees, this application proposes a fruit tree irrigation system. This method can be applied to fields such as agricultural management, remote agricultural monitoring, and smart irrigation. Figure 1 As shown, the system includes:

[0033] An infrared thermal imager is connected to an image acquisition device to preprocess images of the target plant and then sends the preprocessed images to the processing unit.

[0034] In this embodiment, the target plants mainly include fruit trees and other canopy plants with complex canopy structures. The images of the target plants primarily focus on capturing the canopy structure. An infrared thermal imager, acting as a front-end image temperature preprocessing terminal, communicates with the image acquisition device. After the image acquisition device acquires an image of the target plant, the infrared thermal imager preprocesses the image and sends the preprocessed image to the processing unit. The purpose of preprocessing is to convert the infrared radiation energy in the target plant image into a processable electrical signal, for example, converting the original visible light image into an infrared thermal image, thereby facilitating analysis by the processing unit.

[0035] As a feasible implementation method, users can use their mobile phones as image acquisition devices to take pictures of the canopy of the target plant. The infrared thermal imager communicates with the mobile phone through a communication interface, thereby enabling the preprocessing of the image of the target plant.

[0036] The processing unit is configured to: receive an image of a target plant; identify the canopy in the image of the target plant based on a pre-trained neural network model to determine the canopy temperature data of the target plant; determine a target strategy for calculating the water stress index of the target plant based on microenvironment information; and calculate the water stress index of the target plant based on the canopy temperature data and the target strategy in order to irrigate the target plant.

[0037] The images of the target plant received include visible light images and infrared thermal images. In this embodiment, after receiving the image of the target plant, the processing unit identifies the canopy in the image of the target plant using a pre-trained neural network model to determine the canopy temperature data of the target plant. The pre-trained neural network model can be an existing mature convolutional neural network model (such as CNN), which only needs to be able to perform semantic segmentation, canopy region identification, and temperature feature extraction based on visible light and infrared thermal images. Pre-training here means that the neural network model must be able to identify different types of fruit trees. For example, for walnut trees, the neural network model needs to be able to identify at least the type of walnut canopy in the image. If necessary, it can be pre-trained based on the shape features of the walnut canopy to ensure that the neural network model can determine the type of walnut tree based on the canopy in the image.

[0038] For canopy vegetation, since water is absorbed through the roots and ultimately lost through transpiration from the leaves (canopy), canopy temperature directly reflects the water balance. Monitoring canopy temperature provides a more real-time and accurate reflection of water stress in fruit trees than monitoring soil moisture. Canopy temperature data includes the average temperature of the canopy area, the maximum and minimum canopy temperatures, and a histogram of temperature distribution in the canopy area.

[0039] Furthermore, canopy temperature data also includes microenvironment information. Taking an orchard as an example, the microenvironment of an orchard is explained below. The orchard microenvironment refers to the small-scale, localized climatic and environmental conditions with unique physical, chemical, and biological characteristics formed within the orchard under the macro-regional climate background due to the interaction of factors such as topography, vegetation, water bodies, soil, and human management. It directly surrounds the fruit trees and their associated organisms and is the most direct and core ecological space affecting fruit tree growth and development, fruit quality formation, and the occurrence of pests and diseases. Microenvironment information is used to determine the target strategy for calculating the water stress index of the target plants, facilitating irrigation based on the water stress index.

[0040] In this embodiment, microenvironment information specifically refers to information used to characterize the local climate and environmental conditions for the growth of the target plant. Examples include the actual solar radiation intensity received by different parts of the target plant's canopy (the difference in light intensity between sun-facing and shaded leaves, and between outer and inner leaves. This difference directly affects stomatal opening and closing and transpiration rate, and is an important factor to consider when assessing water stress), the temperature and humidity distribution at different locations within the canopy (a relatively hot and humid microenvironment often forms inside a dense canopy, while the canopy edge is well-ventilated and has lower humidity; this can be specifically characterized by temperature and humidity data at different locations on the leaf surface), and the mutual shading and ventilation effects between adjacent plants (which can be characterized by the distance between plants).

[0041] It should be noted that while microenvironmental information can be represented by conventional values ​​such as temperature, humidity, and distance, it expresses environmental information about the target plant within a small, specific ecological space. It focuses more on the target plant itself and its immediate vicinity, rather than acquiring macro-meteorological data specific to a particular region (for example, conventional methods might only acquire local meteorological data and leaf temperature, while this embodiment considers radiation intensity, leaf / plant spacing, and other data in addition to the temperature at different leaf locations to reflect the environmental information of the target plant's current ecological space). This is a major improvement over conventional monitoring and irrigation methods that rely solely on macro-meteorological data. This allows the calculation strategy determined based on microenvironmental information to more accurately reflect the true environmental conditions of the target plant, thus providing a more reliable decision-making basis for the adaptive calculation of the water stress index. Finally, based on the determined calculation strategy and the canopy temperature data calculated in the previous step, the water stress index of the target plant is calculated for irrigation.

[0042] In one feasible implementation, the processing unit is further configured to: determine the corresponding water stress level based on the calculated water stress index; generate and output visualization information, which includes at least: the canopy outline of the target plant, a pseudo-color temperature distribution map superimposed on the canopy outline, the value of the water stress index, and the water stress level; wherein the water stress level includes a level used to indicate irrigation demand.

[0043] In this embodiment, the water stress index includes a water stress level, which is used to indicate irrigation. In one feasible implementation, such as... Figure 1 As shown, the fruit tree irrigation system also includes an irrigation unit for irrigating target plants according to the water stress level. Simultaneously, the water stress index, water stress level, and the canopy outline of the target plants collected in the aforementioned steps, along with the pseudo-color temperature distribution map superimposed on the canopy outline, can all be displayed on a display screen or a smart APP terminal to enhance the level of intelligence.

[0044] Optionally, the image of the target plant includes a visible light image and an infrared thermal image. The processing unit identifies the canopy in the image of the target plant based on a pre-trained neural network model to determine the canopy temperature data of the target plant, including: performing semantic segmentation processing on the visible light image to extract the canopy region mask matrix of the target plant; mapping the canopy region mask matrix to the infrared thermal image to determine the pixel set corresponding to the canopy region in the infrared thermal image; and based on the original radiation values ​​of the pixel set corresponding to the canopy region, calling the plant surface emissivity model and the atmospheric correction model for processing to determine the canopy temperature data of the target plant.

[0045] In this embodiment, the processing unit achieves accurate extraction of canopy temperature through dual-modal image collaborative analysis. Specifically, firstly, using the high-resolution texture information of the visible light image, a semantic segmentation model is used to accurately identify the canopy boundary and generate a pixel-level canopy region mask matrix. Then, the mask matrix is ​​mapped to an infrared thermal image to determine the pixel set corresponding to the canopy region in the infrared thermal image. Finally, the original radiation values ​​of these canopy pixels are corrected by combining the plant's unique surface emissivity characteristics and real-time atmospheric conditions, converting the original radiation values ​​into the actual canopy surface temperature to obtain the canopy temperature data of the target plant, providing a high-quality temperature data foundation for subsequent water stress analysis.

[0046] Optionally, the processing unit determines a target strategy for calculating the water stress index of the target plant based on the microenvironment information, including: calculating the water stress index of the target plant using a theoretical model when equivalent environmental parameters are available; wherein the local climate and environmental conditions represented by the equivalent environmental parameters are the same as those represented by the microenvironment information; calculating the water stress index of the target plant using an empirical model when equivalent environmental parameters are unavailable and historical canopy temperature data matching the microenvironment information is pre-stored in the local database; and calculating the water stress index of the target plant using a simplified model when equivalent environmental parameters are unavailable and historical canopy temperature data matching the microenvironment information does not exist in the local database.

[0047] The formula for calculating the Water Stress Index (PWSI) is shown in Formula 1:

[0048] (Formula 1)

[0049] The PWSI ranges from 0 to 1, representing water conditions from no water stress to no transpiration. In Formula 1, The observed canopy temperature difference is determined based on the canopy temperature data from the preceding steps; specifically, it is the difference between the average canopy temperature and the air temperature. ll dT represents the crown temperature difference under conditions of no water stress.ul dT represents the crown temperature difference under non-transpiration conditions. ll and dT ul The calculation strategy is determined based on microenvironment information through one of three models: theoretical, empirical, or simplified. Specifically, as mentioned above, the water stress index of the target plant is calculated using these models, which is equivalent to calculating dT. ll and dT ul The process.

[0050] This embodiment describes how to determine the target strategy for calculating the water stress index of a target plant based on microenvironment information. Specifically, considering computational complexity and data acquisition difficulties, three scenarios are considered. The first scenario involves calculating the water stress index of the target plant using a theoretical model when equivalent environmental parameters are available. Equivalent environmental parameters are those representing the same local climate and environmental conditions as those represented by the microenvironment information. These equivalent environmental parameters specifically include, for example, net radiation (R0). n ), aerodynamic impedance (r a ), canopy resistance (r cp Parameters such as leaf area index (LAI), air density, air heat capacity, and hygroscopic constant are difficult to obtain. These parameters are numerous, complex, and involve a large number of time dimensions; they cannot be fully obtained solely through infrared thermal images, and some parameters require specific permissions for access. For example, some parameters are taken from databases of specific plant species, while others are from local meteorological data. This is typically achieved by setting up external interfaces to connect to other databases and determining if these data can be retrieved. This approach is usually used when the application requires extremely high precision in understanding vegetation mechanisms and when complete information on the target plant's canopy physiology and microenvironment can be obtained or calculated.

[0051] The second approach involves calculating the water stress index of the target plant using an empirical model when equivalent environmental parameters cannot be accessed and historical canopy temperature data matching the microenvironment information is pre-stored in the local database. This means that when the target plant type is the same and local historical data is available, the model switches to an empirical model based on statistical regularities, using historical baseline data to calculate the water stress index. Whether the data matches the microenvironment information can be determined by setting a threshold. For example, if the ambient temperature for the walnut trees is 19℃, the average canopy temperature is 16℃, and the date is 10:00 AM in mid-July of a certain year, then historical data from the same period in July at 10:00 AM, with ambient temperatures between 18-20℃ and average canopy temperatures between 15.8-16.2℃, can be compared. If a match is found, it is considered a match to the microenvironment information. It should be noted that the above examples using date, ambient temperature, and average canopy temperature are for illustrative purposes; in actual matching, the conditions for matching based on microenvironment data are more numerous and require higher precision.

[0052] The third approach is to use a simplified model to calculate the water stress index of the target plant when the above two methods are not feasible, that is, when it is impossible to retrieve equivalent environmental data that is the same as the local climate and environmental conditions represented by the microenvironment information, and it is also impossible to find data that matches the microenvironment information in historical data.

[0053] Furthermore, when the target strategy employs a theoretical model, the treatment unit calculates the water stress index of the target plant based on canopy temperature data and the target strategy. This includes: calculating the temperature difference between the first and second crowns using the Penman-Monteith equation based on equivalent environmental parameters; and calculating the water stress index based on the temperature difference between the first and second crowns and the canopy temperature data. Here, the temperature difference between the first and second crowns is the difference between the canopy temperature and the ambient temperature under no water stress conditions, and the temperature difference between the second and third crowns is the difference between the canopy temperature and the ambient temperature under no transpiration conditions.

[0054] In this embodiment, the theoretical model calculates the temperature difference dT of the first crown using the Penman-Monteith equation and the energy balance equation. ll The temperature difference between the second crown and dT ul Among them, dT ll The calculation method is obtained by assuming that the canopy resistance is under potential evapotranspiration conditions, as shown in Formula 2:

[0055] (Formula 2)

[0056] In Formula 2, R n Net radiation; r a ρ is the aerodynamic impedance; ρ is the air density; cp γ is the air heat capacity; γ is the wet / dry constant; Δ is the slope of the saturated vapor pressure as a function of temperature; r cp The canopy resistance under potential evapotranspiration conditions is calculated using Formula 3:

[0057] (Formula 3)

[0058] In Formula 3, r min The minimum porosity of the blade can be determined through model parameter optimization; LAI e The effective leaf area index is calculated as shown in Formula 4:

[0059] (Formula 4)

[0060] In Formula 4, LAI stands for Leaf Area Index.

[0061] dT ul This is obtained by assuming that the canopy drag is infinite, i.e., the canopy drag r c =+∞, as shown in Formula 5:

[0062] (Formula 5)

[0063] In Formula 5, the parameter r a The aerodynamic impedance is improved by taking into account the influence of thermal factors on the aerodynamic impedance, as shown in Formula Six:

[0064] (Formula 6)

[0065] In Formula 6, z is the reference height; d is the zero-plane displacement; For momentum exchange roughness length; denoted as the heat exchange roughness length; u is the wind speed at the reference height.

[0066] In this embodiment, using real-time acquired equivalent environmental parameters (air temperature, humidity, wind speed, net radiation, and leaf area index, etc.), two theoretical limit values ​​are calculated using the Penman-Monteith equation: the first crown temperature difference assuming the plant is under conditions of sufficient water supply and unrestricted transpiration; and the second crown temperature difference assuming the plant completely stops transpiration and the transpiration cooling effect is zero. Subsequently, the actual crown temperature difference is calculated by combining measured crown temperature data, and the three crown temperature difference values ​​are substituted into the classical water stress index formula to obtain a quantitative index characterizing the current degree of water stress. Under complete data conditions, this method can provide the most accurate water stress diagnosis results and is suitable for scientific research monitoring or refined scenarios requiring high precision.

[0067] Furthermore, when the target strategy employs an empirical model, the processing unit calculates the water stress index of the target plant based on canopy temperature data and the target strategy. This includes: calculating the saturated vapor pressure difference based on the current air temperature and humidity; retrieving historical non-water stress baselines from the local database that match the microenvironment information; the baselines characterize the relationship between the saturated vapor pressure difference and the canopy temperature difference; determining the first and second canopy temperature differences based on the saturated vapor pressure difference and the historical non-water stress baselines; and calculating the water stress index based on the first and second canopy temperature differences and the canopy temperature data. The first canopy temperature difference is the difference between the canopy temperature and the ambient temperature under no-water stress conditions, and the second canopy temperature difference is the difference between the canopy temperature and the ambient temperature under no-transpiration conditions.

[0068] In this embodiment, the current saturated vapor pressure difference is first calculated based on real-time monitored air temperature and humidity. Then, a historical non-moisture stress baseline matching the current microenvironment information (such as season, phenological period, and weather type) is retrieved from the local database. This baseline is based on a statistical relationship model between the saturated vapor pressure difference and the canopy temperature difference, fitted from long-term observation data. This model can calculate the theoretically maximum canopy temperature difference under current environmental conditions, corresponding to no moisture stress and no transpiration. Finally, these theoretical maximum values ​​and the measured canopy temperature difference are substituted into the following formula. The empirical model fully utilizes the statistical regularities in local historical data to achieve precise adaptation to the localized climatic characteristics of specific orchards.

[0069] First crown temperature difference dT ll The temperature difference between the second crown and dT ul The calculation method is as follows:

[0070] (Formula 7)

[0071] (Formula 8)

[0072] Plant transpiration can only be completely suppressed when the atmospheric water vapor reaches a supersaturated state. Therefore, the saturated water vapor pressure difference at the non-water stress baseline is set as the water vapor pressure gradient, which is the difference between the saturated water vapor pressure difference at the current temperature and the saturated water vapor pressure difference when the temperature is set to the temperature plus the lower baseline intercept. The resulting crown temperature difference is taken as the crown temperature difference under non-transpiration conditions.

[0073] (Formula Nine)

[0074] (Formula 10)

[0075] In Equations 7 to 10, coefficients a and b are the slope and intercept of the historical non-moisture stress baseline, respectively, which depend on the solar altitude angle; VPD is the saturated vapor pressure difference; VPG is the vapor pressure gradient; Ta is the air temperature; and RH is the air humidity.

[0076] Finally, when the target strategy uses a simplified model, the processing unit calculates the water stress index of the target plant based on the canopy temperature data and the target strategy, including: obtaining the reference temperature without water stress and the reference temperature without transpiration calibrated in the current microenvironment; and calculating the water stress index based on the canopy temperature data, the reference temperature without water stress, and the reference temperature without transpiration.

[0077] In this embodiment, the user is prompted and guided to perform rapid calibration in the current microenvironment to obtain the calibrated reference temperature for no water stress and no transpiration. As a feasible implementation, the user is guided to select a healthy leaf, uniformly spray water to form a water film, and obtain its stable temperature under the current environment as the reference temperature for no water stress; another healthy leaf is selected, coated with Vaseline to completely suppress transpiration, and its stable temperature is obtained as the reference temperature for no transpiration. These two reference temperatures essentially utilize artificially created extreme-condition leaves to directly "measure" the theoretical minimum and maximum temperatures that the plant can reach under the current local environmental conditions. Then, the two reference temperatures are substituted into Formula 1 to calculate the water stress index. This simplified model requires no meteorological sensors or historical data support; benchmark calibration can be completed with just one simple on-site operation, providing a robust and reliable fallback calculation solution for the system in scenarios where data is completely scarce.

[0078] Furthermore, Figure 2 A schematic flowchart of a fruit tree irrigation method according to an embodiment of this application is shown, applicable to the fruit tree irrigation system mentioned in any of the above embodiments. The method includes:

[0079] S101, preprocess the image of the target plant.

[0080] S102, based on a pre-trained neural network model, identifies the canopy in the image of the target plant and determines the canopy temperature data of the target plant.

[0081] S103, Based on microenvironment information, determine the target strategy for calculating the water stress index of the target plant.

[0082] S104 calculates the water stress index of the target plants based on canopy temperature data and target strategy, in order to irrigate the target plants.

[0083] This embodiment proposes a portable monitoring system for fruit tree irrigation. Users simply need to connect an infrared camera to an image acquisition device to photograph the target plant to obtain relevant information for subsequent irrigation. Furthermore, regarding canopy plants, image recognition of the target plant's canopy reveals microenvironmental information, characterizing the local climate and environmental conditions, in addition to canopy temperature data. This considers the local climate conditions implicit in the image and allows for the determination of the optimal calculation strategy for the target plant's water stress index based on the microenvironmental information. This ensures the scientific validity and reliability of the output results under different data conditions, overcoming the technical limitation of poor universality when calculating the water stress index using only a single strategy.

[0084] Based on the above, Figure 2 Accordingly, this embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. Figure 2 The method shown.

[0085] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.

[0086] Based on the above, Figure 2 The method shown, and Figure 2 To achieve the above objectives, this application also provides an electronic device, which can be configured on a computer side, etc. The device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-described objectives. Figure 2 The method shown.

[0087] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0088] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0089] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0090] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented through hardware. By applying the solution of this embodiment, compared with related technologies, a portable monitoring facility for fruit tree irrigation is first proposed. Users only need to connect an infrared camera to an image acquisition device to photograph the target plant to obtain relevant information about the target plant for subsequent irrigation. Furthermore, regarding the treatment of canopy plants, by performing image recognition on the canopy of the target plant, in addition to canopy temperature data, microenvironmental information characterizing the local climate and environmental conditions of the target plant's growth is determined. This takes into account the local climate conditions implied in the image of the target plant, and then determines the optimal calculation strategy for the water stress index of the target plant based on the microenvironmental information. This ensures the scientific validity and reliability of the output results under different data conditions, solving the technical pain point of poor universality in calculating the water stress index using only a single strategy.

[0091] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0092] In the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0093] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A fruit tree irrigation system, characterized in that, include: An infrared thermal imager is connected to an image acquisition device to preprocess images of the target plant and then sends the preprocessed images to the processing unit. The processing unit is configured as follows: Receive an image of the target plant; The canopy of the target plant is identified in the image based on a pre-trained neural network model to determine the canopy temperature data of the target plant; wherein, the canopy temperature data also includes microenvironment information used to characterize the local climate and environmental conditions for the growth of the target plant; Based on the microenvironment information, a target strategy for calculating the water stress index of the target plant is determined; Based on the canopy temperature data and the target strategy, the water stress index of the target plant is calculated in order to irrigate the target plant.

2. The system according to claim 1, characterized in that, The images of the target plant include visible light images and infrared thermal images; The processing unit identifies the canopy in the image of the target plant based on a pre-trained neural network model, and determines the canopy temperature data of the target plant, including: The visible light image is subjected to semantic segmentation processing to extract the canopy region mask matrix of the target plant; The canopy region mask matrix is ​​mapped onto the infrared thermal image to determine the pixel set corresponding to the canopy region in the infrared thermal image; Based on the original radiation values ​​of the pixel set corresponding to the canopy region, the plant surface emissivity model and atmospheric correction model are called for processing to determine the canopy temperature data of the target plant.

3. The system according to claim 1, characterized in that, The processing unit determines a target strategy for calculating the water stress index of the target plant based on the microenvironment information, including: When equivalent environmental parameters are available, the water stress index of the target plant is calculated using a theoretical model; wherein the local climate and environmental conditions represented by the equivalent environmental parameters are the same as the local climate and environmental conditions represented by the microenvironment information. When the equivalent environmental parameters cannot be accessed and historical canopy temperature data matching the microenvironment information is pre-stored in the local database, the water stress index of the target plant is calculated using an empirical model. When the equivalent environmental parameters cannot be accessed and there is no historical canopy temperature data matching the microenvironment information in the local database, the water stress index of the target plant is calculated using a simplified model.

4. The system according to claim 3, characterized in that, When the target strategy employs the theoretical model, the processing unit calculates the water stress index of the target plant based on the canopy temperature data and the target strategy, including: Based on the equivalent environmental parameters, the temperature difference between the first crown and the second crown is calculated using the Penman-Monteith equation. The water stress index is calculated based on the first crown temperature difference, the second crown temperature difference, and the crown temperature data; Wherein, the first crown temperature difference is the difference between the crown temperature and the ambient temperature under conditions of no water stress, and the second crown temperature difference is the difference between the crown temperature and the ambient temperature under conditions of no transpiration.

5. The system according to claim 3, characterized in that, When the target strategy employs the empirical model, the processing unit calculates the water stress index of the target plant based on the canopy temperature data and the target strategy, including: Calculate the saturated water vapor pressure difference based on the current air temperature and humidity; From the local database, retrieve historical non-moisture stress baselines that match the microenvironment information; these baselines characterize the relationship between saturated vapor pressure differential and crown temperature differential. Based on the saturated vapor pressure difference and the historical non-moisture stress baseline, the temperature difference between the first crown and the second crown is determined; The water stress index is calculated based on the first crown temperature difference, the second crown temperature difference, and the crown temperature data; Wherein, the first crown temperature difference is the difference between the crown temperature and the ambient temperature under conditions of no water stress, and the second crown temperature difference is the difference between the crown temperature and the ambient temperature under conditions of no transpiration.

6. The system according to claim 3, characterized in that, When the target strategy employs the simplified model, the processing unit calculates the water stress index of the target plant based on the canopy temperature data and the target strategy, including: Obtain the reference temperature for no moisture stress and the reference temperature for no transpiration calibrated under the current microenvironment; The water stress index is calculated based on the canopy temperature data, the reference temperature without water stress, and the reference temperature without transpiration.

7. The system according to claim 1, characterized in that, The processing unit is further configured to: Based on the calculated water stress index, the corresponding water stress level is determined; Generate and output visualization information, which includes at least: the canopy outline of the target plant, a pseudo-color temperature distribution map superimposed on the canopy outline, the value of the water stress index, and the water stress level; The water stress level includes a level used to indicate irrigation demand.

8. A method for irrigating fruit trees, applied to a fruit tree irrigation system as described in any one of claims 1-7, characterized in that, The method includes: Preprocess the image of the target plant; The canopy of the target plant is identified in the image based on a pre-trained neural network model to determine the canopy temperature data of the target plant; the canopy temperature data also includes the microenvironment information of the target plant. Based on the microenvironment information, a target strategy for calculating the water stress index of the target plant is determined; Based on the canopy temperature data and the target strategy, the water stress index of the target plant is calculated in order to irrigate the target plant.

9. An electronic device, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, The processor implements the method of claim 8 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of claim 8.

Images