Crop moisture utilization rate estimation method and system based on unmanned aerial vehicle remote sensing data

Abstract

The invention provides a crop moisture utilization rate estimation method and system based on unmanned aerial vehicle remote sensing data, and the method comprises the steps: obtaining meteorological data of a crop planting region, and laser radar point cloud data, multispectral data and thermal imaging data of a target crop group in the crop planting region; according to the multispectral data, the thermal imaging data and the meteorological data, obtaining the corresponding target evapotranspiration of the target crop group in each day; inputting the laser radar point cloud data, the multispectral data and the thermal imaging data into the crop population overground part dry matter accumulation amount prediction model to obtain overground part dry matter weight prediction data; according to the overground part dry matter weight prediction data corresponding to different prediction moments, performing fitting to obtain a target dry matter accumulation amount corresponding to the target crop group in each day; and obtaining the crop water utilization rate of the target crop group according to the target evapotranspiration and the target dry matter accumulation amount. The precision and efficiency of crop water utilization rate estimation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of crop moisture detection, and in particular to a crop moisture utilization rate estimation method and system based on unmanned aerial vehicle (UAV) remote sensing data. Background Art

[0002] Water Use Efficiency (WUE) is an important indicator for evaluating crop productivity and water use capacity. Existing WUE measurement methods mainly include field measurement methods, crop model simulation methods and remote sensing estimation methods.

[0003] However, existing WUE measurement methods, such as field measurement methods, are usually accompanied by destructive sampling, which is time-consuming and inefficient, making continuous observation difficult. Model simulation methods are based on crop models with regionalized genetic parameters, and the estimation accuracy of WUE at the field scale is still insufficient. Although satellite platforms can estimate WUE over large areas, their resolution is low and it is difficult to meet the needs of precision agriculture.

[0004] Therefore, there is an urgent need for a crop water use efficiency estimation method and system based on UAV remote sensing data to solve the above problems. Summary of the Invention

[0005] In response to the problems existing in the prior art, the present invention provides a crop water use efficiency estimation method and system based on UAV remote sensing data.

[0006] The present invention provides a method for estimating crop water use efficiency based on UAV remote sensing data, comprising: Acquiring meteorological data for a crop-growing area, as well as laser radar point cloud data, multispectral data, and thermal imaging data of a target crop population within the crop-growing area, wherein the laser radar point cloud data, the multispectral data, and the thermal imaging data are collected using a remote sensing sensor based on an unmanned aerial vehicle; Obtaining a target evapotranspiration of the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data; Inputting the lidar point cloud data, the multispectral data, and the thermal imaging data into a crop population aboveground dry matter weight prediction model to obtain aboveground dry matter weight prediction data output by the crop population aboveground dry matter weight prediction model, wherein the crop population aboveground dry matter weight prediction model is trained based on a machine learning model; fitting the aboveground dry matter weight prediction data corresponding to different prediction moments to obtain the target dry matter accumulation amount of the target crop group on each day; The crop water use efficiency of the target crop group is obtained according to the target evapotranspiration and the target dry matter accumulation.

[0007] According to a crop water use efficiency estimation method based on UAV remote sensing data provided by the present invention, obtaining the target evapotranspiration corresponding to the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data includes: Obtaining, based on the multispectral data, a normalized difference vegetation index, albedo, and atmospheric emissivity corresponding to the crop planting area; Obtaining the surface radiation temperature corresponding to the crop planting area according to the thermal imaging data; Calculating the instantaneous net surface radiation flux of the crop planting area according to the normalized difference vegetation index, the albedo, the atmospheric emissivity and the surface radiation temperature; Calculating the instantaneous soil heat flux of the crop planting area according to the instantaneous surface net radiation flux, the albedo, the surface radiation temperature and the normalized difference vegetation index; Calculating a surface temperature gradient value based on the meteorological data, and calculating an instantaneous sensible heat flux in the crop planting area based on a linear relationship between the surface radiation temperature and the surface temperature gradient value; The instantaneous evapotranspiration is calculated based on the instantaneous net surface radiation flux, the instantaneous soil heat flux and the instantaneous sensible heat flux, and the instantaneous evapotranspiration is upscaled to obtain the target evapotranspiration corresponding to the target crop group on each day.

[0008] According to a method for estimating crop water use efficiency based on UAV remote sensing data provided by the present invention, the target evapotranspiration is calculated based on a crop group daily evapotranspiration model, and the formula of the crop group daily evapotranspiration model is: ; ; ; in, represents the instantaneous evapotranspiration, represents the instantaneous net surface radiation flux, represents the instantaneous soil heat flux, represents the instantaneous sensible heat flux, represents the evaporation fraction, represents the target evapotranspiration, represents the latent heat of vaporization of water, represents the net surface radiation flux corresponding to the crop planting area every day, It represents the soil heat flux corresponding to the crop planting area on each day.

[0009] According to a crop water use efficiency estimation method based on drone remote sensing data provided by the present invention, the aboveground dry matter weight prediction data corresponding to different prediction moments are fitted to obtain the target dry matter accumulation amount corresponding to the target crop group on each day, including: Obtaining the maximum aboveground dry matter weight, the first accumulation time, and the second accumulation time based on the aboveground dry matter weight prediction data corresponding to different prediction moments, wherein the maximum aboveground dry matter weight is determined based on the aboveground dry matter weight prediction data corresponding to each prediction moment; the first accumulation time is the time corresponding to when the target dry matter accumulation amount reaches the maximum dry matter accumulation rate, and the second accumulation time is the time corresponding to when the aboveground dry matter weight prediction data reaches the maximum value; Based on a preset aboveground dry matter weight fitting model, the maximum aboveground dry matter weight, the first accumulation time, the second accumulation time, and the aboveground dry matter weight prediction data corresponding to each prediction moment are fitted to obtain the target dry matter accumulation amount corresponding to the target crop group on each day, wherein the preset aboveground dry matter weight fitting model is constructed based on a beta function.

[0010] According to a crop water use efficiency estimation method based on UAV remote sensing data provided by the present invention, the formula of the preset aboveground dry matter weight fitting model is: ; ; ; ; in, represents the target dry matter accumulation amount, represents the maximum dry matter accumulation rate; represents the first accumulation time, which is the time when the target dry matter accumulation amount reaches the maximum dry matter accumulation rate; represents the second accumulated data, which is the time when the above-ground dry matter weight prediction data reaches the maximum above-ground dry matter weight; represents the time corresponding to the aboveground dry matter weight prediction data at the current moment, represents the maximum aboveground dry matter weight, It represents the predicted data of aboveground dry matter weight.

[0011] According to a method for estimating crop water use efficiency based on drone remote sensing data provided by the present invention, the crop population aboveground dry matter weight prediction model is trained by the following steps: Acquiring sample remote sensing data, wherein the sample remote sensing data includes historical lidar point cloud data, historical multispectral data, and historical thermal imaging data of the crop planting area; generating an aboveground dry matter weight label corresponding to the sample remote sensing data according to the aboveground dry matter weight of the crop population at a historical moment in the crop planting area; constructing a training sample set based on the sample remote sensing data and the aboveground dry matter weight labels corresponding to the sample remote sensing data; The automated machine learning model is trained based on the training sample set to obtain a prediction model for the aboveground dry matter weight of the crop population.

[0012] The present invention also provides a crop water use efficiency estimation system based on UAV remote sensing data, comprising: A remote sensing data acquisition module is used to obtain meteorological data of the crop planting area, as well as laser radar point cloud data, multispectral data, and thermal imaging data of the target crop population in the crop planting area, wherein the laser radar point cloud data, the multispectral data, and the thermal imaging data are collected by remote sensing sensors based on drones; an evapotranspiration calculation module, configured to obtain a target evapotranspiration corresponding to the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data; an above-ground dry weight prediction module, configured to input the lidar point cloud data, the multispectral data, and the thermal imaging data into a crop population above-ground dry weight prediction model to obtain above-ground dry weight prediction data output by the crop population above-ground dry weight prediction model, wherein the crop population above-ground dry weight prediction model is trained based on a machine learning model; a daily dry matter accumulation calculation module, configured to obtain the target dry matter accumulation of the target crop group on each day based on the aboveground dry matter weight prediction data corresponding to different prediction moments; The water use efficiency estimation module is used to obtain the crop water use efficiency of the target crop group according to the target evapotranspiration and the target dry matter accumulation.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for estimating crop water utilization rate based on drone remote sensing data as described above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for estimating crop water use efficiency based on drone remote sensing data as described above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for estimating crop water use efficiency based on drone remote sensing data.

[0016] The present invention provides a crop water use efficiency estimation method and system based on unmanned aerial vehicle (UAV) remote sensing data. The method and system obtain meteorological data, lidar point cloud data, multispectral data, and thermal imaging data of the crop planting area through UAV remote sensing sensors. The multispectral, thermal imaging, and meteorological data are then used to calculate the daily evapotranspiration of the target crop population. Simultaneously, the point cloud, multispectral, and thermal imaging data are input into a machine learning-trained aboveground dry matter weight prediction model for the crop population to obtain aboveground dry matter weight prediction data, and the daily dry matter accumulation is fitted. Finally, the crop water use efficiency is calculated by combining the evapotranspiration and dry matter accumulation, thereby improving the accuracy and efficiency of crop water use efficiency estimation and providing strong support for precision agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flow chart of a method for estimating crop water use efficiency based on UAV remote sensing data provided by the present invention; Figure 2 This is a schematic diagram of the structure of the crop water use efficiency estimation system based on UAV remote sensing data provided by the present invention; Figure 3 This is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0020] Existing WUE measurement methods mainly include field measurement methods, crop model simulation methods and remote sensing estimation methods. Among them, field measurement methods include direct measurement methods, gas exchange measurement methods, stable isotope technology, pot weighing method and eddy correlation technology. The direct measurement method is to calculate WUE at the field scale by directly measuring field water consumption and measured yield; eddy correlation technology is the technology that can measure WUE on a larger scale among all field measurement methods; crop model simulation methods are usually driven by meteorological data, integrating soil parameters, crop parameters, and field management measures to simulate water consumption and yield, and calculate WUE at a single point scale; remote sensing estimation methods usually use satellite platforms to estimate WUE. Satellite platforms use data to produce gross primary productivity (GPP) and surface evapotranspiration (ET) products, and directly calculate regional ecosystem WUE through the ratio of GPP to ET.

[0021] However, among the existing WUE measurement methods mentioned above, direct measurement and gas exchange measurement have the narrowest applicability, being applicable only to instantaneous WUE measurements at the leaf scale. Stable isotope techniques offer high accuracy but are costly and technically demanding. Potted plant weighing methods have the narrowest applicability, being applicable only to potted plants. Furthermore, while field measurement methods can produce accurate WUE results, the process typically involves destructive sampling, resulting in lengthy and inefficient measurements and difficulty in continuous observation.

[0022] While existing model simulation methods have improved estimation efficiency compared to field measurements, crop models based on regionalized genetic parameters still have limited accuracy at the field-scale for WUE. While satellite platforms can estimate WUE over large areas, these methods often suffer from low resolution, making them difficult to meet the demands of precision agriculture.

[0023] To address the aforementioned challenges in the existing technologies, the present invention combines multimodal remote sensing data with evapotranspiration modeling to accurately estimate water use efficiency at the varietal level for corn, providing information technology support for the large-scale screening of high-water-efficiency corn germplasm resources and varieties. It should be noted that the crop water use efficiency estimation method based on drone remote sensing data provided by the present invention can also be applied to other crops and is not specifically limited to specific crop types.

[0024] Figure 1 This is a flow chart of the crop water use efficiency estimation method based on UAV remote sensing data provided by the present invention, as shown in FIG. Figure 1 As shown, the present invention provides a method for estimating crop water use efficiency based on UAV remote sensing data, comprising: Step 101: Acquire meteorological data of a crop planting area, as well as lidar point cloud data, multispectral data, and thermal imaging data of a target crop group in the crop planting area, wherein the lidar point cloud data, the multispectral data, and the thermal imaging data are collected based on a remote sensing sensor of an unmanned aerial vehicle.

[0025] In the present invention, meteorological data can be obtained through professional data provided by local meteorological stations in the crop planting area. These meteorological stations are usually equipped with high-precision meteorological monitoring equipment and can record various meteorological elements including temperature, humidity, wind speed, wind direction, precipitation and light intensity in real time and accurately.

[0026] In terms of remote sensing data collection, the present invention uses a drone as a remote sensing platform equipped with a series of remote sensing sensors to obtain lidar point cloud data, multispectral data and thermal imaging data of target crop groups in the crop planting area.

[0027] In this invention, lidar point cloud data is acquired using a lidar sensor integrated into a drone. The lidar emits a laser beam and measures its return time, accurately calculating the distance and three-dimensional coordinates of the target object. During flight, the lidar continuously scans the crop planting area, generating a large amount of point cloud data. This data records spatial information about the crop and its surroundings, including its height, crown width, and volume, in the form of points. This data provides rich data support for 3D modeling and analysis of crop growth conditions.

[0028] Multispectral data is acquired by drone-mounted multispectral cameras, which capture images across multiple spectral bands, ranging from visible light to near-infrared. By analyzing the reflectance characteristics of crops in images of different bands, spectral information can be obtained, providing insights into crop growth, chlorophyll content, moisture status, and other physiological and biochemical indicators.

[0029] Thermal imaging data is acquired using thermal infrared sensors on drones. These sensors detect infrared radiation emitted by objects and convert it into thermal images. In crop-growing areas, different objects (such as crops, soil, and water bodies) emit infrared radiation of varying intensities due to temperature differences. Thermal imaging technology allows for intuitive visualization of these temperature distributions.

[0030] Step 102: Obtain the target evapotranspiration of the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data.

[0031] In this paper, the Surface Energy Balance Algorithm for Land (SEBAL) is used to construct a daily evapotranspiration model for crop populations. SEBAL is based on the principle of surface energy balance, which states that solar radiation received by the surface is distributed between sensible heat flux, latent heat flux (i.e., evapotranspiration), and soil heat flux. SEBAL then estimates surface evapotranspiration through a series of algorithms and formulas, combined with multispectral data, thermal imaging data, and meteorological data.

[0032] In this paper, multispectral data is used to calculate surface parameters such as the Normalized Difference Vegetation Index (NDVI). These parameters describe crop growth and vegetation cover, which in turn influence surface energy distribution and evapotranspiration. Thermal imaging data is also used to obtain the surface radiation temperature of crops, a key parameter for calculating sensible and latent heat fluxes. By processing and analyzing thermal images, surface temperature information can be accurately extracted. Furthermore, collected meteorological data, such as temperature, humidity, wind speed, and solar radiation, are input into the SEBAL model to provide the necessary boundary conditions and driving factors for the model's calculations.

[0033] In this paper, the daily evapotranspiration model for crop populations, constructed based on SEBAL, calculates the surface sensible heat flux, latent heat flux (i.e., evapotranspiration), and soil heat flux based on the surface energy balance principle and the aforementioned input parameters. During the calculation process, the model considers the influence of various factors such as surface vegetation cover, soil properties, and meteorological conditions. Through iterative calculations and parameter adjustments, the instantaneous evapotranspiration for each remote sensing image acquired by the drone is calculated according to the surface energy balance equation. , and used as the residual to perform upscaling calculation, and then converted into the target evapotranspiration within a day .

[0034] Step 103: input the lidar point cloud data, the multispectral data, and the thermal imaging data into a crop population aboveground dry weight prediction model to obtain aboveground dry weight prediction data output by the crop population aboveground dry weight prediction model, wherein the crop population aboveground dry weight prediction model is obtained based on training of a machine learning model.

[0035] In this paper, the crop population above-ground dry matter weight prediction model is trained using a machine learning model. The training process utilizes a large amount of historical data, including lidar point cloud data, multispectral data, thermal imaging data, and corresponding actual above-ground dry matter weight (AGB) measurements. This data is divided into a training set and a validation set. The training set allows the model to learn the mapping relationship between the data, while the validation set is used to evaluate the model's performance and generalization ability.

[0036] During training, the machine learning model continuously adjusts its parameters to minimize the error between predicted and actual values. Through repeated iterations and optimization, the model gradually learns the complex relationships between different data types (lidar point cloud, multispectral, thermal) and aboveground dry matter mass. For example, the model may discover a strong positive correlation between certain lidar point cloud features (such as crop height distribution) and biomass, while changes in reflectance in certain multispectral bands are associated with biomass growth trends.

[0037] The trained crop population aboveground dry matter weight prediction model is capable of predicting new input data. When collected LiDAR point cloud data, multispectral data, and thermal imaging data are fed into the model, it calculates the predicted aboveground dry matter weight of the crop population based on previously learned patterns and rules, and outputs the predicted value as data.

[0038] Step 104 : fitting the aboveground dry matter weight prediction data corresponding to different prediction moments to obtain the target dry matter accumulation amount of the target crop group on each day.

[0039] In this paper, a machine learning model is used, combined with lidar point cloud data, multispectral data and thermal imaging data, to obtain the aboveground dry matter weight (AGB) prediction data corresponding to different prediction times. These data reflect the biomass status of the target crop population at different time points and serve as the basis for subsequent fitting and calculation of dry matter accumulation.

[0040] The beta function offers excellent flexibility and adaptability, capable of describing trends in a variety of forms. In this paper, the beta function is selected to fit AGB data at different forecast times, effectively capturing the dynamics of AGB changes over time. Furthermore, the beta function is used as input for fitting the predicted AGB data at different forecast times. The goal of the fitting process is to find the optimal combination of beta function parameters that most accurately describes the AGB trend over time. This fitting process generates a continuous AGB curve, which reflects the dry matter accumulation process of the target crop population throughout its entire growth cycle.

[0041] Furthermore, on the fitted AGB variation curve, each point corresponds to a specific time and AGB value, and the slope represents the rate of change of AGB at that point in time. The slope of the AGB variation curve corresponding to the date the drone data was acquired can be calculated. This slope value represents the daily dry matter accumulation, i.e., the target dry matter accumulation, and thus the daily dry matter accumulation sequence of the target crop population throughout its entire growth cycle is obtained.

[0042] Step 105 : obtaining the crop water use efficiency of the target crop group according to the target evapotranspiration and the target dry matter accumulation.

[0043] In the present invention, The daily water use efficiency can be calculated by taking the target evapotranspiration of the drone on the day of flight as and target dry matter accumulation The estimated formula is: .

[0044] The crop population water utilization rate calculated by the present invention can greatly improve the efficiency of selecting high-water-efficiency materials or varieties, thereby saving manual sampling costs and promoting the high-water-efficiency breeding process.

[0045] The present invention provides a crop water use efficiency (WUE) estimation method based on unmanned aerial vehicle (UAV) remote sensing data. The method uses UAV remote sensing sensors to obtain meteorological data, lidar point cloud data, multispectral data, and thermal imaging data from crop planting areas. The multispectral, thermal, and meteorological data are then used to calculate the daily evapotranspiration of a target crop population. Simultaneously, the point cloud, multispectral, and thermal imaging data are input into a machine learning-trained aboveground dry matter weight prediction model for the crop population to obtain aboveground dry matter weight prediction data, and the daily dry matter accumulation is fitted. Finally, the crop water use efficiency (WUE) is calculated by combining the daily evapotranspiration and daily dry matter accumulation, thereby improving the accuracy and efficiency of WUE estimation and providing strong support for precision agriculture.

[0046] Based on the above embodiment, obtaining the target evapotranspiration of the target crop group on each day according to the multispectral data, the thermal imaging data, and the meteorological data includes: Obtaining, based on the multispectral data, a normalized difference vegetation index, albedo, and atmospheric emissivity corresponding to the crop planting area; Obtaining the surface radiation temperature corresponding to the crop planting area according to the thermal imaging data; Calculating the instantaneous net surface radiation flux of the crop planting area according to the normalized difference vegetation index, the albedo, the atmospheric emissivity and the surface radiation temperature; Calculating the instantaneous soil heat flux of the crop planting area according to the instantaneous surface net radiation flux, the albedo, the surface radiation temperature and the normalized difference vegetation index; Calculating a surface temperature gradient value based on the meteorological data, and calculating an instantaneous sensible heat flux in the crop planting area based on a linear relationship between the surface radiation temperature and the surface temperature gradient value; The instantaneous evapotranspiration is calculated based on the instantaneous net surface radiation flux, the instantaneous soil heat flux and the instantaneous sensible heat flux, and the instantaneous evapotranspiration is upscaled to obtain the target evapotranspiration corresponding to the target crop group on each day.

[0047] In the present invention, the normalized difference vegetation index NDVI and albedo are obtained based on multispectral data. α and atmospheric emissivity ε , among which, the Normalized Difference Vegetation Index (NDVI) can be calculated by the reflectance of specific bands (such as red light and near-infrared bands) in multispectral data. The Normalized Difference Vegetation Index (NDVI) can reflect the coverage and growth status of vegetation. Albedo α Indicates the ability of the surface to reflect solar radiation. It is also calculated through multispectral data. The size of the albedo is affected by many factors such as surface cover type and vegetation conditions. ε It can be estimated based on NDVI, which reflects the absorption and emission characteristics of the atmosphere to thermal radiation.

[0048] At the same time, the surface radiation temperature is obtained based on thermal imaging data T s , where the surface radiation temperature T s It can be directly measured through thermal imaging data. It is the temperature corresponding to the thermal radiation emitted by surface objects and can reflect the thermal conditions of the surface.

[0049] Furthermore, the obtained normalized difference vegetation index NDVI, albedo α , atmospheric emissivity ε and surface radiation temperature T s , combined with the surface energy balance principle, the instantaneous surface net radiation flux is calculated R n Instantaneous surface net radiation flux R n The difference between the total radiation energy received by the surface and the total radiation energy lost, the instantaneous surface net radiation flux R n The calculation formula is: ; ; ; ; ; ; in, represents the incident shortwave radiation, and are the incident and outgoing long-wave radiation, respectively; S c is the solar constant, approximately 1367 W / m²; θ is the solar zenith angle (related to the time, longitude and latitude of the remote sensing image acquisition); τ is the atmospheric transmittance, generally between 0.6-0.8; is the reflectivity of red light, is the reflectivity in the near-infrared band, ε is the atmospheric emissivity; is the surface emissivity, which is generally 0.98–0.99 in vegetated areas and slightly lower in bare land; σ is the Stefan-Boltzmann constant, T a is the atmospheric temperature (unit K), provided by the weather station. R n It is primarily related to the atmospheric radiation and surface thermal emission capacity, and is estimated using input meteorological data.

[0050] Then, according to the instantaneous net surface radiation flux R n , albedo α , surface radiation temperature T s The empirical relationship function between the normalized difference vegetation index NDVI is used to calculate the instantaneous soil heat flux. G Instantaneous soil heat flux G It represents the rate of heat transfer inside the soil, and the formula is: ; Furthermore, the surface temperature gradient is calculated dT , and calculate the instantaneous sensible heat flux based on the linear relationship H In the present invention, the surface temperature gradient value dT It reflects the rate of change of surface temperature with height and can be obtained through a series of calculations based on meteorological data (such as wind speed, temperature, etc.) and surface characteristics (such as vegetation height, surface roughness, etc.). Specifically, using meteorological data from the local weather station, the friction velocity for neutral stability is calculated. The initial value of the ground friction was calculated by measuring the average height of vegetation (cm) at locations near the weather station. z 0. Furthermore, the influence of surface roughness can be eliminated by converting the near-surface wind speed into the value of the mixing height (200m). An initial estimate of the aerodynamic drag is used to infer r a The first value of . The atmospheric stability correction is obtained iteratively for each pixel, and a series of iterations are required to determine the corrected friction velocity _corr and corrected aerodynamic drag r a_corr The new value of is then numerically stable according to the Monin-Obukhov length criterion.

[0051] In the present invention, the surface temperature gradient value dT The calculation relies on the energy balance framework of the sensible heat flux model, especially in terms of its relationship with the friction velocity and aerodynamic drag r a According to the atmospheric stability theory, the instantaneous sensible heat flux H It can be expressed as: ; ; ; in, is the air density; is the constant pressure specific heat capacity constant, which is about 1004 J / (kg K); is the aerodynamic drag, Measuring height for wind speed; d is the zero plane displacement height, usually 2 / 3 of the surface vegetation height; is the momentum roughness length, which varies depending on the surface type; k is the von Karman constant (about 0.41), u is the wind speed. The instantaneous sensible heat flux H This reflects the physical process of less resistance and more intense sensible heat exchange under rough surface and strong wind conditions. Subsequently, stability correction was performed using the Monin-Obukhov similarity theory, and the corrected _corr and r a_corr . Ultimately, the temperature gradient The acquisition is based on the coupling relationship between sensible heat flux and aerodynamic drag. Meteorological and surface structure parameters such as friction velocity and roughness play a decisive role in this relationship, thus forming a complete set of estimation paths.

[0052] In the present invention, the instantaneous sensible heat flux H The surface radiation temperature can be used T s and surface temperature gradient dT The linear relationship between them is calculated by combining the selected cold pixels and hot pixels (divided according to NDVI values, such as NDVI>80% for cold pixels and NDVI<20% for hot pixels). H It represents the heat exchanged between the earth's surface and the atmosphere through conduction and convection.

[0053] In the present invention, according to the surface energy balance principle, instantaneous evapotranspiration Equal to the instantaneous net surface radiation flux ) minus the instantaneous soil heat flux and instantaneous sensible heat flux , instantaneous evapotranspiration It represents the amount of water released from the Earth's surface into the atmosphere through evaporation and plant transpiration.

[0054] Furthermore, the instantaneous evapotranspiration Perform upscaling calculations to obtain daily evapotranspiration , that is, the target evapotranspiration. Since the energy components estimated by the crop group daily evapotranspiration model are all instantaneous values, in order to obtain the daily scale evapotranspiration, it is necessary to introduce the evaporation fraction The instantaneous evapotranspiration is calculated by upscaling the time ratio based on the parameters such as temperature, humidity, and humidity, so as to obtain the evapotranspiration of the target crop group on each day. .

[0055] Based on the above embodiment, the target evapotranspiration is calculated based on the crop group daily evapotranspiration model. The formula of the crop group daily evapotranspiration model is: ; ; ; in, represents the instantaneous evapotranspiration, represents the instantaneous net surface radiation flux, represents the instantaneous soil heat flux, represents the instantaneous sensible heat flux, represents the evaporation fraction, represents the target evapotranspiration, represents the latent heat of vaporization of water, represents the net surface radiation flux corresponding to the crop planting area every day, It represents the soil heat flux corresponding to the crop planting area on each day.

[0056] On the basis of the above-mentioned embodiment, the above-ground dry matter weight prediction data corresponding to different prediction moments are fitted to obtain the target dry matter accumulation amount of the target crop group corresponding to each day, including: Obtaining the maximum aboveground dry matter weight, the first accumulation time, and the second accumulation time based on the aboveground dry matter weight prediction data corresponding to different prediction moments, wherein the maximum aboveground dry matter weight is determined based on the aboveground dry matter weight prediction data corresponding to each prediction moment; the first accumulation time is the time corresponding to when the target dry matter accumulation amount reaches the maximum dry matter accumulation rate, and the second accumulation time is the time corresponding to when the aboveground dry matter weight prediction data reaches the maximum value; Based on the preset aboveground dry matter weight fitting model, the maximum aboveground dry matter weight, the first accumulation time, the second accumulation time, and the aboveground dry matter weight prediction data corresponding to each prediction moment are fitted to obtain the target dry matter accumulation amount corresponding to the target crop group on each day, wherein the preset aboveground dry matter weight fitting model is constructed based on the beta function In the present invention, the aboveground dry matter weight (AGB) prediction data of the target crop group at different prediction moments are obtained through machine learning models or other prediction methods. These data reflect the dry matter accumulation of crops at different growth stages.

[0057] Furthermore, the maximum aboveground dry matter weight was determined W max In the present invention, based on the aboveground dry matter weight prediction data at each prediction moment, the maximum value, i.e. the maximum aboveground dry matter weight, is found by beta function curve fitting. W max , which represents the maximum aboveground dry matter weight that the crop can reach during its growth cycle. Then, determine the first accumulation time and the second accumulation time , where the first accumulation time Corresponding maximum daily dry matter accumulation The time point when the maximum dry matter accumulation rate reaches its maximum; the second accumulation time Corresponding aboveground dry matter weight prediction data Reach maximum aboveground dry matter weight The time point when the dry matter of the aboveground part of the crop stops accumulating.

[0058] In the present invention, the beta function is used as the basis for the preset dry matter accumulation fitting model. The beta function is flexible and adaptable, and can describe the changing trends of various shapes. It is suitable for fitting the dry matter accumulation process during the crop growth cycle. Specifically, The aboveground dry matter weight prediction data corresponding to each prediction time is used as input and substituted into the Beta function model for fitting to obtain the maximum aboveground dry matter weight. W max , first accumulation time and the second accumulation time Based on the above embodiment, the formula of the preset dry matter accumulation fitting model is: ; ; ; ; in, represents the target dry matter accumulation amount, represents the maximum dry matter accumulation rate; represents the first accumulation time, which is the time when the target dry matter accumulation amount reaches the maximum dry matter accumulation rate; represents the second accumulated data, which is the time when the above-ground dry matter weight prediction data reaches the maximum above-ground dry matter weight; represents the time corresponding to the aboveground dry matter weight prediction data at the current moment, represents the maximum aboveground dry matter weight, Represents the aboveground dry matter weight prediction data In the present invention, a continuous dry matter accumulation curve can be obtained by fitting, which describes the dry matter accumulation process of the target crop population throughout the growth cycle. t > hour, dw / dt will be negative, reflecting the shedding of senescing leaves; c m is the maximum dry matter accumulation rate, Based on the fitted dry matter accumulation curve, the present invention can calculate the dry matter accumulation of the target crop population on each day. These values reflect the dry matter accumulation of the crop at different growth stages. The present invention uses a beta function model, which can be adjusted and optimized according to different crops and growth conditions to improve the accuracy and reliability of the fit.

[0059] Based on the above embodiment, the crop population aboveground dry matter weight prediction model is trained by the following steps: Acquiring sample remote sensing data, wherein the sample remote sensing data includes historical lidar point cloud data, historical multispectral data, and historical thermal imaging data of the crop planting area; generating an aboveground dry matter weight label corresponding to the sample remote sensing data according to the aboveground dry matter weight of the crop population at a historical moment in the crop planting area; constructing a training sample set based on the sample remote sensing data and the aboveground dry matter weight labels corresponding to the sample remote sensing data; The automated machine learning model is trained based on the training sample set to obtain a prediction model for the aboveground dry matter weight of the crop population.

[0060] In this study, we first collected historical LiDAR point cloud data, multispectral data, and thermal imaging data from crop-growing areas. Together, these data constitute a sample remote sensing dataset. The LiDAR point cloud data provides three-dimensional structural information, such as crop height and crown width; the multispectral data contains the reflectivity of crops in different spectral bands, reflecting their physiological and ecological characteristics; and the thermal imaging data records crop surface temperature, which is closely related to crop moisture status and transpiration.

[0061] Furthermore, a corresponding aboveground dry matter weight label was generated for each set of sample remote sensing data based on the measured aboveground dry matter weight of crop populations at historical moments in the crop-growing region. This measured data is typically obtained through ground surveys, sampling, and laboratory analysis, and has high accuracy. In the present invention, the aboveground dry matter weight label serves as the target value for model training, guiding the model to learn the mapping relationship between sample remote sensing data and aboveground dry matter weight.

[0062] Furthermore, the sample remote sensing data is paired with the corresponding aboveground dry matter weight labels to form a training sample set. Each sample contains both remote sensing information and biomass labels. During the construction of the training sample set, data preprocessing, such as denoising, normalization, and feature extraction, may be performed to improve data quality and model training effectiveness.

[0063] In this paper, automated machine learning (AutoML) technology is used to build the model using the H2O modeling platform in a Python environment. AutoML can automatically call various supervised and unsupervised algorithms, such as deep learning, tree ensembles, and generalized low-rank models, providing great flexibility for model training. AutoML can also automatically optimize model hyperparameters, improving the model's predictive performance and generalization ability by searching for the optimal hyperparameter combination.

[0064] In this method, a training sample set is fed into an automated machine learning model for model training. During the training process, the model continuously learns the mapping relationship between the sample remote sensing data and the aboveground dry matter weight, gradually adjusting its parameters to minimize the error between the predicted value and the true value.

[0065] To prevent model overfitting, the present invention uses a five-fold cross-validation method, randomly dividing the training sample set into five subsets of equal size. Four of these subsets are used for model training each time, and the remaining subset is used for model validation. Through multiple cross-validations, the stability and generalization ability of the model can be evaluated. After training and optimization, a crop population aboveground dry matter weight prediction model is obtained. The model can receive new sample remote sensing data as input and output the corresponding aboveground dry matter weight prediction value, providing timely and accurate data support for agricultural production.

[0066] The following describes the crop water utilization rate estimation system based on UAV remote sensing data provided by the present invention. The crop water utilization rate estimation system based on UAV remote sensing data described below and the crop water utilization rate estimation method based on UAV remote sensing data described above can be referenced to each other.

[0067] Figure 2 This is a structural diagram of the crop water use efficiency estimation system based on UAV remote sensing data provided by the present invention. Figure 2As shown, the present invention provides a crop water use efficiency estimation system based on UAV remote sensing data, including a remote sensing data acquisition module 201, an evapotranspiration calculation module 202, an aboveground dry matter weight prediction module 203, a daily dry matter accumulation calculation module 204 and a water use efficiency estimation module 205, wherein the remote sensing data acquisition module 201 is used to obtain meteorological data of a crop planting area, as well as lidar point cloud data, multispectral data and thermal imaging data of a target crop group in the crop planting area, wherein the lidar point cloud data, the multispectral data and the thermal imaging data are acquired based on the remote sensing sensor of the UAV; the evapotranspiration calculation module 202 is used to obtain the target crop group based on the multispectral data, the thermal imaging data and the meteorological data. The target evaporation corresponding to each day; the aboveground dry weight prediction module 203 is used to input the lidar point cloud data, the multispectral data and the thermal imaging data into the aboveground dry weight prediction model of the crop group, and obtain the aboveground dry weight prediction data output by the aboveground dry weight prediction model of the crop group, wherein the aboveground dry weight prediction model of the crop group is obtained based on the training of the machine learning model; the daily dry matter accumulation calculation module 204 is used to fit the aboveground dry weight prediction data corresponding to different prediction moments to obtain the target dry matter accumulation corresponding to the target crop group on each day; the water utilization efficiency estimation module 205 is used to obtain the crop water utilization efficiency of the target crop group based on the target evaporation and the target dry matter accumulation.

[0068] The crop water use efficiency estimation system based on unmanned aerial vehicle (UAV) remote sensing data provided by the present invention obtains meteorological data, lidar point cloud data, multispectral data, and thermal imaging data of the crop planting area through UAV remote sensing sensors; then uses the multispectral, thermal imaging, and meteorological data to calculate the daily evapotranspiration of the target crop population; simultaneously, the point cloud, multispectral, and thermal imaging data are input into a machine learning-trained aboveground dry matter weight prediction model for the crop population to obtain dry matter weight prediction data, and the daily dry matter accumulation is fitted; finally, the crop water use efficiency is calculated by combining the evapotranspiration and dry matter accumulation, thereby improving the accuracy and efficiency of WUE estimation and providing strong support for precision agriculture.

[0069] The system provided in the embodiment of the present invention is used to execute the above-mentioned method embodiments. Please refer to the above-mentioned embodiments for the specific process and detailed content, which will not be repeated here.

[0070] Figure 3 A schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 3As shown, the electronic device may include: a processor (Processor) 301, a communication interface (Communications Interface) 302, a memory (Memory) 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304. The processor 301 can call logic instructions in the memory 303 to execute a crop water use efficiency estimation method based on UAV remote sensing data, the method comprising: obtaining meteorological data of a crop planting area, and laser radar point cloud data, multispectral data, and thermal imaging data of a target crop group in the crop planting area, wherein the laser radar point cloud data, the multispectral data, and the thermal imaging data are collected based on a remote sensing sensor of a UAV; obtaining a target evapotranspiration corresponding to the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data; inputting the laser radar point cloud data, the multispectral data, and the thermal imaging data into an aboveground dry matter weight prediction model for the crop group to obtain aboveground dry matter weight prediction data output by the aboveground dry matter weight prediction model for the crop group, wherein the aboveground dry matter weight prediction model for the crop group is obtained by training a machine learning model; fitting the aboveground dry matter weight prediction data corresponding to different prediction moments to obtain a target dry matter accumulation corresponding to the target crop group on each day; and obtaining the crop water use efficiency of the target crop group based on the target evapotranspiration and the target dry matter accumulation.

[0071] Furthermore, the logic instructions in the aforementioned memory 303 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0072] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the crop water utilization rate estimation method based on drone remote sensing data provided by the above methods, the method comprising: obtaining meteorological data of a crop planting area, and lidar point cloud data, multispectral data and thermal imaging data of a target crop group in the crop planting area, wherein the lidar point cloud data, the multispectral data and the thermal imaging data are collected based on remote sensing sensors of drones; according to the multispectral data, the thermal imaging data The target evaporation of the target crop group on each day is obtained by fitting the aboveground dry weight prediction data corresponding to the target crop group at different prediction moments; the target evaporation of the target crop group on each day is obtained by inputting the lidar point cloud data, the multispectral data and the thermal imaging data into the aboveground dry weight prediction model of the crop group to obtain the aboveground dry weight prediction data output by the aboveground dry weight prediction model of the crop group, wherein the aboveground dry weight prediction model of the crop group is obtained by training based on a machine learning model; the target dry matter accumulation of the target crop group on each day is obtained by fitting the aboveground dry weight prediction data corresponding to different prediction moments; the crop water use efficiency of the target crop group is obtained according to the target evaporation and the target dry matter accumulation.

[0073] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the crop water utilization rate estimation method based on drone remote sensing data provided in the above embodiments, the method comprising: obtaining meteorological data of a crop planting area, and lidar point cloud data, multispectral data and thermal imaging data of a target crop group in the crop planting area, wherein the lidar point cloud data, the multispectral data and the thermal imaging data are collected based on the remote sensing sensor of the drone; obtaining the target crop group based on the multispectral data, the thermal imaging data and the meteorological data The target evapotranspiration of the target crop group on each day is obtained; the lidar point cloud data, the multispectral data and the thermal imaging data are input into a crop group aboveground dry weight prediction model to obtain aboveground dry weight prediction data output by the crop group aboveground dry weight prediction model, wherein the crop group aboveground dry weight prediction model is obtained based on training of a machine learning model; according to the aboveground dry weight prediction data corresponding to different prediction moments, the target dry matter accumulation amount corresponding to the target crop group on each day is obtained by fitting; according to the target evapotranspiration and the target dry matter accumulation amount, the crop water use efficiency of the target crop group is obtained.

[0074] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0075] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for estimating crop water use efficiency based on UAV remote sensing data, characterized in that: include: Acquiring meteorological data for a crop-growing area, as well as laser radar point cloud data, multispectral data, and thermal imaging data of a target crop population within the crop-growing area, wherein the laser radar point cloud data, the multispectral data, and the thermal imaging data are collected using a remote sensing sensor based on an unmanned aerial vehicle; Obtaining a target evapotranspiration of the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data; Inputting the lidar point cloud data, the multispectral data, and the thermal imaging data into a crop population aboveground dry matter weight prediction model to obtain aboveground dry matter weight prediction data output by the crop population aboveground dry matter weight prediction model, wherein the crop population aboveground dry matter weight prediction model is trained based on a machine learning model; fitting the aboveground dry matter weight prediction data corresponding to different prediction moments to obtain the target dry matter accumulation amount of the target crop group on each day; The crop water use efficiency of the target crop group is obtained according to the target evapotranspiration and the target dry matter accumulation.

2. The method for estimating crop water use efficiency based on UAV remote sensing data according to claim 1, characterized in that: The step of obtaining a target evapotranspiration of the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data includes: Obtaining, based on the multispectral data, a normalized difference vegetation index, albedo, and atmospheric emissivity corresponding to the crop planting area; Obtaining the surface radiation temperature corresponding to the crop planting area according to the thermal imaging data; Calculating the instantaneous net surface radiation flux of the crop planting area according to the normalized difference vegetation index, the albedo, the atmospheric emissivity and the surface radiation temperature; Calculating the instantaneous soil heat flux of the crop planting area according to the instantaneous surface net radiation flux, the albedo, the surface radiation temperature and the normalized difference vegetation index; Calculating a surface temperature gradient value based on the meteorological data, and calculating an instantaneous sensible heat flux in the crop planting area based on a linear relationship between the surface radiation temperature and the surface temperature gradient value; The instantaneous evapotranspiration is calculated based on the instantaneous net surface radiation flux, the instantaneous soil heat flux and the instantaneous sensible heat flux, and the instantaneous evapotranspiration is upscaled to obtain the target evapotranspiration corresponding to the target crop group on each day.

3. The method for estimating crop water use efficiency based on UAV remote sensing data according to claim 2, characterized in that: The target evapotranspiration is calculated based on a crop group daily evapotranspiration model, and the formula of the crop group daily evapotranspiration model is: ； ； ； in, represents the instantaneous evapotranspiration, represents the instantaneous net surface radiation flux, represents the instantaneous soil heat flux, represents the instantaneous sensible heat flux, represents the evaporation fraction, represents the target evapotranspiration, represents the latent heat of vaporization of water, represents the net surface radiation flux corresponding to the crop planting area every day, It represents the soil heat flux corresponding to the crop planting area on each day.

4. The method for estimating crop water use efficiency based on UAV remote sensing data according to claim 1, characterized in that: The step of fitting the aboveground dry matter weight prediction data corresponding to different prediction moments to obtain the target dry matter accumulation amount of the target crop group on each day includes: Obtaining the maximum aboveground dry matter weight, the first accumulation time, and the second accumulation time based on the aboveground dry matter weight prediction data corresponding to different prediction moments, wherein the maximum aboveground dry matter weight is determined based on the aboveground dry matter weight prediction data corresponding to each prediction moment; the first accumulation time is the time corresponding to when the target dry matter accumulation amount reaches the maximum dry matter accumulation rate, and the second accumulation time is the time corresponding to when the aboveground dry matter weight prediction data reaches the maximum value; Based on a preset aboveground dry matter weight fitting model, the maximum aboveground dry matter weight, the first accumulation time, the second accumulation time, and the aboveground dry matter weight prediction data corresponding to each prediction moment are fitted to obtain the target dry matter accumulation amount corresponding to the target crop group on each day, wherein the preset aboveground dry matter weight fitting model is constructed based on a beta function.

5. The method for estimating crop water use efficiency based on UAV remote sensing data according to claim 4, characterized in that: The formula of the preset aboveground dry matter weight fitting model is: ； ； ； ； in, represents the target dry matter accumulation amount, represents the maximum dry matter accumulation rate; represents the first accumulation time, which is the time when the target dry matter accumulation amount reaches the maximum dry matter accumulation rate; represents the second accumulated data, which is the time when the above-ground dry matter weight prediction data reaches the maximum above-ground dry matter weight; represents the time corresponding to the aboveground dry matter weight prediction data at the current moment, represents the maximum aboveground dry matter weight, It represents the predicted data of aboveground dry matter weight.

6. The method for estimating crop water use efficiency based on UAV remote sensing data according to claim 1, characterized in that: The crop population aboveground dry matter weight prediction model is obtained by training through the following steps: Acquiring sample remote sensing data, wherein the sample remote sensing data includes historical lidar point cloud data, historical multispectral data, and historical thermal imaging data of the crop planting area; generating an aboveground dry matter weight label corresponding to the sample remote sensing data according to the aboveground dry matter weight of the crop population at a historical moment in the crop planting area; constructing a training sample set based on the sample remote sensing data and the aboveground dry matter weight labels corresponding to the sample remote sensing data; The automated machine learning model is trained based on the training sample set to obtain a prediction model for the aboveground dry matter weight of the crop population.

7. A crop water use efficiency estimation system based on UAV remote sensing data, characterized in that: include: A remote sensing data acquisition module is used to obtain meteorological data of the crop planting area, as well as laser radar point cloud data, multispectral data, and thermal imaging data of the target crop population in the crop planting area, wherein the laser radar point cloud data, the multispectral data, and the thermal imaging data are collected by remote sensing sensors based on drones; an evapotranspiration calculation module, configured to obtain a target evapotranspiration corresponding to the target crop group on each day based on the multispectral data, the thermal imaging data, and the meteorological data; an above-ground dry weight prediction module, configured to input the lidar point cloud data, the multispectral data, and the thermal imaging data into a crop population above-ground dry weight prediction model to obtain above-ground dry weight prediction data output by the crop population above-ground dry weight prediction model, wherein the crop population above-ground dry weight prediction model is trained based on a machine learning model; a daily dry matter accumulation calculation module, configured to obtain the target dry matter accumulation of the target crop group on each day based on the aboveground dry matter weight prediction data corresponding to different prediction moments; The water use efficiency estimation module is used to obtain the crop water use efficiency of the target crop group according to the target evapotranspiration and the target dry matter accumulation.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the method for estimating crop water utilization efficiency based on drone remote sensing data as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for estimating crop water use efficiency based on drone remote sensing data as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for estimating crop water use efficiency based on drone remote sensing data as claimed in any one of claims 1 to 6 is implemented.

Images